Computing Project Repository - New pages [en]

Systems Administration as a Self-Organizing System: the Professionalization of SA via Interest & Advocacy Groups

Mark — Sat, 18 Nov 2006 11:21:26 GMT

Summary:

Systems Administration as a Self-Organizing System: the
Professionalization of SA via Interest & Advocacy Groups
Strata R. Chalup <strata@virtual.net>

Abstract:
Lacking a single vendor or institution both motivated and powerful enough
to provide a structure for collaboration, early Unix System Administrators
had to build their own structures. The gradual coalescence of
a cohesive SA community out of the mish-mash of academic, commercial, and
government Unix environments was among the most fruitful of these
bootstrapping efforts.

Introduction

System administration is clearly shown to be a member of the
engineering professions by the plethora of sysadmin membership organizations,
each focused on particular specialties, credentials, or topics. There
is no overarching organization today that serves the entire sysadmin
community. As the community has become larger and more diverse,
it has become common for SAs to be members of several professional
societies according to their training, specialties, and general career
focus. Thus you have general groups for consultants (ICCA), for
auditors (CoBIT), for
academically-focused organizations that form a baseline for
college-educated professionals (ACM, IEEE), system research and practice
organizations (USENIX, SAGE, LOPSA), regional umbrella groups (BayLISA,
$groupname), and operating-system advocacy
groups (a variety of Linux User Groups, or LUGs).

If we make a common analogy of a system administrator as a doctor,
we find there is no American Medical Association equivalent. ACM and, to a
lesser extent, IEEE serve as a surrogate for a truly sysadmin-centric
overarching organization. USENIX is the AMA equivalent for unix-oriented
system administrators, but is largely ignored by desktop computer admins
and network admins. The latter primarily rally around IETF and NANOG,
which are working groups and discussion lists, rather than membership organizations.
(confirm on IETF)

The past decade has seen a sharp rise in the number of
vendor certification programs offered, and many successful attainees
tend to congregate in vendor-specific forums and at vendor-sponsored
conferences, such as Oracle's annual conference. Among the most prominent
examples of certification-centered communities are those revolving
around Microsoft's operating systems and their Microsoft Certified Professional
(MCP) program. Sun, Red Hat, Oracle, and Cisco are among the most
prominent vendors offering these programs. Some independent organizations
have responded by creating their own training, such as SAGE's cSAGE
certification. Training vendors have also embraced user communities
as a way of publicizing their own certificate programs. The Linux
Professional Institute (LPI) and SANS both are examples of vendors which
do not have membership-driven groups, but nonetheless provide support
frameworks such as mailing lists, conferences, and regional meetups as
a combination 'virtual user group' and upselling opportunity.

Establishing the Problem Space

Managing complex computer infrastructures requires a support structure
of peers who can serve as mentors, inspiration, and sounding boards
for solutions and wild ideas. Early Unix platforms provided some
of the same computational timesharing facilities generally
associated with mainframes of the day. The mainframe operators
enjoyed a combination of standardization within specific vendor
communities, coupled with a high level of formal and informal
organizational support from the platform providers. The operator
community was correctly identified as a valuable resource by
vendors of expensive mainframes, and cultivating the community was
part of most vendors' official sales strategy. Groups such as
Digital Equipment Corporation's "DECUS" and Control Data Corporation's
"PLATO" provided an early model for structured interaction between
systems staff and vendors, and for peer to peer networking between
SAs from different sites.

Meanwhile, Unix SAs were not "feeling the love". In sharp contrast to
the active vendor support enjoyed by mainframe SAs, the pioneers
managing Unix environments of the 1970's and 1980's were generally
either ignored or actively discouraged by vendors.
Even Digital Equipment Corporation, whose internal Unix Education Group
(UEG) had created a Bell Labs V7 port for the VAX and played key roles
in the creation of 4.1BSD, did not consider any Unix version
supported until its own release of Ultrix in 1984. Early SA's
resignedly kept a VMS or RSTS-E boot tape handy in case their
systems required hardware support from DEC. Soon DEC's PDPs and
VAXes were joined by an ever-growing assortment of Motorola 68000
platforms from established players as well as startups.

Ultimately the strong research orientation surrounding early users
of Bell Labs' most viral software release would find expression
in peer-reviewed the initial model for what would become the Usenix
Annual Technical conference in the late 1970's, and spin off the
Large Installation System Administrators (LISA) conference in 1987.

April 1987 The first LISA Workshop, chaired by Rob Kolstad and Alix Vasilatos, brings in 75 attendees

BayLISA founding year 1991. Tradition of meetings being on the 3rd
Thursday of the month will be followed by other groups such as $groupname;
may date from DECUS days.

SAGE, the System Administrators' Guild, started in early 1992 by a group
of San Francisco Bay Area sysadmins, many of whom were also BayLISA
founders, becomes a Special Technical Group of USENIX in June 1992.

SAGE: "...to serve the system administration community by:
* Offering conferences and training to enhance the technical and managerial capabilities of members of the profession
* Promoting activities that advance the state of the art or the community
* Providing tools, information, and services to assist system administrators and their organizations
* Establishing standards of professional excellence and recognizing those who attain them"

$groupname 1993: "$GROUPNAME is an organization for UNIX system
administrators in New Jersey formed to facilitate information
exchange pertaining to the field of Unix system administration.
$GROUPNAME is not affiliated with a particular hardware or software
vendor or company." (via their site)
Summer 1993 Tom Limoncelli shares airplane home from Usenix with Matt Blaze. Tom talks about possibly creating a local LISA-like group. Matt mentions that Tommy Reinhold (a co-worker) is also interested. Tommy and Tom talk in the next week and try to put something together. Tom posts all over the net an announcement that a mailing list called GSLISA has started, and its purpose is to figure out how to create such a group. Tom posts first Draft of the charter, and invents the name $GROUPNAME as a temporary solution. Much to his chagrin, people say they want to use it permanently. July 22, 1993 About 30 people met at The Princetonian Diner, sang, drank, and made merry. Oh, they also worked out details of creating said group. Names were discussed. A provisional board was elected in November of 1993 and the rest, as they say, is $history.

DC-SAGE, Back Bay LISA, and similar regional groups: many are claimed
by various parent orgs, but strictly maintain their independence despite
similar naming conventions. http://www.dc-sage.org/ (expand, give
backstory)

SIG-BEER, a venerable DC-SAGE tradition. "SIG-beer, as in "Special
Interest Group - Beer" ala ACM, or as in "send the BEER signal to that process"
(via website)

2002, David Albans moves to the SF Bay Area from Washington, DC, and starts
SIG-BEER-WEST in February, continuing the 3rd Saturday tradition. Still going strong in
2006.

LUGs now serve where sysadmin groups once trod. ;-) There is bottom-up
organization, too, as small regional LUGs merge to create meta-event
aggregators such as SVLUG (Silicon Valley) or BALE (Bay Area Linux
Events).

2000 - 2005 The Cultural Revolution / Agonizing ReAppraisal: SAGE splinters,
LOPSA forms, primarily around the different viewpoints of research/cutting-edge sysadmin vs trade-union-like nurturing of the profession. There were also subgroups within both groups who argued strongly for maintaining an elite core, rather than reaching out inclusively. Some positions favored an approach based on seniority, others on topic, wishing to limit potential morphing of either organization by an influx of Windows-centric admins (rephrase for better PC)

LOPSA: "Our mission is to advance the practice of system administration; to support, recognize, educate, and encourage its practitioners; and to serve the public through education and outreach on system administration issues."

In 2000, the DECUS U.S. Chapter incorporated as the independent user group Encompass.

**** Trade & Industry Groups ****

1. IT focused groups:
http://www.itaa.org/
http://www.simnet.org
http://www.naspa.com/
http://www.techforum.com
http://www.cmiiug.com

Also http://www.campconferences.com/

Platform focused groups / shows - maybe more historical than current:
http://www.share.org/About/ - IBM
DECUS - Dec
http://www.techmentorevents.com/ - windows
http://www.jcp.org - java

Also - http://www.sun.com/bigadmin/home/index.html - Sun

**** References ****

$GROUPNAME, http://www.groupname.org/
USENIX Community Services timeline, http://www.usenix.org/about/history/community.html
USENIX Firsts timeline, http://www.usenix.org/about/history/firsts.html
LISA, http://www.usenix.org/events/byname/lisa.html
LOPSA, http://www.lopsa.org/
AT&T System V, http://en.wikipedia.org/wiki/UNIX_System_V
AT&T Version 7, http://en.wikipedia.org/wiki/Unix

High Volume Services notes

Mark — Thu, 09 Nov 2006 07:54:25 GMT

Summary:

== Terminology ==

Explain the difference between the following?

* High Performance Computing
* High Volume Services
* High Availability Services

Hardware, software redundancy - clustered environment. Monitoring
take action on node/service failure. Load-balancing

=== HPC ===

*High Performance Computing*

HPC is the term used to describe the use of networked clusters of
computers to take a large computational (usually scientific, number crunching) problem, divide it into to smaller chunks, distribute those chunks to
the nodes in the
cluster(s), and then solve those chunks in parallel. HPC sometimes
requires very high-bandwidth and/or low-latency interconnects between
nodes to address the target problems.
High Performance computinga also refers to the use of supercomputers and computer
clusters that is, computing systems comprised of multiple processors linked
together in a single system.

Grid computing - combining resources of (heterogeneous) IT components
into one pool of shared resources.

=== HVS ===
*High Volume Services*
Applications or services that are designed to handle a large number of requests/operations. This is achieved by implementing strategies to share the volume/load betweeen separete service hosts (load balancing), often based on queuing theory priciples.

HVS is focused on serving large amounts of requests, and therefore
scalability. A typical concern here is having appropriately large
network throughput but server resources must also be scaled to provide
the resources to answer the requests. Load-balancing is often used in
this context to attain a high-efficiency processing of requests among a
pool of resources. There is often a discussion of certain metric
objectives (e.g. response time).

This includes the management of services, the economics and the performance
tuning of services. The ability to deliver reliable service levels depends on
both the resources that are available in the data centre and the pattern of
demand driven by the users.

Can cope with high loads, i.e a webserver which gets alot of requests, may
cope with high load using webfarms.

=== HAS ===

*High Availability Services*
Applications or services designed for "mission-critical" situations, where it is of the essence that the service always is available for the clients. To achieve this a mixture of redundancy and failure detection are implemented in the system responsible for delivering the service.

Services that have to be redundant. Services that always have to be
available.
HAS is about ensuring that a service is available according to a certain
target service level. This takes the form of redundancy at some level to
accommodate for failures of individual components. Load-balancing is
often used in this context to route requests to an available server.
There is often a discussion of what "available" means in terms of metric
goals (e.g. response time) here as well.
High availability Service: refers to service offering a high level of service
uptime with assured quality.

Also called "RAS" (reliability, availability, serviceability) or "fault
resilient," it refers to a multiprocessing system that can quickly recover from
a failure. There may be a minute or two of downtime while one system switches
over to another, but processing will continue. This is not the same as fault
tolerant, in which redundant components are designed for continuous processing
without skipping a heartbeat.

High availability also refers to being able to service a component in the system
without shutting down the entire operation. See clustering and hot fix.
http://www.pcmag.com/encyclopedia_term/0,2542,t=high+availability&i=44246,00.asp
This term gets more and more related to Virtual servers it seems.
[http://www-935.ibm.com/services/us/index.wss/offerfamily/its/a1000424 - IBM]
Nice Link: http://www.linuxvirtualserver.org/HighAvailability.html

== Resource Sharing ==

The essence of high performance and therefore high volume is resource management. What resources are being referred to in the names above?

=== Technologies ===

We want to utilize/share

* CPU
* Disk
* Memory
* Network

between different tasks/users/processes. What technologies are available to do this in each of the three cases above?

==== General ====

P2P - e.g. bittorrent, distributed data dissemination

Live server migration - virtualization for seamless service

Load-balancing clusters operate by having all workload come through one or more
load-balancing front ends, which then distribute it to a collection of back end
servers. Although they are primarily implemented for improved performance, they
commonly include high-availability features as well. Such a cluster of
computers is sometimes referred to as a server farm.

The term "Grid Engine" often pops up in this context. Grid engines differs from
clusters by being able to work alone (no one element is purely shared), adding
power or new items if it needs to

==== CPU ====

For sharing CPU, a job distribution system (often associated with HPC
clusters) is often used.

* Beowulf cluster
* GridEngine for parallel computing.
* Load balancer to balance requests to multiple servers.
* Multitasking at OS level.
* MPI (Message Passing Interface)
* Parallel systems (blade racks)
* Mainframe computing
* CPU caching

Putting CPUs into a single computer and have all of them work together
simultaneously to solve problems. This procedure is now known as parallel
computing. Machines providing the highest level of performance today are all
parallel computers, not sequential machines.

==== Disk ====

* SAN
* NAS
* RAID

Filesystems and filesystem attributes are used to share disks. On a
network level, you can create a storage area network or use
network-attached storage to provide a single pool of disk to a large
number of clients. Various RAID levels can be used to achieve certain
results, an example is striping which can improve read/write operations.
Distributed storage techniques are in development which some believe
hold promise for moving the challenge of storage from the edge into the
network.

[http://en.wikipedia.org/wiki/Distributed_file_system Distributed filesystems]

Storage Area Network(SAN) (low level - disk appears as locally attached)
nfs/samba (high level - mounted over network)
network-attached storage(nas - os provides only filesharing)

http://en.wikipedia.org/wiki/SCSI#SCSI_interface_overview
Max devices, speeds etc.

http://en.wikipedia.org/wiki/Fiber_channel
Fibre Channel Protocol (FCP) is the interface protocol of SCSI on the Fibre
Channel. Topology, Protocols etc.

==== Network ====

* Routing policies.
* QoS parameters
* Providing (Provisioning)
* Use the time multiplexing property in ethernet (Maximum transmission unit)
* Gb Ethernet, Infiniband, ++

Might be required to use fast connection such as fiber( connecting to a SAN
storage for example). Redundant links.

* SAN (Storage Area Network) - uses a protocol (e.g. ISCSI - SCSI over IP) to make it look as though network disks are attached

Why would we do this? If the internal bus is faster than the IP network?
Does size matter?

What are the limits of storage on a regular SCSI bus? See above. But with ISCSI we can have unlimited number. Depends on te implementation of the ISCSI client. In Linux uses 16M for caching which limits the number of devices that can be connected in practice,

We can also use Fibre channel. How does this work?

A SAN is normally connected to a different interface than the regular network.
(Not always, but should be?) e.g. ISCSI can run on regular ethernet, so
a second net card on the computer. Or fibre interface.

This is a lower level protocol resource management. Looks like a local device to the computer rather than a network service. Here you need something in the kernel to support to this.

* NAS (Network Attached Storage) This is a service running over a regular network protocol, e.g. sambda, NFS, NetApp, Firewire, shared drives.
Normally this would run over the same network interface as normal traffic
(but does not have to).

This is more high level resource management. High level service. AFS, DFS.
Distributed file system. Don't need anything special in the kernel (except
standard service drivers).

Is there a clear distinction between a SAN/NAS?

==== Backup ====

SAN is more flexible for backup, since we don't *have* to hvae a filesystem.
A NAS would "always" be a mounted filesystem. So we can so "dump" to a SAN?

== Server Technology ==

=== What is a service? ===

Buzzword? Usually over network? A service is a "process" at any level
.e.g at OS level, or business level? A running application (application level services)? Network level services. A daemon?
Work done by one entity for another?

Culture difference -- ISPs think of everythign exxcept the infrastructure as a service. UNix admins think of a service as a daemon (app. service).

We can be flexible.

=== What is the difference between a server and a workstation? ===

A workstation needs a graphics card, colour screen, sound, ... keyboard

A server is not necessarily about peformance, but saving money on the above.

So servers are usually more expensive - why? Expensive hardware. (RAM - registered RAM, fault tolerant, faster, error correction) (MTBF longer?) (Disk -- two classes of disk. Disks are tested at the factory and the higher reliability/performance ones are sold differently with higher spec. So we have some guarantees)

Physical size? (Not necessarily).
Server - longer lifetime, more fault tolerant,
Low level redundancy
IBM sell their mainframes for low level redundancy
Server serves more people, higher importance (subjective view)
Greater requirements for server.
Cost - workstation should be (can be) cheaper

Is is question of scale? How often can we afford to replace hardware?
If we split into two tiers, servers and workstations, we can change workstations
more often (cheaper) than servers.

Is this a strategy for the vendors to sell to us, or a strategy for us to
solve our problems?

* Why do we need clusters?
We can extend them rather than replace them. Cheaper.
Cost of maintaining a cluster built from ad hoc.

Power supplies - power factors are better for expensive hardware than for
cheap. We can reduce the electricity bill.

=== What is the reason for server-blade design? What is wrong with a regular PC? ===

* Blades are small, can fit a lot into a small space
* Hot swappable
* Power distribution (common to all blades in a chassis)
* Common switch, internal network
* We can make local VLANs internally in a chassis, or link to outside network
* Designed for redundancy - two of everything (power, switch, ...)
* Separate network for management console

====Bottlenecks?====

* Network - sharing common network - how does the speed internally compare to link externally? Gb ethernet? Could be a bottleneck?

What is faster than Gb ether? Infiniband - high speed, short distance. Each technology has its own design trade-offs. Short distance/long distance. PIck a tech that works "Best".

* Disk
* Memory
* CPU

= Questions for next time =

* Uptime Institution Tier Level System??? Ranking for datacentres?

* VPNs what roles to these play in a data centre?

* What role does a SLA play in data centre?

* How do queues apply? Where are they? Can we use queueing theory?

Notes on research

Mark — Mon, 30 Jan 2006 07:24:05 GMT

Summary:

== Visits ==

It takes a year to follow up on a vist from a researcher, or a visit to some other college.

Lisa conference report

Mark — Tue, 03 Jan 2006 08:24:14 GMT

Summary:

== LISA'05 ==

On Friday, December 2nd 2005, four brave members of the staff at Oslo University College depart for San Diego. Their objective:

: To attend the [http://www.usenix.org/events/lisa05/ Large Installation System Administration Conference (LISA '05)].

You can follow their exciting journey, as they eat, sleep and inhale the adventure that is LISA. Watch this space to get updated information as the events unfold.

=== General remarks ===

This year's LISA conference was the best in a long time. We made many good contacts with old friends and colleagues.

=== Student Impressions ===

==== Espen ====

The most valuable for me at LISA'05 was the virtualization BOF where I got
in contact with the "virtualization community". I signed up for an
attendee mailinglist and discussed my master thesis description with Steve
Traugott in infrastructures.org. Monica Lam's talk about "automatic PC
desktop management with virtualization technology" was also interesting,
and I discussed a while with Costa Sapuntazakis (Lam's phd student). Both
Traugott and Sapuntazakis were interested in my thesis, and encouraged me
to establish e-mail contact with them when I started to work on it. These
contacts will probably be valuable resources in my master thesis.

I got in touch with Jeff Polo in Google before the google BOF, and also
other google employees at the BOF. We mostly discussed the possibility for
internship in Google in Norway. I am going to send e-mails to the people I
talked to, and hopefully get more information about Google in Norway and
how to best get in touch with them.

I attended several other sessions as well, but only for my own interest
(shell programming, os x, network black ops, backup, etc). It felt very
valuable to attend sessions that was related to my master thesis, and on a
conference like this I think it is important to have a topic that one can
consider as "my own field" - in my case, virtualization.

==== Maurice ====

Our part of the USENIX sponsoring agreement was to stuff bags on
Saturday before the conference, as well as handing out printed material
(tutorials) to conference participants on 2 other days. This gave me the
chance to get to know both USENIX staff and others - some of them I knew
as book authors, or I had heard about them as authorities in the
sysadmin field. Exciting to meet them personally! Even during our
working days it was possible to attend tutorial sessions, and some of
them were a delight: interesting material, presented in a great manner.
Attending the more academic sessions (invited/refereed papers) allowed
me to have an update on recent research, and experiencing "live" the
interactive nature of these talks by means of q&a mostly at the end, but
sometimes even during the presentations.

Being able to attend LISA was not only useful for my Master thesis (got
several useful contacts, plus opinions and hints from experienced
people), but also an ice-breaker for future conferences. I am very
grateful for both the support HiO/IU (a special thanks to you, Mark!!)
and USENIX provided - it was really a sweet deal!

==== Ilir ====

he trip to LISA was carried out in the right time for me. It was a period
when I was a little unsure about my professional orientation. The
selection of technical sessions and tutorials was wide, yet I managed to
join in a couple of them. We as a group of voluntary students were lucky
to be able to join any of them for free. I attended "Network Black Ops:
Extracting Unexpected Functionality from Existing Networks Dan Kaminsky,
DoxPara Research" and of special interest was the Network Activity
Visualizing part of the lecture, as my Masters Project is going to involve
a lot of network traffic analysis. I also found interesting to get more
familiar with MACs, as it was one of the main laptop brands used by
participants in LISA. The lecture "About the Integration of Mac OS X
Devices into a Centrally Managed UNIX environment by Anton Schultschik,
ETH Zürich" contained interesting informations and appealed me to join.
Appreciation of security is crucial for an SA, so I decided to get some
information, and joined "What's the Worst That Could Happen? by Peyton
Engel, Berbee Information Networks Corporation". Though not very
technical, it provided plenty of useful information regarding broad
concepts of security issues.
"Picking Locks with Cryptology Matt Blaze, University of Pennsylvania" was
another useful lecture, which gave insights on technologies used in
eavesdropping links, and ways to secure the information exchange.

Masters programme BoF was a good way to portray our work at Oslo
University college.
Meeting and finding potential employers was a very important moment. I got
the chance to talk to Google company representatives, who were hiring.
Exchanging information with them was very useful. Additionally, there were
many job postings in message boards which had useful information regarding
job inquiries.

I am thankful to you for providing/introducing me with this opportunity to
meet with professionals, and see some real-life applications of what we
read from books. I must say it has been often discouraging to see such
high-level experts, worrying whether I will personally be any close to
that, and at the same time encouraging one since it opened some doors of
opportunity, where I could see myself working, applying and developing my
knowledge.

==== Kyrre ====

Kyrre joined a tutorial on Linux System Administration in Production
Environemnts in order to update himself on the lates techologies used
in the industry. He also attended a workshop on system administration
education. This workshop was productive in many ways as the community
got together and discussed problems and challenges. Kyrre recieved a
great deal of positive feedback and suggestions for the software tool
MLN which is designed by Kyrre and developed at IU.

==== Siri ====

I attended several of the technical sessions, but particularly enjoyed
the sessions on System Administration Theory, and the session on Network
Visualization.

In addition to the technical sessions, I also attended a couple of BoF's
(Virtualization and Cfengine), in addition to the BoF where the Oslo
University College Master programme was presented. I also attended a
so-called Work-in-progress-session, where each presenter only had five
minutes to present the essence of their work. This allowed 11 people to present
their work during less than two hours, which was a very interesting and
entertaining session.

During the conference, I met several interesting people, and got to
discuss issues strongly related to the topic of my Phd-project.

=== Day 7, Friday ===

Mark takes the masters students out to dinner in the San Diego old town.

=== Day 6, Thursday ===

Mark gave a paper in the morning and a tutorial in the afternoon.

=== Day 4, Tuesday ===

Mark gave a tutorial in the morning.

==== System admin education workshop ====

Below are the DRAFT notes from this year's workshop. Thanks to everyone who was able to make it. Please let me know of corrections.

Thanks,
Steven

9-10 Introductions
- Problems
* How to connect our graduates to jobs
* Scott: How to build good problem-solving instead of recipes
* David: no money for course development -- drives towards
shared platform -> agree on a platform so we could share exercises
more easily
* Elizabeth: testing & assessment; external validation of testing
and assessment: SAGE certification standards/tests
* Matt: advocacy of education efforts
* Steven: how to connect us to others (in particular, to ACM)
* Steven: no web & ftp teaching materials

- Issue that seems common: lack of Windows administration expertise

10-12 Virtualization
- mln "My Linux Network"; Kyrre runs close to 100 hosts with mln. Built
on UML, so there are some stability issues because UML is still heavily
under development
- Two other products in the same space: vulab & vcml (by Zach Lowry: built on
QEMU); VMware or VirtualPC-based solutions are also possible.
- lab issues: logistics (physical space, setup time) x
cost (of hardware & space) = pedagogical payoff
- A solution adopted by everyone would need to be low/no cost and
cross-platform (e.g., VMware, VirtualPC, etc)

Assessment/Testing/Evaluation
- 'group riders' -
* weight group-work less than individual metrics (ie, homeworks 20%,
tests 80%)
* exams-per-assignment (ie, grade per topic is a combination)
* inter-group evaluations/peer evaluations
* add a notebook component per person
- Automated assignment grading
* some are doing it, basically by pre-setting test systems
* Acuitus handles tracking by looking at the following:
- machine's behavior,
- change ontology (ie, machine state),
- student belief state (ie, what the student thinks the problem is)
- David does interviews/demonstrations with each group of students to
do assessment
- David is moving _towards_ recipes _because_ of the 'just google it'
phenomenon (i.e., they search for 'email configuration' and find AIX
docs, even though they are working on, say, Solaris)
- Another common strategy is to give recipes at the beginning and
wean them from them over the course of the program.
- Some do pre-tests and post-tests to show improvement (this is
basically a required approach for accreditation). Some would like to
do 'test-out' approaches for earlier labs, for example. Using such
tests for diagnostic purposes seem to be a good theme. It might be
useful to have a common (ie, across this group) diagnostic test, or
perhaps a modular set.
- Big issues for us: (our) time, dealing with 'out of the ordinary'
approaches for solutions (e.g., instead of fixing the 'broken' box,
break the correct one
to match the broken one), breaking down larger projects into
(checkable) milestones,
- David has a vi tutorial he gives
- Modules people would like to see:
* LDAP (Aeleen Frisch did an LDAP module at Oslo this Fall)
* ActiveDirectory (an equivalent module to the LDAP one)
* Basic stuff: cd, editors, etc
* (Kyrre) a way to more easily and accurately describe overlap
between courses among schools
* Backups (and restores): Curt has some materials that may be
sharable/helpful. Aeleen did a backup module for Oslo also.
* Authentication & Authorization
- External assessment
* a more generic tool like SAGE cerification would be viewed as a very
positive thing
* one view is that a particular exam (within a course or program)
is hard to be exchanged with an external one
* 'nice to have', not necessary
* could be nice as an integrative exam (within a 'track', 'thread',
'concentration' or 'emphasis' program)

12:30-1:30 lunch

1:30-3 ACM IT Curriculum (and community issues)
* sysadm-education recommended curriculum models
* How to connect our graduates to jobs
- bringing students to LISA has mixed success
- internships are a great idea
- Oslo has had some success where students' final
projects (which include an industry customer) ends up
in job offers for those students (ie, job offers from the
customer).
- visibility and credibility (of the field) is an issue
* Advocacy of education efforts
* How to connect us to others (in particular, to ACM)
* Mark suggests a consortium idea -- grouping together to
- apply for grants/funding
- publish papers together
- (collaborate in general)
- Consortium of System Administration
with the overall view of raising visibility of System Administration
education and research
* Oslo:
MS program:
- has graduated one class (of 12)
- second class has ~15 students
- next has ~15 students
- two goals of program: terminal degree _and_ leave open
option of research (ie, PhD)
- currently 2 PhD students, expecting to graduate in 2007
* Alva currently has 4 PhD students, Yizhan Sun (almost finished),
Ning Wu (post-qualifiers), Mark Chiarini (pre-quals, post-courses),
and Hengke Susanto (working on autonomics).

3-3:30 Break

3:30-5 Textbooks & Materials
* Quite a few _are_ using Mark's Principles book as the core
text (both w/in the workshop and outside)
* Scott's school has access to 'Books 24/7'
* Steven's school has access to Safari (safari.oreilly.com)
* Technical book vs textbook: Kyrre sees a difference in quality
of reports in the original sources (recipes vs textbook)
* Curt is using outside reviewers of student papers (eg, students
in a different course will give feedback on the writing)
* Firewalls and Intrusion Detection (Kyrre is using)
- Northcutt's Network Intrusion Detection
- Ziegler's Linux Firewalls

* Ethics: some have a few chapters, but there don't seem to
be any good Ethics texts for System Administration (The Gift
of Fire seems to be fairly popular). Curt shilled 'Ethics and
Computing' by KW Boyer. The SAGE Code of Ethics is too
general and high-level to be the basis for a course in Ethics
for System Administrators.
* Security: Bishop, Secrets and Lies
* Networking:
- Peterson & Davies
- Kurose & Ross - for undergraduates
- Fourazan - good on details
- Tanenbaum
- Leon Garcia's: Communication Networks
- Alva has a lecture called 'Breaking IP'
* Another missing area: how to do write about research.
- problem is missing the scientific context
- Steven needs to dig up the reference on research methods
for CS paper that he has lying in his office somewhere
* Unix Fundamentals:
- Glass' Unix for Programmers and Users
- Mark Sobel's book Unix System V
- It's hard to beat Kernighan and Pike
- David would be interested in collaborating on a book for this
- Mark offerred to let people start from his notes to put
together an online text/version

* Mark offered to let us host our notes & materials on the Oslo
'System Administration Resources' site so that we can more easily
collaborate/share materials (http://www.iu.hio.no/SystemAdmin/). Contact
Matt if we want to put content up.

* Another possibility is the Drupal site set up last year (or conference
wiki?).

* A site for sharing this information might be something useful for the
Consortium to have as a main(?) focus.

* Norway's Network of Excellence called 'Magix' -- they are motivated
to produce teaching resources. They are going to make a virtual
laboratory to share the resources of ~30 universities. They might
be able to assemble a scheme wherein the Consortium could access
the resources as well.

* If someone is interested in looking at other student projects (e.g.,
Master's Theses), some way of sharing those might be something the
Consortium could do.

* Another idea is cross-institution collaboration on projects.
* YAI is doing external audits of curriculum

Thoughts on this year:

* good discussion today
* a paper idea would be to write up issues & challenges (and even a survey)
of virtualization and System Administration, Security, & Distributed
Computing Education (Kyrre and Matt, among others, are interested in that)
* The consortium idea has some support. Oslo will set up a wiki. One goal
would be to have it visually appealing (hint: anyone have a student project
to do web design?)

Attendees:

Curt Freeland - curt@cse.nd.edu

Steven Jenkins - jenkinss@etsu.edu

Scott Orr - sorr@cs.iupui.edu

Elizabeth Zwicky - zwicky@acuitus.com

Steve VanDevender - stevev@uoregon.edu, Univ of Oregon

David Byers - davby@ida.liu.se

Kyrre Begnum - Oslo kyrre@iu.hio.no

Matt Disney - Oslo matthew.disney@iu.hio.no

Terry Countryman - GIT terry@cc.gatech.edu

Mark Burgess - mark@iu.hio.no (afternoon)

Alva Couch - couch@cs.tufts.edu

Rik Farrow - rik@usenix.org

=== Day 3, Monday ===

Mark gave a tutorial in the afternoon and attended the workshop in the afternoon.

==== Notes from the configuration management workshop ====

http://homepages.inf.ed.ac.uk/group/lssconf/config2005/index.html

=== Day 2, Sunday ===

This was a day working on private things.

=== Day 1, Saturday ===

There really is not much happing today, as far as LISA is conserned. We just registered, and use the oportunity to get online as well. We used some of the time earlier today to do some work.

==== Simen's notes ====

Visualization can be used to improve understanding of complex matters. The idea is to create a simple interface that can be used to set up a simple scene. The framework can be controlled with only a few parameters. The interface can presented both as a 3-dimensional and a classic 2D graphical user interface (GUI). After the initial setup of the display, the visualization is controlled by modifying only a small number of parameters. The system will consist of two main parts.

# The visualization engine
# The plug-ins (monitors)

The visualization engine is just a simple interface with which the plug-ins will communicate. The communication will be done through a network socket. This allows for distributed plug-ins. It also allows for the plug-ins to be written in any language, as long as it can send and receive messages through the network.

Each plug-in will typically represent one input to the system. It may represent a single machine or network, or it may represent a property of a machine or network. By using several monitors, a complex picture of the whole system can be monitored with relative ease.

{| cellspacing="3"
|- valign="top"
|width="400" bgcolor="#eeeedd" style="border:1px solid #c5cabd;padding:1em;padding-top:0.5em;"|

==== Mark's summary of our discussion ====

Simen's visualization
---------------------

Visualization can cover everything from the model of policy in terms
of graphs and promises to data observed and measured as consequences
of those choices. Therefore we should combine models from reliability
theory (fault/cause trees) with time series models and anomaly detectors,
logging engines and controllers.

Tie this to a cfengine user base. Combine a policy-based view of
visualization with a control interface for setting policy? A good idea
is to build the research around a prototype that will run on
Unix and Windows. Then people (esp cfenigne users) will start to use
this and we will get a whole bunch of feedback from real users.

Policy-based visualization means being able to decide how events and
states are visualized, so as to maximize the comprehensibilty of the
data. Here are some ideas for visualizing complex data.

# Tunnel vision (3d tunnel - curvature and walls and floor are paramters)
# Fish eye vision (retinax)
# Syslog with multimedia objects

In terms of abstractions, we want to attach multimedia representations to:

# Discrete (graphs, events)
# Continuum inputs (time series etc)

===== Siri + Kyrre overlap project =====

The only example of autonomous agents that is currently in widespread
use is BGP. Make a promise and Maude simulation of BGP. How do
promises get amde in Maude?

Game payoff - how does it fit in?
Payoff is a subjective function of the measurements.

There is an analogy with routing

- Metrics (or utility factor for decision-making)
- Policy (bgp)

OSPF etc are like legacy systems that simply assume cooperation
tacitly. As a result they do not scale well etc. Compare how OSPF and
RIP networks look in terms of promises. RIP only promises its nearest
neighbours. OSPF promises to listen to any flooded message.

Maud is a realization of the policy ideas that is more concrete than
Event Calculus?? We should look at this other work by people in London
using temporal logics and try to relate it.

Kyrre and Siri should formulate a version of promises in Maude
language so we can explain our work to the Oslo group in their
own language.

Promises can be viewed as logic, but the are also a stepping stone to
methods of approximation. e.g. a promise may be subject to noise and
fluctuations from the environment, thus, while the ideas might be
logically consistent in spirit, the details could fall short of this
idea.

How do our projects fit into the magix goals? We want to make a
Maude/promise presentation for the MAGIX/Promise workshop, and also to
engage a discussion with UiO.. This is about networking and about
policy.

|}

===== Specifying networks and systems in Maude =====

[http://maude.cs.uiuc.edu/ Maude] is a formal specification and analysis tool based on rewriting logic.

==== Pictures ====

[[Image:Bilde(64).jpg|10px|thumb|Description]]

The whole gang (except me, who is taking the picture) was joined by Æleen Frish for a dicussion about nothing in particular. Later we went on a litter excursion to somewhere and looked at some seals. It was nice weather. We also ended up eating lunch.

[[Image:LISA_Scenery1.jpg|50px]]
[[Image:LISA_Scenery2.jpg|thumb|Description]]

EU Network of Excellence Activities

Mark — Tue, 27 Dec 2005 11:04:09 GMT

Summary:

<h4>Activities Autumn 2005 </h4>

<table border=1 cellpadding=10>
<tr><td>
22 Aug - -- --- 2005 </td><td> New masters students start Network and System Administration.

<tr><td>
12 Sep - 15 Sep 2005 </td><td> M. Burgess at [[ICINCO 2005]] conference in Barcelona
<tr><td>
14 Sep - 22 Oct 2005 </td><td> Distinguished visitor Aeleen Frisch arrives.
<tr><td>
27 Oct - 28 Oct 2005 </td><td> Norwegian Network Seminar (SIMULA/Oslo)
<tr><td>
23 Oct - 27 Oct 2005 </td><td> M. Burgess at DSOM conference in Barecelona
<tr><td>
04 Des - 09 Des 2005 </td><td> USENIX/LISA conference

</table>

LISA05 Blog

Simenhag — Wed, 30 Nov 2005 15:42:27 GMT

Summary:

== LISA'05 Blog ==

On Friday, December 2nd 2005, four brave members of the staff at Oslo University College depart for San Diego. Their objective:

: To attend the [http://www.usenix.org/events/lisa05/ Large Installation System Administration Conference (LISA '05)].

You can follow their exciting journey, as they eat, sleep and inhale the adventure that is LISA. Watch this space to get updated information as the events unfold.

=== Day 4, Tuesday ===

Below are the DRAFT notes from this year's workshop. Thanks to everyone who was able to make it. Please let me know of corrections.

Thanks,
Steven

9-10 Introductions
- Problems
* How to connect our graduates to jobs
* Scott: How to build good problem-solving instead of recipes
* David: no money for course development -- drives towards
shared platform -> agree on a platform so we could share exercises
more easily
* Elizabeth: testing & assessment; external validation of testing
and assessment: SAGE certification standards/tests
* Matt: advocacy of education efforts
* Steven: how to connect us to others (in particular, to ACM)
* Steven: no web & ftp teaching materials

- Issue that seems common: lack of Windows administration expertise

10-12 Virtualization
- mln "My Linux Network"; Kyrre runs close to 100 hosts with mln. Built
on UML, so there are some stability issues because UML is still heavily
under development
- Two other products in the same space: vulab & vcml (by Zach Lowry: built on
QEMU); VMware or VirtualPC-based solutions are also possible.
- lab issues: logistics (physical space, setup time) x
cost (of hardware & space) = pedagogical payoff
- A solution adopted by everyone would need to be low/no cost and
cross-platform (e.g., VMware, VirtualPC, etc)

Assessment/Testing/Evaluation
- 'group riders' -
* weight group-work less than individual metrics (ie, homeworks 20%,
tests 80%)
* exams-per-assignment (ie, grade per topic is a combination)
* inter-group evaluations/peer evaluations
* add a notebook component per person
- Automated assignment grading
* some are doing it, basically by pre-setting test systems
* Acuitus handles tracking by looking at the following:
- machine's behavior,
- change ontology (ie, machine state),
- student belief state (ie, what the student thinks the problem is)
- David does interviews/demonstrations with each group of students to
do assessment
- David is moving _towards_ recipes _because_ of the 'just google it'
phenomenon (i.e., they search for 'email configuration' and find AIX
docs, even though they are working on, say, Solaris)
- Another common strategy is to give recipes at the beginning and
wean them from them over the course of the program.
- Some do pre-tests and post-tests to show improvement (this is
basically a required approach for accreditation). Some would like to
do 'test-out' approaches for earlier labs, for example. Using such
tests for diagnostic purposes seem to be a good theme. It might be
useful to have a common (ie, across this group) diagnostic test, or
perhaps a modular set.
- Big issues for us: (our) time, dealing with 'out of the ordinary'
approaches for solutions (e.g., instead of fixing the 'broken' box,
break the correct one
to match the broken one), breaking down larger projects into
(checkable) milestones,
- David has a vi tutorial he gives
- Modules people would like to see:
* LDAP (Aeleen Frisch did an LDAP module at Oslo this Fall)
* ActiveDirectory (an equivalent module to the LDAP one)
* Basic stuff: cd, editors, etc
* (Kyrre) a way to more easily and accurately describe overlap
between courses among schools
* Backups (and restores): Curt has some materials that may be
sharable/helpful. Aeleen did a backup module for Oslo also.
* Authentication & Authorization
- External assessment
* a more generic tool like SAGE cerification would be viewed as a very
positive thing
* one view is that a particular exam (within a course or program)
is hard to be exchanged with an external one
* 'nice to have', not necessary
* could be nice as an integrative exam (within a 'track', 'thread',
'concentration' or 'emphasis' program)

12:30-1:30 lunch

1:30-3 ACM IT Curriculum (and community issues)
* sysadm-education recommended curriculum models
* How to connect our graduates to jobs
- bringing students to LISA has mixed success
- internships are a great idea
- Oslo has had some success where students' final
projects (which include an industry customer) ends up
in job offers for those students (ie, job offers from the
customer).
- visibility and credibility (of the field) is an issue
* Advocacy of education efforts
* How to connect us to others (in particular, to ACM)
* Mark suggests a consortium idea -- grouping together to
- apply for grants/funding
- publish papers together
- (collaborate in general)
- Consortium of System Administration
with the overall view of raising visibility of System Administration
education and research
* Oslo:
MS program:
- has graduated one class (of 12)
- second class has ~15 students
- next has ~15 students
- two goals of program: terminal degree _and_ leave open
option of research (ie, PhD)
- currently 2 PhD students, expecting to graduate in 2007
* Alva currently has 4 PhD students, Yizhan Sun (almost finished),
Ning Wu (post-qualifiers), Mark Chiarini (pre-quals, post-courses),
and Hengke Susanto (working on autonomics).

3-3:30 Break

3:30-5 Textbooks & Materials
* Quite a few _are_ using Mark's Principles book as the core
text (both w/in the workshop and outside)
* Scott's school has access to 'Books 24/7'
* Steven's school has access to Safari (safari.oreilly.com)
* Technical book vs textbook: Kyrre sees a difference in quality
of reports in the original sources (recipes vs textbook)
* Curt is using outside reviewers of student papers (eg, students
in a different course will give feedback on the writing)
* Firewalls and Intrusion Detection (Kyrre is using)
- Northcutt's Network Intrusion Detection
- Ziegler's Linux Firewalls

* Ethics: some have a few chapters, but there don't seem to
be any good Ethics texts for System Administration (The Gift
of Fire seems to be fairly popular). Curt shilled 'Ethics and
Computing' by KW Boyer. The SAGE Code of Ethics is too
general and high-level to be the basis for a course in Ethics
for System Administrators.
* Security: Bishop, Secrets and Lies
* Networking:
- Peterson & Davies
- Kurose & Ross - for undergraduates
- Fourazan - good on details
- Tanenbaum
- Leon Garcia's: Communication Networks
- Alva has a lecture called 'Breaking IP'
* Another missing area: how to do write about research.
- problem is missing the scientific context
- Steven needs to dig up the reference on research methods
for CS paper that he has lying in his office somewhere
* Unix Fundamentals:
- Glass' Unix for Programmers and Users
- Mark Sobel's book Unix System V
- It's hard to beat Kernighan and Pike
- David would be interested in collaborating on a book for this
- Mark offerred to let people start from his notes to put
together an online text/version

* Mark offered to let us host our notes & materials on the Oslo
'System Administration Resources' site so that we can more easily
collaborate/share materials (http://www.iu.hio.no/SystemAdmin/). Contact
Matt if we want to put content up.

* Another possibility is the Drupal site set up last year (or conference
wiki?).

* A site for sharing this information might be something useful for the
Consortium to have as a main(?) focus.

* Norway's Network of Excellence called 'Magix' -- they are motivated
to produce teaching resources. They are going to make a virtual
laboratory to share the resources of ~30 universities. They might
be able to assemble a scheme wherein the Consortium could access
the resources as well.

* If someone is interested in looking at other student projects (e.g.,
Master's Theses), some way of sharing those might be something the
Consortium could do.

* Another idea is cross-institution collaboration on projects.
* YAI is doing external audits of curriculum

Thoughts on this year:

* good discussion today
* a paper idea would be to write up issues & challenges (and even a survey)
of virtualization and System Administration, Security, & Distributed
Computing Education (Kyrre and Matt, among others, are interested in that)
* The consortium idea has some support. Oslo will set up a wiki. One goal
would be to have it visually appealing (hint: anyone have a student project
to do web design?)

Attendees:

Curt Freeland - curt@cse.nd.edu

Steven Jenkins - jenkinss@etsu.edu

Scott Orr - sorr@cs.iupui.edu

Elizabeth Zwicky - zwicky@acuitus.com

Steve VanDevender - stevev@uoregon.edu, Univ of Oregon

David Byers - davby@ida.liu.se

Kyrre Begnum - Oslo kyrre@iu.hio.no

Matt Disney - Oslo matthew.disney@iu.hio.no

Terry Countryman - GIT terry@cc.gatech.edu

Mark Burgess - mark@iu.hio.no

Alva Couch - couch@cs.tufts.edu

Rik Farrow - rik@usenix.org

=== Day 3, Monday ===

Notes from the configuration management workshop
at

http://homepages.inf.ed.ac.uk/group/lssconf/config2005/index.html

=== Day 2, Sunday ===

This was a day working on private things.

=== Day 1, Saturday ===

There really is not much happing today, as far as LISA is conserned. We just registered, and use the oportunity to get online as well. We used some of the time earlier today to do some work.

==== Simen's notes ====

Visualization can be used to improve understanding of complex matters. The idea is to create a simple interface that can be used to set up a simple scene. The framework can be controlled with only a few parameters. The interface can presented both as a 3-dimensional and a classic 2D graphical user interface (GUI). After the initial setup of the display, the visualization is controlled by modifying only a small number of parameters. The system will consist of two main parts.

# The visualization engine
# The plug-ins (monitors)

The visualization engine is just a simple interface with which the plug-ins will communicate. The communication will be done through a network socket. This allows for distributed plug-ins. It also allows for the plug-ins to be written in any language, as long as it can send and receive messages through the network.

Each plug-in will typically represent one input to the system. It may represent a single machine or network, or it may represent a property of a machine or network. By using several monitors, a complex picture of the whole system can be monitored with relative ease.

{| cellspacing="3"
|- valign="top"
|width="400" bgcolor="#eeeedd" style="border:1px solid #c5cabd;padding:1em;padding-top:0.5em;"|

==== Mark's summary of our discussion ====

Simen's visualization
---------------------

Visualization can cover everything from the model of policy in terms
of graphs and promises to data observed and measured as consequences
of those choices. Therefore we should combine models from reliability
theory (fault/cause trees) with time series models and anomaly detectors,
logging engines and controllers.

Tie this to a cfengine user base. Combine a policy-based view of
visualization with a control interface for setting policy? A good idea
is to build the research around a prototype that will run on
Unix and Windows. Then people (esp cfenigne users) will start to use
this and we will get a whole bunch of feedback from real users.

Policy-based visualization means being able to decide how events and
states are visualized, so as to maximize the comprehensibilty of the
data. Here are some ideas for visualizing complex data.

# Tunnel vision (3d tunnel - curvature and walls and floor are paramters)
# Fish eye vision (retinax)
# Syslog with multimedia objects

In terms of abstractions, we want to attach multimedia representations to:

# Discrete (graphs, events)
# Continuum inputs (time series etc)

===== Siri + Kyrre overlap project =====

The only example of autonomous agents that is currently in widespread
use is BGP. Make a promise and Maude simulation of BGP. How do
promises get amde in Maude?

Game payoff - how does it fit in?
Payoff is a subjective function of the measurements.

There is an analogy with routing

- Metrics (or utility factor for decision-making)
- Policy (bgp)

OSPF etc are like legacy systems that simply assume cooperation
tacitly. As a result they do not scale well etc. Compare how OSPF and
RIP networks look in terms of promises. RIP only promises its nearest
neighbours. OSPF promises to listen to any flooded message.

Maud is a realization of the policy ideas that is more concrete than
Event Calculus?? We should look at this other work by people in London
using temporal logics and try to relate it.

Kyrre and Siri should formulate a version of promises in Maude
language so we can explain our work to the Oslo group in their
own language.

Promises can be viewed as logic, but the are also a stepping stone to
methods of approximation. e.g. a promise may be subject to noise and
fluctuations from the environment, thus, while the ideas might be
logically consistent in spirit, the details could fall short of this
idea.

How do our projects fit into the magix goals? We want to make a
Maude/promise presentation for the MAGIX/Promise workshop, and also to
engage a discussion with UiO.. This is about networking and about
policy.

|}

===== Specifying networks and systems in Maude =====

[http://maude.cs.uiuc.edu/ Maude] is a formal specification and analysis tool based on rewriting logic.

==== Pictures ====

[[Image:Bilde(64).jpg|10px|thumb|Description]]

The whole gang (except me, who is taking the picture) was joined by Æleen Frish for a dicussion about nothing in particular. Later we went on a litter excursion to somewhere and looked at some seals. It was nice weather. We also ended up eating lunch.

[[Image:LISA_Scenery1.jpg|50px]]
[[Image:LISA_Scenery2.jpg|thumb|Description]]

Online Resources

Mark — Mon, 26 Sep 2005 12:11:51 GMT

Summary:

==Internal to our group==

[http://project.iu.hio.no/ICINCO_2005 ICINCO 2005 Complete Proceedings Online]

- [http://project.iu.hio.no/ICINCO_2005/BPC/Papers/W2_683_Valente.pdf Short cut to annealing paper]

[http://project.iu.hio.no/IM99 IM 1999 Complete Proceedings Online]

[http://project.iu.hio.no/IM2003 IM 2003 Complete Proceedings Online]

[http://project.iu.hio.no/IM2005 IM 2005 Complete Proceedings Online]

Minutes from meeting 20th Sept 2005

Mark — Tue, 20 Sep 2005 10:00:28 GMT

Summary:

==FOU i gruppen==

I den nye modellen for høgskoler og universiteter blir FOU stadig vitkigere for oss. Vi må derfor være flinkere til å drive FoU.

"Time management" er et problem her -- vi er ikke flink nok til å skyve fra oss trivielle forstyrrelser for å fokusere på noe av større betydning.

- Vi lar oss stresse av studenter og admin forstyrrelser.

- Vi må finne roen til å forske og produsere.

- Jobb med FoU når du jobber mest effektivt (tidlig eller sent på dagen)

- Skyv trivia til side, ikke omvendt.

Mark sier: la oss ikke være Afrika, som ikke klarer å løfte seg selv ut av fattigdom pga dårlig management, eller manglende vilje til samarbeid. Vi må bli flinkere til å styre med de ressursene vi har.

==Posters==

Alle bør tenke på a lage en A1 poster om sitt arbeid. Det er overraskende vanskelig men verdifullt av flere grunner.

- Det synliggjør FoU overfor studenter og ledelse

- Det inspirerer studentene at vi jobber med noe mer enn undervisning

- Det forbereder anledninger som åpne dager. forskningstorget osv

- Det motiverer studenter å ta del i vår suksess.

==Ny 6th framework prosjekt==

Mark er blitt bedt om vi vil være med på nok et EU prosjekt, av et konsortium av universiteter og firmaer fra EU/Norge. Gruppen tilpasser Linux til spesielle formål. Skolelinux er med, og NOKIA, og et italiensk firma som lager musikk-applikasjoner for Linux.

Dette kan bety penger for videreutvikling av cfengine til tilpasning av konfigurasjoner til spesielle formål.

Søknaden sendes inn nå.

==Post 45 og høyvolumtjeneste lab==

Jeg har søkt om midler til å etablere en høyvolumtjeneste lab. Vi trenger kompetanse om storskala webtjenester osv. Opera software vil gjerne være med på dette, og det er kjerne kompetanse som IU bør ha. Kanskje andre firmaer (IBM..) også vil delta.

==FoU Admin==

Narve Strand er ansatt i admin til å bruke tid på FoU admin. Hva slags tjenester ønsker vi av FoU admin? Hvordan kan dette lette belastningen for oss og hjelpe oss til å løfte oss selv ut av `fattigdom'.

==Revival of seminar==

Vi kan kanskje holde seminarer i 8 etg på møterommet, vha prosjektor. Disse møtene er viktig for å holde liv i FoU.

==MARS - multi agent robotic systems==

Seminar som Mark var på i Barcelona. Dette var et fokus for all den kunnskapen vi har. Veldig (overraskende) nyttig.

- Kontrollsystemer / agenter (cfengine)

- Logikk (belief systems/modallogikk)

- Signalbehandling (anomalidetekjson)

==Til slutt==

Som en konklusjon: alle bør planlegge en poster.

Alle bør tenke seg hvordan de kan bedre bruke sin tid for å produsere FoU.

Group stuff

Mark — Tue, 20 Sep 2005 09:42:07 GMT

Summary:

[[Minutes from meeting 20th Sept 2005]]

ICINCO 2005

Mark — Sun, 18 Sep 2005 16:06:41 GMT

Summary:

This is a very brief conference report. Mark presented a poster/short paper on cfengine as a multiagent software robot.

A full proceedings is available on CDROM.

This was a surprisingly relevant conference. I went to see what robotics can teach us about configuration management, and what we might be able to teac them. I found a lot of interest. The conference had two parts of special interest:

==MARS - multi-agent robotic systems==

- Here I met [http://www.doc.ic.ac.uk/~klc Keith Clark] of imperial college and his interesting language related work. We talked about examining the language design of cfengine, casting an expert eye over our models. He has interesting work which we should look at.

- I also met Joaquim Filipe - who works on the logics of agents, belief systems etc. He was keen to stay in touch -- he would be a great person to invite, linking together our group's work on logic and agents.

This part of the conference shows how logic and control systems fit neatly
together and can contribute to system administration through agent behavioural models and language descriptions.

- There is an interesting and very friendly group at the Technion in Israel.

- Graph notions of "rigidity" and "persistence" used in metric constraint spaces mirror our work on "Promises".

==Signal processing==

Here there were many ideas presented that relate to anomaly detection. Signals could be digital or pseudo-continuous.

- Use of simulated annealing as a smoothing method on offline data. This is something that Hårek and Kyrre could/should collaborate on. Hårek has a lot of expertise in statistical mechanics.

- Independent component analysis, distinct from PCA. Is this a useable technique.

Open questions: people are good at doing off-line analyses, but can these techniques be translated into realtime methods?

From our side, cfengine was received with interest.

(Mark)

Chapter: Use of Web Services in management

Mark — Thu, 15 Sep 2005 18:21:23 GMT

Summary:

Use of Web Services in management
---------------------------------

1. Motivation
- Evolution of SNMP failed
- No need for dedicated technologies
- Many skilled developers
- Many tools

2. Introduction to Web Services
- Web services components
- Protocol stack
- Data types
- Main w3c specifications
- Structure WSDL definition

3. Web Services for management
- Example of Web Services based management
- Operations
- Parameters

4. Advanced features
- Transactions
- Security
- Choreography / Orchestration

5. Standardization activities
- OASIS WSDM, WSRF
- OGSI
- Parley-X
- IRTF-NMRG

6. Performance
- Comparison to SNMP

Master level tutorials and tutors

Mark — Wed, 07 Sep 2005 06:33:34 GMT

Summary:

The purpose of a tutor is to guide students through their college duration with advice about learning. It is not to teach course material, but rather answer and pose questions to the students that enable students to get the most out of their stay.

==Tutorial==

A tutorial is a private meeting, not a lecture. It is more informal. Students could bring coffee or tea to such a meeting, but they should come expecting to work. Tutorials are important for creating the feeling of more individual contact between student and teacher. Ideally tutorials would be conducted in groups of two or three, but this is not feasible in terms of resources.

==Group Tutorials==

Each class (year) should have a tutorial once every two weeks in which at least one of the staff is present. The tutorials should enourage participation.

<ol>
<li>The tutorials should measure attendance.

<li>In the first session, all students should present themselves: where are they from, what are their ambitions and expectations of the course? It is important for each student to know that they are expected to communicate and give of themselves, not sit passively in classes. The tutorials should aim to help break through the shyness barrier.

<li>Advise students to practice mathematics and writing skills.

<li>What can each student contribute top the group: special skills? Social activities? Job experience?

<li>The tutorial staff member(s) can pose general questions about the course materials that they want to test the students on in an "ad hoc" way. This could be to provoke a special point of view, or demonstrate an example of something.

<li>Tutorials may be conducted as field trips.

<li>Tutorials should help students in the area of "time-management". How can they organize their time efficiently?

<li>Students can be given responsibility for organizing events like seminars or industry days.

<li>Students could be given responsibility for organizing study groups. e.g. each student in the class organizes one group per week in which the important topics of the week are summarized.

</ol>

==Tutors==

Tutors are assigned to give personal and general advice to students, not to solve course problems. Tutors should keep a close contact with the administration and study board to advise students in personal matters.

Tutoring is an intensive and tiring process. We should aim to make students self-sufficient, not molly-coddle them!

==Field trips==

Visits and field trips should be encouraged. Students should be reminded that they represent the College in all respects. They should dress and behave accordingly.

MAGIX Network of Excellence Activities

Mark — Thu, 18 Aug 2005 14:57:31 GMT

Summary:

<h4>Activities Autumn 2004 </h4>

<table border=1 cellpadding=10>
<tr><td>
22 Aug - -- --- 2005 </td><td> New masters students start Network and System Administration.

<tr><td>
12 Sep - 15 Sep 2005 </td><td> M. Burgess at [[ICINCO 2005]] conference in Barcelona
<tr><td>
14 Sep - 22 Oct 2005 </td><td> Distinguished visitor Aeleen Frisch arrives.
<tr><td>
27 Oct - 28 Oct 2005 </td><td> Norwegian Network Seminar (SIMULA/Oslo)
<tr><td>
23 Oct - 27 Oct 2005 </td><td> M. Burgess at DSOM conference in Barecelona
<tr><td>
04 Des - 09 Des 2005 </td><td> USENIX/LISA conference

</table>

Chapter: Licensing agreements

Mark — Thu, 11 Aug 2005 10:49:29 GMT

Summary:

Synopsis:

Summary of different software licensing schemes, covering both
typical commercial licenses (e.g. Microsoft "Select"), including
dedicated, floating, and site licenses, as well as widespread
"Free software" licenses such as GPL, (new) BSD, CC, MIT, etc.
Discussion of social, contractual, legal, managerial and
oranizational implications of different licensing schemes for
both users and network/system adminstrators. Also som
discussion on the implications of the increasing use of
technical measures for IP-management and to ensure license
compliance (e.g. technologies such as NGSCP, and product
activation keys, as used by Adobe and Microsoft.)

Chapter: Ontologies and environment

Mark — Mon, 27 Jun 2005 05:23:15 GMT

Summary:

Knowledge Engineering - Models and Ontologies

1. Introduction and Motivation

1.1 Reconnect the Business with the Network

1.2 Managing Telecom Systems: Elements, Networks, OSSs and BSSs

1.3 Managing Non-Telecom Systems: Storage Systems

1.4 The Role of Policy Management and Process Automation

1.5 Stovepipe Components, Stovepipe Knowledge

1.6 Knowledge Lifecycle Management

2. Current Approaches in Network and System Management

2.1 Legacy Systems and Their Problems

2.2 Standards Efforts

2.3 Newer Management Architectural Styles and Their Problems

3. Constructing a Knowledge Framework

3.1 Types of Knowledge

3.2 Contextual Knowledge

3.3 Connecting Models, Knowledge and Semantics

4. Using Knowledge to Orchestrate Behavior

5. Conclusions and References

Chapter: System Administration and the Business Process

Mark — Sat, 25 Jun 2005 08:37:26 GMT

Summary:

This chapter will be about integrating system administration into business models. Its placement and contents will depend a little on some of the
other contributions.

Chapter: Outsourcing

Mark — Mon, 13 Jun 2005 10:53:36 GMT

Summary:

Decision and Control factors for IT-sourcing

Dr. Guus Delen

ABSTRACT
In my ten years of experience as outsourcing consultant I have grown interested by the paradox that still many IT-outsourcings fail, whilst in that period ample experience has been built up. This chapter gives a summary of my work on that subject, in which two research questions are addressed:
1. the scientific quest for finding which factors influence success or failure of a sourcing.
2. the consultants question on how to apply the understanding gained on those factors in order to raise the success of actual sourcing trajects
This chapter starts with an introduction and definitions of the various types of sourcing. The second section gives an elaboration on the development, and economic relevance of IT-sourcing. Thereafter I threat the set of 10 Sourcing factors which was my start-hypothesis for the research. This set has been validated in a field survey. Section 4 summarises the result if this validation, which shows that 9 of the 10 factors are relevant for sourcing success, that this set is rather complete. The validation also resulted in an indication of the relative strength of each specific factor. In the last section the Sourcing-factors monitor is introduced which makes use of the apparent strength of the 9 relevant factors in order to predict and monitor sourcing success in actual cases.

Chapter: System administration and business processes: ITIL

Mark — Fri, 10 Jun 2005 11:15:34 GMT

Summary:

Abstract of ITIL chapter in Handbook NSA

IT service management is becoming more important due to a number of reasons. The most important of these reasons are the still growing use of ICT within the primary process of organizations, The growth of ICT outsourcing, the pressure on accountability for the IT of organization, (propelled by SOX). All these trends provide a pressure on the (internal or external) ICT service provider to formalize their service offerings and act as a business

ITIL has become the de-facto standard for ICT service management processes. ITIL is a set of best practices that can help an organization with structuring and standardizing the management of their technical ICT infrastructure.

I propose the following structure for the chapter:

A definition of IT services and service management

a short history of the origin of ITIL, the development of the best practices and the growth of worldwide usage of ITIL

a summary of the two sets of most widely used ITIL processes: the service delivery set (Incident management, Problem management, Configuration management, Change management, Release management and service desk) and the service support set (Capacity management, IS service continuity management, Availability management, service level management, financial management)
Positioning of ITIL between other models like CMM, COBIT, ASL, BiSL, MOF, E-TOM

Shortcomings or misperceptions of ITIL and the limits of ITIL

Implementing ITIL processes, do’s and don’ts

The human factor in IT service management

Chapter: Use of XML in management

Mark — Thu, 02 Jun 2005 04:49:02 GMT

Summary:

1.What is XML-based Network Management?

1.1Definition of XNM

1.2Merits and Demerits of XNM

1.3Related Work

1.4Chapter Organization

2Background Knowledge

2.1XML Technologies

XML, DOM &SAX, XPath, WSDL, SOAP, UDDI, etc

2.2Network Management Tasks

3Application of XML technologies to Network Management

3.1Information Model

3.2Information Instrumentation

3.3Communication

3.4Analysis

3.5Presentation

4Architecture of XNMS

4.1Overall Architecture

4.2Manager Architecture

4.3Agent Architecture

4.4Gateway Architecture

5Implementation Experience

5.1Manager

5.2Agent

5.3Gateway

6Summary & Future Work

7References

Eclipse, JBuilder og IDE i opplæringen

Simenhag — Thu, 26 May 2005 23:04:42 GMT

Summary:

''This page has been removed''

Chapter: Backup and redundancy

Mark — Mon, 23 May 2005 14:02:36 GMT

Summary:

Backups

Backup Strategies

Fundamental Entities

Tactical Considerations

Snapshots

Scope: Single system vs. Enterprise approaches

Resource Requirements

Media for Backups: Tape and Other Technologies

Backup Subtleties (open files and other reliability issues)

Chapter: Storage Technologies

Mark — Mon, 23 May 2005 14:02:02 GMT

Summary:

Storage: Technologies and Best Practices

Introduction: A Brief Historical Perspective

Computing Without Permanent Storage

Current Storage Technologies

A Generalized I/O Subsystem

Buses: PATA, SATA, SCSI

Controller and Disk Technology

RAID

NATs

SANs

Storage in Use

File Systems

Network I/O Considerations

I/O Performance

Data Access Patterns

Data Placement

Evas forslag

Evav — Mon, 23 May 2005 12:53:33 GMT

Summary:

Jeg tror det er helt nødvendig at studentene får noe programmeringserfaring FØR de lærer om forskjellige typer programmeringsspråk. Hvilket språk dette er spiller ikke så stor rolle.
Mitt forslag er at studentene kan følge JAVA-undervisningen første halvdel av kurset
(t.o.m. uke 42, jfr. forelesningsnotater og fremdriftsplanen t.o.m. kap.7) og når ingeniør og IKT-studentene går videre med objektorientering i JAVA, går studentene i Programmeringsspråk istedet over til å lære om flere (typer) språk. Da har de forhåpentligvis lært såpass om programmeing generelt at de har tilstrekkelig med knagger å kunnskapen om nye språk på.
Med hensyn til oppgaver og øvinger kan det lages et eget opplegg tilpasset disse studentene.

Link til forelesningsnotater:
http://www.iu.hio.no/~kjetilg/uvstoff/intro/progintro.html

Link til fremdriftsplan:
http://www.iu.hio.no/~kjetilg/Undervisningsstoff/h2001/uvplan.htm#fplan

Kirstens forslag

Kirstenr — Fri, 20 May 2005 13:02:18 GMT

Summary:

Kurset skal gi en teoretisk og praktisk innføring i programmering og ulike programmeringsspråk.
Målet er å tilegne seg en begrepsmessig forståelse gjenom praktiske øvelser.
Kurset inneholder følgende:

''' Generelt om programmering:''' </br>

<UL>
<LI>
Om utvikling av programvare fra problemdefinisjon til ferdig produkt</LI>
<LI>Feilsøking og retting, testing</LI>
<LI>Vedlikeholdbar og utvidbar kode. Kodestandarder.</LI>
<LI>Utviklingsverktøy og utviklingsmiljøer.</LI>
<LI>Ulike plattformer.</LI>
<LI>Sammenligning av programmeringsspråk:
Objektorienterte vs. prosedyreorienterte språk.</LI>
</UL>

'''En oversikt over populære programmeringsspråk'''
<UL>
<LI>Java</LI>
<LI>PHP</LI>
<LI>C#</LI>

.......
<LI>Scriptspråk</LI>
</UL>
'''Gøy med programmering'''
<UL>

<LI>Websider med (X)HTML og litt funksjonalitet - Forms, javascript (samarbeid med kurset 'Webprosjekt')</LI>
<LI>Lego Mindstorms: Bygg og programmer en robot, og se programmet kjøre.</LI>
</UL>
'''Hands-on: Bruk av et eller to språk i praksis (feks PHP, java)'''

Øvelser i:

<UL>
<LI>å skrive små programsnutter fra første dag</LI>
<LI>
hvordan kompilere og kjøre et program </LI>
<LI>hvordan bruke en debugger</LI>
<LI>hvordan skrive funksjoner</LI>
<LI>hvordan definere variabler </LI>
<LI>hvordan utføre enkle matematiske beregninger</LI>
<LI>if/else </LI>
<LI>for og while-løkker</LI>
<LI> hvordan lage input/output</LI>
<LI>bruk av arrays</LI>
</UL>

Frodes forslag

Mark — Fri, 20 May 2005 12:21:25 GMT

Summary:

== Data typer (frode tøff type) ==

Demonstrere i praksis hva typer er og viktigheten av og ha typer.

Demonstrere begrensingene av typer - de er tilnærminger av virkeligheten
o integere som ruller rundt
o integere som ikke kan representere hele den dynamiske bredden som trenges - dvs norsk statsbudsjett og kostnadene til en riegel
o flyttall som har begrenset bredde, demonstrere at a+b = a i enkelte tilfeller når a er veldig stor og b er veldig liten men ikke null.

Konvertering fra en type til en annen, konsekvenser
gutt+gutt != barn
pike+gutt == barn

Atomiske typer i motsetning til sammensatte typer.
strenger, representasjon
arrayer
stakker
køer

gøy med arrayer stakker og køer, mange mange eksempler må viser for å banke inn i hodene en intuitiv forståelse av disse viktige datatypene og hvor de skal brukes

to dimensjonale arrayer, representasjon av bilder (burde være visuelt og fengende)
-konkuranse, la studentene "oppdage" run-line enkoding (enkel kompresjon)

Objecter og records kan passe inn her, men bør kanskje bare nevnes og presenteres separat.

Chapter: Anomaly Detection

Mark — Wed, 18 May 2005 07:05:40 GMT

Summary:

The "Anomaly Detection" chapter

* Introduction and Motivations:
--> what is anomaly detection and its applications
--> confusions and misunderstanding
--> misuse of the term

* Basic Anomaly Detection Techniques:
--> Statistic-based Approaches
--> Specification-based Approaches
--> False positives versus false negatives

* Profiling the Features:
--> Feature Selection
--> The Machine Learning Approaches

* Performance Evaluation:
--> Analytical Approaches (e.g., Entropy-based Metrics)
--> Data-centric Benchmarks and Experiments (e.g., Data set)
--> Data Manipulation for Anomaly Detection

* Issues: The usefulness of Anomaly Detection
--> On Efficiently/Practically Building the Long-term Profiles
--> Explanation of Detected Anomalies
--> Responing the Anomalies

* Examples: the Systems
--> two or three examples such as (Self/Why 6), NIDES, Elisha.

* Summary and Future Research Topics

Nytt fag: Programmeringsspråk (Anvendt Datateknologi

Frodes — Tue, 17 May 2005 16:03:27 GMT

Summary:

Simens forslag

Simenhag — Fri, 13 May 2005 12:11:50 GMT

Summary:

Jeg tror vi må lage et opplegg hvor studentene kan få en enkel og grei forståelse for hva et programmeringsspråk er, hva det kan brukes til og hva det ''ikke'' kan brukes til. Her er noen forslag til emner jeg mener vi må ha med, og som kan flettes sammen med/overlapper med det Mark har skrevet:

* Hva er et programmeringsspråk
* Hva slags typer progarmmeringsspåk finnes
* Forskjell på kildekode og program (exe)
** Compilerte språk
*** Hva er kompilering? C++ -> obj, java -> bytecode
*** Hva er linking? obj -> exe
** Interpreterte språk
** Eksempler på noen språk (det finnes minst 2500, se [http://www.oreilly.com/pub/a/oreilly/news/languageposter_0504.html The History of Programming Languages] og [http://people.ku.edu/~nkinners/LangList/Extras/langlist.htm The Language List])
*** '''Assembly'''
*** '''C/C++'''
*** '''Java'''
*** '''C#'''
*** Simula
*** '''PHP'''
*** '''Javascript'''
*** '''Pascal'''
*** ADA
*** Delphi
*** Basic (VisualBasic)
*** VBScript
*** '''Perl'''
*** Tcl/Tk
*** Python
*** Ruby
*** bash/tcsh/sh
*** APL
*** '''Postscript'''
*** Cobol
*** Fortran
*** Prolog
*** Lisp
** Variabler
** Tester
** Løkker
**
** Struktruer og algoritmer
*** Her lager vi et enkelt opplegg, og introduserer kun enkle algoritmer og strukturer. Kanskje en enkel sorteringsalgoritme ([http://www.cs.princeton.edu/~ah/alg_anim/gawain-4.0/BubbleSort.html bubble sort?]) eller noe slikt.
* Hva er et API og hva kan det brukes til?
* Hva er et GUI og hvordan kan jeg lage et?
* Objektorientering
* Programstack
* Minnehåndtering i forskjellige språk

Mark's forslag

Mark — Fri, 13 May 2005 09:40:25 GMT

Summary:

<ol>
<li>Introduksjon: hva er programmering? [[PS-Oppgaver 1]]

<ul>
<li>Litt historie?
<li>Hva er en operasjon?
<li>Intruksjoner som koder
<li>Input/output
<li>Hva er det som tolker?
<li>Er DNA et dataprogram?
</ul>

<li>Datamaskinens oppbygning [[PS-Oppgaver 2]]

<ul>
<li>Litt historie
<li>CPU, Minne, Disk, skjerm, adresser
<li>Modellen: variable <-> minne, CPU <-> operatorer, skjerm/disk <-> streams/files
</ul>

<li>Maskinspråk, tolket og kompilerte språk [[PS-Oppgaver 3]]

<ul>
<li> Løkker (hvordan ser de ut: JMP, while, ...)
<li> Subrutiner - hoppe i en nummerert liste, stack
<li> Avgjørelser - forgrening, veivalg
<li> Informasjonsflyt i et program: variable/minne
</ul>

<li>Unix shell og DOS ??

<ul>
<li> Sammensetninger av programmer som kommuniserer
<li> Pipe tolket som en subrutine
<li> Løkker, subrutiner, avgjørelser
<li> Informasjonsflyt i et program
<li>Scripting generelt
</ul>

<li>Høynivåspråk

<ul>
<li>Datatyper - hva er de?
<li>Interpreterte/kompilerte språk
<li> Fordeler og ulemper med kompilasjon/tolkning
<li> Pascal,C,Java - hva brukes de til?
<li> Vis: Prolog/ML - hva brukes de til?
<li> Nevne: Lisp, COBOL - hva brukes de til?
</ul>

<li>Organisering av et program

<ul>
<li>Hvordan håndtere kompleksitet i intrukser?
<li>Procedural
<li>Objektorientert
<li>Hendelsesbasert
</ul>

<li>Et enkelt dataspråk: (velg en...C?)

<li>Lokale og globale variable

<li>

<li>

<li>
<li>
<li>Deklarasjonsspråk: HTML,XML
<li>Hvor går fremtidens språk?
</ol>

PS-Oppgaver 1

Mark — Thu, 12 May 2005 12:06:25 GMT

Summary:

Ideer...

<ol>

<li>Naturlig språk i vesten bruker bokstaver til å representere lyder,
slik at sammensetninger lage fonetiske ord. Finn ut hvordan tegn
brukes i:
<ol>
<li>Egyptiske hieroglypher
<li>Kinesiske symboler
</ol>

<li>Lag pseudokode av en typisk dag på IU... hva er operasjonene? Hvilke abstraksjoner kan man lage for repetisjon osv..

</ol>

Nytt fag: Programmeringsspråk (Anvendt Datateknologi)

Mark — Thu, 12 May 2005 11:02:54 GMT

Summary:

Målet med dette kurset er å gi en grunnleggende forståelse av hva programmering er og dreier seg om, for studenter med svak bakgrunn i data og matematikk.

<b>Planlegging:</b>

- Mark
- Simen
- Ulf
- Frode
- Geir
- Hårek
- Eva

<b>Undervisning:</b>

- Mark
- Simen
- Hårek
- Frode

<b>Lage oppgaver</b>

- Mark
- Hårek
- Simen
- Frode

{| cellspacing="3"
|- valign="top"
|width="700" bgcolor="#eeeedd" style="border:1px solid #c5cabd;padding:1em;padding-top:0.5em;"|

<h4>Status 10 Juni</h4>

Ved siste møtet, gikk vi in for [http://www.iu.hio.no/teaching/materials/AD001A et opplegg].

<h4>Status 25 Mai</h4>

Ved siste møtet bestemte vi oss for at kurset skulle kjøres delvis sammen med vanlig programmeringskurs, for å spare ressurser, i hvertfall i det første året. (Senere kan vi revurdere dette).

-''4 timer forelesninger sammen med Eva/KJetil per uke, i 7 uker.'' Etter disse ukene, skulle vi slutte med disse timene og fortsette kun med de neste punktene.

-''2 timer oversikts/inspirasjonsforelesninger per uke (samkjørt med introkurs).''

''-2 timer med lab/øvinger med egne oppgaver.''

Det ble bestemt at et eget oppgavesett skulle utvikles, hvor man eventuelt plukker fra faget til Eva/Kjetil, og legger til andre oppgaver. Disse oppgavene må trene studentene hardt i grunnleggende ferdigheter og samtidig få dem til å reflektere over begreper.

'''Til neste møte 9. Juni''': skaffe aktuelle bøker, og fortsette å skissere opplegg.

<font color=red>Det er fortsatt ikke bestemt noen lærer i dette faget.</font>

<h4>Forslag til emner</h4>

[[Mark's forslag]]

[[Simens forslag]]

[[Evas forslag]]

[[Frodes forslag]]

[[Kirstens forslag]]

<h4>Links</h4>

There are nice introductions in English about programming languages in all kinds of places.

[[http://eu.wiley.com/WileyCDA/WileyTitle/productCd-0470864125.html The Encyclopaedia of Computer Science]]

[[http://en.wikipedia.org/wiki/Programming_language Wikipaedia definition of programming language]]

[[http://cgibin.erols.com/ziring/cgi-bin/cep/cep.pl Dictionary of languages]]

[[http://www.cs.iastate.edu/~leavens/teaching-prog-lang/home.html Teaching programming languages]]

[[http://www.levenez.com/lang/ Timeline of languages]]

|}

Åpne samarbeidsprosjekter

Mark — Thu, 12 May 2005 10:54:51 GMT

Summary:

''Denne siden brukes som base for utviklingsprosjekter på gruppa, hvor man trenger felles tilgang til åpne informasjon.''

# [[Nytt fag: Programmeringsspråk (Anvendt Datateknologi)]]
# [[Eclipse, JBuilder og IDE i opplæringen]]

Chapter: Network Configuration Management

Mark — Tue, 03 May 2005 05:13:24 GMT

Summary:

Network Configuration Management (outline)

J. Schoenwaelder

1. Introduction to network configuration management

- Terminology and concepts:
- configuration vs. provisioning
- device-local vs. network-wide configuration
- configuration state vs. operational state
- configuration partitioning and locking
- configuration distribution vs. activation

- Approaches:
- Ad-hoc configuration management
- Ad-hoc configuration management with change tracking
- Controlled centralized configuration management
- Autonomic configuration management

- Turning high-level policies into configurations

2. Network management technologies and approaches

- variable-oriented approaches (e.g. SNMP)
- command-oriented approaches (e.g. CLIs)
- object-oriented approaches (e.g. Corba)
- document-oriented approaches (e.g. NetConf)

3. Configuration management transactions

- Impact of Configuration Transactions
- change of configuration state in the target(s)
- change of operational state in the target(s)
- change of operational state in other nodes

- Coordination and locking

- Commit and rollback mechanisms
- hard rollback via reset and reload
- smart least-disruptive rollback
- commit and rollback primitives

3. NetConf Approach

- Architecture:
- layers (secure transport, rpc, operations, config data)

- Protocol details:
- operations
- capabilities
- framing and rpc layer
- transport mappings

- Missing pieces:
- Authorization
- Data modeling
- Transition

4. Conclusion and outlook

- Robust and portable (enterprise) network configuration managers
- Policy translation and translators
- Integration (EPP, XCAP, ...)

Chapter: Management of wireless and ad hoc

Mark — Mon, 25 Apr 2005 09:43:41 GMT

Summary:

Synopsis: Management of Ad-Hoc Networks

Remi Badonnel, Radu State, Olivier Festor

This document presents the synopsis of a chapter on ad-hoc networking for the handbook on network management proposed by Mark Burgess. This document is structured as follows. We introduce mobile ad-hoc networks in Section 1 and highlight the management requirements (what to manage) in Section 2. We present in Section 3 the main management architectures for these networks and detail the management functional objectives illustrated by management solutions in Section 4. Finally, we conclude that chapter in Section 5, with a synthesis and an analysis of research orientations in the network reseach community as well as in the network management research community.

1 Introduction Mobile ad-hoc networks (MANETs) are networks spontaneously deployed from a set of mobile devices without requiring any pre-existing xed infrastructures [24, 39, 28]. The increasing interest in this technology raises new research challenges towards providing them a management framework. These networks are strongly dynamic, composed of heterogeneous devices, constrainted in terms of bandwidth and energy. Current management paradigms and tools must be reconsidered to cope with these characteristics... (1 page)

2 Management Requirements This section describes the management requirements of network providers and users, in order to evaluate the operational state of MANETs.

2.1 Network Space-Time Behavior An ad-hoc network is composed of devices moving in a given surface. The mobility of ad-hoc nodes leads to de ne the space-time behavior of ad-hoc networks. Network topology [4, 13, 17] informs on either geographically and neighborhood relationships among nodes. Global positioning systems or alternative systems of relative coordinates [9, 27] can be used to determine node position and node density in the service area. These information in the time scale can be employed to nd out the most approaching mobility model of the current network, to predict the network topology in a provisioning approach, and to detect intermittent and isolated nodes. Besides, connectivity (one-hop [34, 14] and end-to-end [18, 3]) conveys node relationships and network robustness... (1 page)

2.2 Network Tra c Behavior The ad-hoc network behavior is also de ned by the network tra c and communications [36]. Tra c statistics cannot be divided in subnetworks as in a xed infrastructure and must take into account the routing activity for each network node. The ad-hoc network organization requires tra c distinctions according to new scales: geographic-de ned, cluster-de ned, probe-de ned or neighborhoodde ned. The signi cant constraint in bandwidth impacts on the de nition of quality of service models. [42, 44, 20]...

2.3 Node Fairness and Misbehavior The network ressources are shared in ad-hoc networks. The channel access as well as the routing participation of nodes must be estimated to detect misbehavior and ensure fairness among nodes. Some nodes may not respect the medium access control protocol or may refuse to forward packets: to save their own resources (energy and CPU) [26] and to maintain a monopoly of shared ressources (bandwidth)... (1 page)

3 Management Architectures This section presents the main management architectures for mobile ad-hoc networks.

3.1 Interaction Model Ad-hoc networks require a management plane, adaptable towards node heterogeneity and mobility, robust towards scalability and critical environments, and low-cost towards the constrained ressources. De ning a management plane goes through the selection of convenient interaction models and the impact evaluation of management [32]. An excellent introduction to model ad-hoc networks and evaluating the management interaction models in such networks was outlined in [5, 6]... (1.5 pages - Burgess approach will be detailed.)

3.2 SNMP-based Management Integrating ad-hoc networks into common agent/manager models requires a reorganization of the management plane, often based on clustering. ANMP approach, de ned in [12], proposes an SNMP-compatible architecture based on a three-level hierarchy (manager, cluster heads and simple nodes). It de nes an information model and can be extended to several clustering algorithms. The GUERILLA framework in [35] is a self-management approach where the management intelligence is spread over the network nodes according to their capabilities. This two-tier architecture proposal is composed of nomadic managers (higher tier), that maintain connectivity in the management plane with other managers, and active probes (lower tier), which perform the localized management operations... (2 pages - ANMP approach will be detailed.)

3.3 Policy-based Management Policy-based management approaches are proposed in [23, 29, 30, 11]. The management framework in [29, 30] lies in decentralizing an hybrid provisioning and outsourcing model called COPS-PR using clusters, and introduces three key mechanisms: service location discovery, cluster management, dynamic service redundancy. The approach of [23] is based on service di erentiation where node policy databases are synchronized when two network nodes meet. DRAMA [11, 10] de nes a more centralized architecture but considers realistic use cases and interesting prototype developement... (2 pages - Phanse approach will be detailed.)

4 Management Functional Objectives This section details some valuable management functional objectives in MANETs.

4.1 Monitoring Monitoring ad-hoc networks resides in measuring metrics, collecting the measurement data and processing these measurements. WANMon describes in [26] a monitoring approach for determing the ressources usage (network usage, power usage, memory usage and CPU usage) for a local ad-hoc node. The DAMON approach [31] de nes a distributed monitoring system including seamless support for multiple repositories and auto-discovery of sinks by agents to store measurement data. [15] de nes metrics to characterize the performance of adhoc networks based on timescales for information ow. Distributed ltering, including graph dependency analysis [5], are applied to network tra c data in [2]. Corrective management operations can be performed from this monitoring to achieve QoS guarantees [1]... (2 pages - DAMON approach will be detailed.)

4.2 Con guration The con guration of mobile ad-hoc networks [33] must be performed in a selforganized manner to ensure scalability. A basic expectation in con guring adhoc network is the autocon guration of IP addresses [41, 22]. Two allocation approaches are usually highlighted: con ict-detection allocation and con ictfree allocation. The rst one [25] consists in selecting a potentially-unused IP address and to request the network con gured nodes if this address is already allocated. The second one [43] lies in ensuring the IP address is not yet allocated using disjoint address pools. A third alternative best-e ort solution is to allocate an IP address to a node until a con ict is detected [38, 40]... (2.5 pages - MANETconf and PROPHET approaches will be detailed.)

4.3 Accounting The operation of an ad-hoc network is based on node equitable cooperation. The node participation, such as medium access and packet routing (on the behalf of others), must be controlled to ensure fairness among them. Solutions de ned in [19, 37] detect sel sh hosts and limits an unfair share of the channel bandwidth. At the routing level, [21] details how to mitigate the routing misbehavior using a watchdog that identi es misbehaving nodes and a pathrater that helps routing protocols avoid these nodes. A virtual currency (called nuglets) is introduced in [7, 8] with two modes (packet purse and packet trade) to stimulate routing participation among nodes... (2.5 pages - NUGLETS approach will be detailed.)

5 Conclusion Ad-hoc networks are commonly considered as a subset of regular networks. However, the dynamic of future networks may reverse the tendency: future regular networks "seen" as a subset of ad-hoc networks. The described management approaches permit, in this way, to optimize the management of current and future networks and to tame the edge network management crisis [16]...

Virtual visualization project

Simenhag — Sat, 23 Apr 2005 16:04:38 GMT

Summary:

Today, system administartors have to relay on mainly text oriented tools in order to build and maintain a computer network. For large and complicated network, where portable devices is added and removed all the time, keeping track of everything that happens in the network, might be an close to impossible task.

This projects aim is to simplify the process of building and manage networks by using different types of visualization.

More info to follow.

Chapter: Routing

Mark — Tue, 19 Apr 2005 07:54:01 GMT

Summary:

Chapter on routing, bridging, and switching

This chapter covers what the intrinsic differences are between layer 3
forwarding
("routing") and layer 2 forwarding ("bridging"). For layer 2 forwarding
it discusses
the spanning tree algorithm and source route bridging. For layer 3
forwarding it
discusses link state, distance vector, and path vector routing. It also
discusses
address format (IPv4, IPv6, as well as other protocols such as CLNP and
Appletalk).

Chapter: Diagnostic and fault finding methods

Mark — Tue, 19 Apr 2005 07:53:06 GMT

Summary:

Synopsis

1) A real-world example of system analysis

2) Parallell components and serial components

3) Describing the system - fault trees

4) Calculating probabilities for fault trees

5) Describing the system - the general approach

6) Other methods of system descriptions - graphs

7) Calculating the probability for graphs

8) A real-world example of a graph layout

9) Measures of component importance

Chapter: Control theory and management

Mark — Mon, 11 Apr 2005 20:39:18 GMT

Summary:

A Blueprint for Automating Systems Administration

Aaron B. Brown, Joseph L. Hellerstein, and Alexander Keller

The cost of systems administration often exceeds the cost of Information Technology (IT) systems. Our belief is that automating systems administration can reduce these costs and increase the business value provided by IT by making IT more responsive to business needs.
Since much of system administration revolves around managing change, we focus on the automation needed to deploy changes into an IT infrastructure. This focus is supported by market studies that confirm that Change Management is a central concern for system administrators. Our view is that deploying changes is a process that can be described as a workflow of multiple tasks with various decision points and loops. We show how to analyze the process to identify parts that contribute most to complexity and hence to the cost of system administration. To reduce cost, complex parts of the process must be revised and replaced with automation. In some cases, revised processes can be generated—and often executed—automatically based on dependency information (e.g., the operating system must be installed before the Java virtual machine). We use the term planning to refer to the automated generation of workflows. These workflows must themselves be robust and deal with dynamics, such as changes in policies. It turns out that many of the concerns here are well addressed by control theory, and so we employ control theoretic principles to construct workflows with closed loop controls.
To better illustrate the problem we address, consider the following example drawn from the area of software distribution….
The remainder of this chapter is organized as follows. Section 2 describes how we apply process analysis to software distribution to characterize what automation is required and when it should be applied. Section 3 details how new processes can be generated automatically from dependency information. Section 4 discusses principles for process design based on control theory. Our conclusions are contained in Section 5.

2. Process Analysis

3. Automated Plan Construction

4. Design for Process Dynamics

5. Conclusions

Chapter: Some aspects of network analysis and graph theory

Mark — Fri, 08 Apr 2005 18:01:04 GMT

Summary:

"Some relevant aspects of network analysis and graph theory"

with some discussion of:

- basic concepts
- node degree distribution
- types of topologies and their properties
(emphasis on scale free)

- centrality on directed and undirected graphs
- concepts & math
- known applications (brief) (discuss social networks,
technological networks, and "mixed" ones such as P2P)

- EVC, topography, and spreading on undirected graphs
- roles & topography
- our new spreading stuff
- limited theory so far
- applications: security, efficient spreading of info

- directed graphs: why it's more complicated
- the math from our JACM-wannabe paper
- what we have that is new: TRank I & III
- relevance of these, besides for search
(must be speculative so far)

Chapter: Parallelism and cluster management

Frodes — Thu, 07 Apr 2005 04:43:06 GMT

Summary:

This chapter will cover the basics of parallel and high performance computing and management issues related to everything from the early parallel computers, more recent clusters, to the present grid.

Parallel computing
o why, who and how
o brief history
o factors governing performance
o core management issues in management of high performance resources (computational requirements, user and social factors)

Management of multiprocessors and supercomputers - historic perspective

o TBD

Management of clusters

o TBD

Management of the grid

o TBD

Chapter: Open Technology

Mark — Fri, 25 Mar 2005 21:25:25 GMT

Summary:

Open technology

Open standards

Open specifications

Open software

Open source

The scientific character of software development

Software as a scientific publication

The influence of software patents

The influence of licenses

Case studies in closed software and closed file formats

WordPerfect

Word versions

Laws on Open Technology

Chapter: Email

Mark — Fri, 25 Mar 2005 21:23:23 GMT

Summary:

Mail history

Including pre-Internet mail (UUCP)

Mail architecture

Mail hierarchies

Categories of mail agents

Store and forward (MX)

Mail protocol standards

(E)SMTP, POP, IMAP

Mail format standards

Meaning of headers

Mailbox formats

Mail reliability and scalability

Storage and mail backends

Mail flow control, auditing and accounting

Load balancing and failover

Mail abuse

SPAM and viruses

SPF and Sender Id

Content filtering

Policy filtering

Mail security

Transport security (SSL)

Content security (PGP, S-MIME)

Mail gateways

Mail to news and news to mail gateways

Gateways to non-Internet mail (Lotus Notes, Exchange Server)

Mail features

Aliases and forwarding

Mailing lists

Virtual domains

Mail usage

Conventions (quoting)

Etiquette

Mail server implementations

Sendmail

Postfix

Qmail

Relations with other technologies

Directory services (DNS, LDAP)

Authentication frameworks (Kerberos)

Instant messaging

Client side mail manipulation

User filtering

Chapter: Task and service provision

Mark — Tue, 22 Mar 2005 10:31:54 GMT

Summary:

Service and Task Provisioning

Gabi Dreo Rodosek
University of Federal Armed Forces Munich
Werner-Heisenberg-Weg 39
85579 Neubiberg
E-Mail: dreo@informatik.unibw-muenchen.de

Heinz-Gerd Hegering
Leibniz Supercomputing Center Munich
Barer Str. 21
803333 Munich
E-Mail: hegering@lrz.de

The chapter will deal with following issues:

Discuss the problem area of service and task provisioning
What is difficult and why
How to identify the adequate resources to provision services and tasks upon?
How to identify the necessary steps for the provisioning?
Examples from various area (configuration mgmt, job scheduling, grid computing etc.)
What is necessary to take into account during the specification of SLAs wrt. the dynamic service provisioning aspect? What is important and why?

Review of existing approaches (research, productive usage) with a discussion of drawbacks and benefits

How provisioning needs to interact with capacity planning (processes such as ITIL and tools), discuss capacity planning approaches (research, productive usage)

Chapter:Programs, Executables and Architectures

Mark — Tue, 22 Mar 2005 10:30:27 GMT

Summary:

TITLE (provisional) 'Programs, executables and architectures'
AUTHORS (provisional): Jan Bergstra, Inge Bethke and Alban Ponse,

CONTENT (provisional, some items may be left out if it becomes to long,
connections to other papers may be included as options for doing so
emerge):

- definitions of hardware and software, and a split up of software in
control code and content code (using machine function based control code
algebra),

- an analysis of when a control code is a program and in which case it is
not. A precise definition of what a program is, based on program algebra
and a distinction between programs (= instruction streams) and program
descriptions (often simply called programs). Projection semantics of
program description notations (programming languages) and why this
provides a pragmatic view towards many program notations irrespective of
whether one appreciates their design or not.

- a formalization of clients and services culminating in client/server
composition of threads and its key properties.

- a description of multi-threading and concurrency at a foundational level
using thread algebra. An example of the conept of a virus, an analyisis of
the classic result fo Fred Cohen on virus detection,

- a description of the halting problem as a statement about programs
operating in a formalized architecture.

Bruke Wiki i matematikk

Olavt — Tue, 15 Mar 2005 14:36:52 GMT

Summary:

<h3><center>Bruke Wiki i matematikk?</center></h3>

Klikk på 'Edit' (over) og skriv i skrivefeltet.

Det ser ut til at denne Wikien har støtte for LaTeX (se 'rot'-ikonet i verktøylinjen). Da skal man kunne skrive matematiske formler i denne Wikien.
Men URLen under viser til bruk av Tex for å skrive matematiske formler.

* Se http://meta.wikimedia.org/wiki/MediaWiki_User%27s_Guide:_Editing_mathematical_formulae:TeX_markup

Og TeX og LaTeX er vel ikke det samme?? - Bruker de ulike måter for visning ('rendering') av matematiske symboler?

For info om TeX, se http://en.wikipedia.org/wiki/TeX

For diskusjon om LaTeX, TeX og Texvc se disse URLene:

http://en.wikipedia.org/wiki/Texvc og http://meta.wikimedia.org/wiki/Texvc

Noen eksempler på matematiske symboler hentet fra http://meta.wikimedia.org/wiki/Help:Editing

* ∫ ∑ ∏ √ − ± ∞

* <i>x</i><sup>2</sup>  ≥  0 true

* <math>\sum_{n=0}^\infty \frac{x^n}{n!}</math>

Men det er kanskje best å gå direkte til informasjonen om hvordan man gjengir formler med TeX? - se:

http://meta.wikimedia.org/wiki/MediaWiki_User%27s_Guide:_Editing_mathematical_formulae:TeX_markup

Innspill

Olavt — Tue, 15 Mar 2005 14:30:11 GMT

Summary:

<h3><center>Innspill</center></h3>

Det har slått meg at vi kan gjøre jobben vår lettere og enklere ved å arbeide sammen.
Da vi reviderte fagplaner for noen uker siden, kunne vi ha delt et felles dokument og
skrevet i det. I en Wiki ville alt blitt bevart, og vi kunne sett tidligere forslag og endringer.
På samme måte kan vi lage felles uttalelser i en rekke saker.<br>
Men vi kan også bruke Wikien rent faglig overfor våre studenter. Hva med å integrere LaTex i Wikien slik at vi kan bruke den i fag som matematikk, for eksempel?

Hvis noen er interessert i dette, så håper jeg at de tar seg tid til å legge inn noen synspunkter
på
*[[Bruke Wiki i matematikk]].

Latex er klar -- men av en eller annen grunn må man logge inn for å se resultatet (dette må vi fikse): <math>X = \int_0^\infty f(t) dt</math>

<math>X= \sum_{i=0}^\infty A_i</math>

<math>\Pi_{\tau,\chi}(a\rightarrow b)\otimes \Pi_{\tau,\chi\cap\chi'}</math>

Denne siden kan redigeres av andre: de kan legge inn lenke til andre sider og emner for diskusjon.

Chapter: Application Integration with Semantic Web

Mark — Mon, 14 Mar 2005 18:30:22 GMT

Summary:

Application Integration with the Semantic Web

dr. J.L.M. Vrancken
Delft University of Technology
Faculty of Technology, Policy and Management
josv@tbm.tudelft.nl

dr. C.P.J. Koymans
University of Amsterdam
Faculty of Science
ckoymans@science.uva.nl

Introduction
It occurs very often that people want applications to cooperate that were not designed to be cooperate. Often the applications were designed in different periods of time for different goals by different project teams. This may happen within one organization but nowadays it also occurs between organizations.
Often it is not just two applications but whole sets of hundreds if not thousands of applications. Often there is already a long history of patches and bilateral interfaces resulting in a nice spaghetti of interconnected applications, which is increasingly expensive to maintain.

Integrating applications
Actually the multi-organizational integration problem is not so much different from the single organizational setting, at least technically speaking. Managerially and in decision taking, the multi-organizational setting may be different.
There are two main ideas in the approach to application integration that we describe here.
First of all, the bilateral interfaces should be replaced by one main interface. This will reduce the number of interfaces from O(n2) to O(n), where n denotes the number of applications involved. The benefits of this are especially appreciable when n increases.

The second idea is that such an interface not only caters for connectivity (applications being able to interchange bits) but also for interoperability, which means that applications 'understand' each other's data. For instance the Web Services paradigm essentially offers such an interface up to the level of XML: applications can exchange XML documents. Middleware solutions like CORBA offered an interface up to the object interface. Can one go beyond XML and the object interface. Of course there will always remain something to be done by humans, but the Semantic Web offers a technology that helps in defining semantic properties of data, such that the applications can use this automatically in their information exchange.

Semantic Web can help in application integration but also in the subject of organization-wide or inter-organizational data management, which is closely related to application integration. This chapter gives an introduction to Semantic Web technology (RDF, RDF-S, OWL, ontologies) and the ways it can be applied in application integration and data management. It gives an overview of available tools and examples of organizations that are currently applying this technology. In order to properly position the subjects treated, an architectural introduction will be included which describes the relationship of business processes and information on the one hand and IT-facilities (applications and IT-infrastructure) on the other hand.

This chapter is closely related to several other chapters in the handbook, most notably the chapters on data management and on managing Web Services.

Chapter: Policy based management

Mark — Mon, 14 Mar 2005 17:37:06 GMT

Summary:

Arosha Bandara, Emil Lupu and Morris Sloman Department of Computing, Imperial College London, 180 Queensgate, London SW7 2AZ {a.k.bandara, e.c.lupu, m.sloman}@imperial.ac.uk

'''Abstract:''' ''Policies are rules governing the choices in behaviour of a system. They are often used as a means of implementing flexible and adaptive systems for management of internet services, distributed systems, and security systems. There is also a need for a common specification of security policy for large-scale, multiorganisational systems where access control is implemented in a variety of heterogeneous components. In this chapter we survey both security and management policy specification approaches. We also cover the issues relating to detecting and resolving conflicts which can arise in the policies and approaches for refining high level goals and service level agreements into implementable policies. The chapter briefly outlines some of the research issues that have to be solved for large-scale adoption of policy-based systems''

Topics Covered:

1. Overview Introduction to policy based management, concepts and terminology: what is policy, the different types of policy etc; policy components and policy organisation. Present some example scenarios that can be used as motivating examples to show how policies would be used to provide security, trust and management functionality.

2. General Aspects and Architecture Describe the common features of policy based management systems and describe an architecture for specification, deployment and enforcement of policies.

3. Policies for Security and Trust specification Discuss the requirements of security and trust specification and show how policy based management can be applied in this context. Provide an overview of the various research results that provide security and trust functionality using policies. Use the examples presented in the overview to illustrate the capabilities and limitations of the various approaches.

4. Policies for Management and Control of Networks and Distributed Systems Use the examples presented in the Overview, to describe the requi
rements when using policies to manage networks and distributed systems. Present the existing approaches to meeting these requirements and use the examples to show how they can be used.

5. Ponder Policy Framework The Ponder framework provides a language for specifying both security and management policy in a declarative form. This section will describe this framework and provide examples of how it can be used in each application domain.

6. Tool Support for Policy Based Management Present a summary of the toolkits developed in academia and industry for policy-based management. Examples include, KaOS (IHMC, Florida), Ponder (Imperial), IBM Tivoli, Cisco Assure, HP Openview etc.

7. Policy Analysis and Refinement Present the problems associated with analysis and refinement of policies, and provide an overview of the work being done to address these challenges.

8. Summary and Conclusions

Norske samarbeidskontakter (no)

Mark — Tue, 08 Mar 2005 12:40:56 GMT

Summary:

Contacts in strategic companies and organizations

{| border="1" cellpadding="10"
|-
|NAVN HOS OSS || Firma/organisasjon || NAVN HOS DEM
|-
|Mark Burgess || IBM, Stormaskin divisjon, Norge || Anders Nohre, Per Fremstad
|-
|Mark Burgess || Telenor FoU, Fornebu, Norge || Geoff Canright, Kenth Engø Monsen
|-
|Mark Burgess || Økokrim/Nye Kripos || Torstein Schjerven
|-
|Mark Burgess || Politihøgskolen || Marit Gjerde
|}

Resources for thesis writing

Mark — Tue, 08 Mar 2005 12:24:33 GMT

Summary:

<h2>General structure</h2>

A report has a general structure:

<ul>
<li>Introduction
<li>Motivation and aims
<li>Literature survey of related work
<li>Methodology
<li>Results and their analysis
<li>Conclusions and suggestions for further work
</ul>
Your references to literature should not only be in an isolated "literature survey". In general, your report should have an introduction and conclusion that is easily understood by a non-expert, and a technical part that is detailed enough so that another expert could reproduce your results. Similarly your discussion of the literature can be at a high level in the introduction and conclusions, and at a more detailed technical level in the technical part.

<h2>Experimental methods</h2>

<h4>Control experiments</h4>

A control experiment means to take two identical experimental setups, under identical environmental conditions, change one of them and measure the effect of that change. One of the systems remains unchanged for comparison.

You should mention which environmental factors (factors that you do not control yourself) which can change, and draw them as a cause tree. See [http://www.iu.hio.no/SystemAdmin/analyticalbook.html Analytical Network and System Administration].

<ol>
<li>Comparing the same system at different times is not the same
<ul>
<li>Something might have changed in the environment since the last time it was measured.
</ul>

<li>Need to think about the effect of making multiple measurements on both systems at the same time. When? How?
</ol>

<h4>Checklists for discussing performance measurements</h4>

<ol>
<li>Remember that all measurements are uncertain. Draw a cause/fault tree that shows the different likely sources of uncertainty. Can you estimate how large each source is?
<li>What controls will you use to understand your results?
<li>Remember to `de-trend' your measurements, e.g. remove periodic variations in the average.
<li>Remember to repeat measurements and measure the distribution of values about the mean. Is it symmetrical, asymmetrical? Why?

<li>What is the flow diagram for activity in your system?
<li>Where are the sources and the sinks of activity? (Cause and effect)
<li>What are the overheads versus the processing times? (How does this change under stress testing?)
<li>When several sources come together, how will you demonstrate the effect of bottlenecks in performance?

<li>When there are random arrival events involved in a system:
<ul>
<li>What is the effect of short bursts of many events, or long periods of calm? (Burstiness)
<li> Plot the distribution of inter-arrival times. Is it Poisson, power law etc.? What does this tell you?
<li>If several random processes flow together in a confluence, what is the behaviour of the combined stream?
</ul>

</ol>