Monday Dec 01, 2008

Learnings from Lyon at ICT2008

I have spent this morning looking at the leaflets and notes I took at ICT 2008 in Lyon last week. I have bookmarked many of them at delicious with the "myict2008" tag. These cover mainly grids, distributed computing and knowledge management, there are a couple of consultancy sites as well.

<script type="text/javascript" src="http://feeds.delicious.com/v2/js/DaveLevy/myict2008?title=my%20bookmarks%20from%20ICT%202008&icon=s&count=20&bullet=%C2%BB&sort=date&tags&extended"></script>

I hope you'll find them useful. I have posted them here, using their link roll gadget since you can't enter del.icio.us on a date and this blog entry has both a date URL and a permalink. The lack of a date query is probably one of the reasons that people post links to their blogs. This is the first time I have done it, although the linkroll is in my sidebar on this page and on my archive page.

I hope to write up my notes in a more narrative form, which I'll back date to last week, which is when the conference took place.

tags:

Monday Nov 03, 2008

Building new age clouds

Sohrab Modi introduced three presentations from the Sun Labs on Hadoop & Hbase, and Project Celeste. He also pointed us at http://opensolaris.org/os/project/livehadoop/. I have downloaded this and shall let you know how it goes.

tags:

Tuesday Aug 26, 2008

Building big grids

A colleague of mine, Philipe Trautman presented on winning High Performance Computing deals. He produced some fascinating figures to describe the opportunity. Both Storage and Systems are forecast to grow at double digit rates for the next few years, where as commercial IT is expected to standstill at best. Over 30% of CPUs are going to be bought by HPC solutions during this period and at the moment, 65% of the HPC market is educational and/or research institutes. He outlined Sun's product portfolio consisting of systems, storage, operating systems and interconnects, which can be supplemented by partner products and people. He made the assertion that the real pain is no longer FLOPS, but elsewhere

  • Power & Cooling
  • Space
  • Cluster Management
  • Consolidation
  • Application Scalability & Utilization
  • Data Access including Filesystem selection

and presumably interconnect architecture and selection. Some of these are problems we have been confronting in commercial data centres for a while, albeit on a smaller scale but the last two are new.

Philipe introduced Dr Wolfgang Hafeman, of "Solutions for Research", a subsidiary of T-Systems and thus Deutsche Telekom, who have built and manage an HPC system for researchers in German commerce and academia, using Sun's products. I wonder if I can get the picture he showed, its quite dramatic. Again, this is an example of the right thing done well. Certainly T-System's people have added massive value to the proposition, although often the success of such a piece of business is based on the quality, drive and determination of the project teams. The relationship between the project teams supercedes the relationship between the companies. Its a great example of partnering for the end customer's success.

tags:

Monday Apr 21, 2008

Are "Quants" suitable for grid infrastructure?

Are "Quants" suitable for grid infrastructure? investorwords, a finance dictionary site, defines a quant as "One who performs quantitative analyses". Not so useful; while I have known people bet on horses because they like the name, I have not really heard of people investing in the stock market using this strategy, so evaluating the value of stock using numbers seems pretty basic, and in their eyes a Quant is a person. They define quantitative analyses as "The process of determining the value of a security by examining its numerical, measurable characteristics such as revenues, earnings, margins, and market share." If measuring the value of one stock, it is unlikely that, parallel algorithms are necessary. The data points are too few to warrant applying grid or parallel programming techniques. If analysing a portfolio of many stocks using these techniques, or a whole bank portfolio, then grids become more useful. It really depends upon the size of the portfolio, and the required response time.

A chartist on the other hand, performs "Technical Analyses", which investorwords, defines as "A method of evaluating securities by relying on the assumption that market data, such as charts of price, volume, and open interest, can help predict future (usually short-term) market trends. " Presumably the name chartist, comes from the fact that they use graph representations of the data series they analyse. However, the analysis of history increases the number of data points involved, and when one tries to apply technical analysis to portfolios, the problem of scale and the attributes amenable to parallelism come into the frame. The use of Grid software and hardware architectures probably becomes more useful earlier.

Neither of these techniques however are truly applications; they can be used to support trading decisions, risk evaluation, capital adequacy calculations, and pretty much any decision which involves evaluating today's and tomorrow's value of a financial instrument, be it treasury, equity or derivative.

investorwords is in my sidebar, and in my del.icio.us feed.

tags:

Wednesday Apr 02, 2008

workshoping the future

Over the last two day, we have been in workshops, discussing aspects of the development of the internet. The workshops, their agenda and supporting papers are all hosted at the future internet site. We'll have to wait for the slides to see what agreement was discovered.

I was interested to attend BO6, "Future Internet Research and Experimentation", otherwise known as FIRE where I heard a number of presentations from FP6 funded projects talking about the Grids they'd built, primarily on University sites. There's a lot going on. It's a shame we couldn't find someone to take on Sun's London "grid-for-rent". There was some innovative stuff in the re-provisioning solution.

The other working groups were called Networks, Services, Content and Security. I am eagerly waiting the slides from the plenary sessions that introduced and concluded these workshops.

tags:

Tuesday Dec 11, 2007

Driving change on the internet

The first key note was from Dr. Joao Schwarz Da Silva, a Director from the Commission's ICT. He envisioned a network of services driven by trends easily observable today. These are,

  • Social Networks
  • Digital Production
  • Virtual Worlds
  • Internet of Things

Much of the consideration around social networks seems around how to monetise the size of the network. The value created by cooperation seems always to be under valued. Dr. Da Silva predicted that the growth of social networks and user created content would lead to the growth of what he calls Digital Production. At its most simple, this will be just allowing mashups on a home page, however more complex models such as the tools for machinima or audio manipulation are clearly here today, it'll be interesting to see where this goes.

I am more questioning that virtual worlds will become ubiquitous and powerful problem solving tools. It is clear that World of Warcraft is a hugely popular both social network and digital world, but we have spent 1000's of years devising two dimensional representations of most of the problems we seek to solve. We need new representational metaphors before 3D rendering and virtual worlds become serious problem solving devices. I mentioned this earlier in the year. These criticisms are before considering that a Social Network needs to leverage the wisdom of crowds, or at least the wisdom of huddles. Facebook's visual {book/DVD} shelf works because you can see what both your close friends and strangers say about the books and films you're interested in. You can see what everyone, or at least your friends recommend. An interesting counterpoint though is that if you consider electronic gaming to be a social network, then sharding reduces the wisdom of crowds; you can only learn  from the wisdom of a shard. There's lots of work to do before 3D and/or Virtual Worlds truly take off.

He then looked at how in a network of services, one discovers anything useful. So this is partly how does one discover any content, such as images (tags), houses (attributes) etc., but for services we expect a directory solution. There isn't yet a directory of internet services.

tags:

Monday Dec 10, 2007

Hi Tech in Europe

And back to Brussels for a bit of EU politicking. Last time I traveled through Heathrow and the journey home was terrible. This time I travel through Southampton and the journey out is fine. I get to the Sun Office on the airport estate, do my meeting and have an easy ride into the hotel in the city centre.

I am attending the NESSI AGM. I wrote about it last time I visited Brussels in November, but it is having its AGM over the next two days. I will be commenting on it, but the slides and AV files will appear on its web site.

tags:

About

DaveLevy

Search

Archives
« April 2014
MonTueWedThuFriSatSun
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
    
       
Today