Wednesday November 26, 2008

Managing Torrow's Cloud

An off agenda session on Cloud Computing, kicked off by William Fellows of the 451 Group. I quite like his stacks both of functionality, illustrating what needs to be done and the evolution of the cloud from its partly failed predecessors. The discussion then moved to management, with contributions from IRMOS and the Autonomic Internet project, which sounds a bit IBM'ish but isn't. There's obviously some thinking going on about Service Management for Clouds and networks, looking at life cycle issues (is this just job management, probably not because of birth and death), self functioning, SLAs and QoS issues. It seems to me that Robert Holt's experimentation with SMF is exactly the right thing to do. The features that Sun's Systems Management Facilty add to the operating system are a foundation on which a number of features can be built which meet the need of Cloud managers. The BREIN project which says about itself,

"BREIN takes the e-business concept developed in recent Grid research projects, namely the concept of so-called "dynamic virtual organisations" towards a more business-centric model, by enhancing the system with methods from artificial intelligence, intelligent systems, semantic web etc."

I love the etc. It always makes you think people know exactly what they're doing. They have published a white paper here.... Despite this, these projects and this approach might well enable the automated SLA negociation. Can we create a semweb for SLAs? It always been the fact that sustaining and management science comes after the invention stage, but this was a jolly interesting session, and addressing issues identified by both myself and colleagues at Sun and leading industry commentators as crucial. If we don't/can't automate this stuff, we are going to run out of people.

tags: technology research europe r+d ict2008 SLA cloudcomputing future ict2008

Impressions of the Citie International

Among the things to do better should Sun come to ICT again, is that the hotel should be booked in advance. Its a real pain being so far from my hotel room; I can't return to my room for either power or privacy. The commute is a time consuming pain; I am staying in Vienne which is about 30 minutes away, although the journey takes much longer. The journey in both directions was made harder by a strike on the trains, it was just like old time in England being picketed by the CGT. I hope it's easier today.

The Lyon convention centre is enormous and very good. If we could justify a Sun global training event in Europe, it'd be excellent, I wonder if they rent parts of it?

Even at this conference, they 'ushhered' people to sit below the main walkways when using the main auditorium for break out sessions. Having said it'd be excellent, are there enough hotel rooms in Lyon, as I said I booked late and AMEX couldn't get me in (to Lyon), but it could always be AMEX's fault. The number of breakout rooms might be a constraint and the wireless was poor in a number of rooms and unlike Palau de Congressos de Catlunya in Barcelona, there is no power available in the conference rooms and halls. They claimed 4500 delegates at ICT 2008.

tags: travel ict2008 lyon

Tuesday November 25, 2008

Can Europe keep up?

I then attended a panel discussion on R&D in Europe, which given the attendees was pretty self congratulatory. HP's VP for Labs is a Brit, and was on the panel. The reason I mention this is that he was the only employee of a global IT company i.e. one not quoted in Europe, who spoke in a plenary session. They sort of said "Great Research, no IT manufacturing" , but why? We do have ICT manufacturers in Telco, including Alcatel, Ericsson, Nokia and Seimens.

Can the European NEP's maintain their leadership? What does Europe's computing hardware poverty mean? Can it compensate with a single market, a vibrant software industry and a well educated work force?

It was also shown that not all these advantages are enough. SAP does very little development in Europe these days, and it was said that innovation rate in Europe is too low, despite a world leading position in many areas.

tags: economics research europe r+d ict2008

Visions of Future Computing

After lunch, with wine, it is in France after all, I attended a session called "Visions of Future Computing and Communication Paradigms". Frustratingly this was not video'd and nor can I find the slides on the USB stick they gave us. So you'll have to rely on my memory; I didn't take any notes. The first two speakers, although their presentations weren't designed to show the difference between IT people and computer scientists. Prof. John McCaskill, of BioMIP, the Biomolecular Information Processing Research Group presented on 'Constructive IT', which as far as I can tell starts from chemistry and is looking at new ways of building computers...beyond Silicon. I have to ask what sort of timescales they expect to do anything substantial. The need to change programming models because of large scale multi-threading is one thing, the abolition of silicon is quite another. This stuff just amazed me. He was followed by Micheal Wolf, who illustrated the insights that quantum physics offer to mainly software design. He was followed by Illka Tuomi also at Wikipedia, who presented on "Intellectual Property Processing After the End of Semiconductor Scaling", and his slides are available on his personal web page at meaningprocessing.com. He illustrates some interesting changes in system design after the end of Semiconductor scaling. The session was brought to end by Wendy Hall, who illustrated the holistic nature of ICT futures arguing for a 'Web Science' approach borrowing from many separate disciplines to build an understanding of the technical and social networks that are being built today.

tags: europe future ict2008 technology eu

ICT 2008, Lyon

I got into the conference in time to hear the words of welcome from the Mayor of Lyon, and the opening panel discussion. The panel was chaired by Viviane Reding, European Commissioner for Information Society and Media, and its participants were Luc Chatel, Secrétaire d'Etat chargé de l'Industrie et de la Consommation, France, Esko Aho, Executive Vice President, Nokia Corporation, and Former President of the Finnish Innovation Fund (SITRA), former Prime Minister of Finland and one of the key commentators on FP6, he chaired the group that produced "Information Society Research and Innovation: Delivering results with sustained impact", which was published in September. Also on the panel were Ben Verwaayen, CEO, Alcatel-Lucent, previously of BT, Harold Goddijn, CEO, TomTom and Michel Cosnard, CEO and Chairman, INRIA, representing a research view. The conference has a video link on its site for this session. The panel was called "Setting the ICT Agenda for the Next Decade" , has its own page. The panelists said little of controversy, with Verwaayen arguing that trust and security were keys with Aho arguing for a global dimension, starting from a green perspective to invest in productive knowledge. He also interestingly argued that US leadership was based on entrepreneurialism and commercial innovation. I was surprised, I am not yet convinced that european basic science research is yet competitive with the US. For instance, while researching NESSI's contribution to the EU's Software Industrial policy, I was pointed at China's Shanghai Jiao Tong University's study of Academic Ranking of World Universities. I, and others, have considered the methodology and anomalies, but it illustrates a world domination of scientific excellence in the universities by the USA. However Goddijn, who was there to tell the startup story, stated that his biggest problems in building Tom Tom were not technological, but regulatory compliance, specifically, VAT and patent registration. These comments got a round of applause, and Verwaayen weighed in specifically asking when it might become possible to register patents in the EU in one language. There were further discussions on the public policy dimensions of how innovation enters the economy, discussing public/private partnerships, educational/innovation clusters with much agreement about the short term changes in ICT.

In between the opening sessions and the panel discussion, some video's from Futuris were shown. This focused on the use of ICT in health care delivery. I have argued previously that the UK's investment in i-health care has been too focused on record keepting and NHS cost control, so it was good to see a couple of case studies showing the innovative use technology in improving the ill and injured's lives. I can't find the specific video on the Futuris site, but Futuris is an EU sponsored TV show broadcast on the Euronews channel. Leave me a comment if you find it.

tags: europe future ict2008 technology eu

Monday November 24, 2008

Vienne

I am in staying in Vienne to attend the EU's ICT 2008 conference, to be held in Lyon which is a biennial conference of Europe's top IT researchers in commerce and academia, convened by the Commission of the European Union.

tags: travel

Wednesday November 19, 2008

Mobile Viewing

I have just attempted to read a recent Guardian article, the Coolest Quartier in Paris. I was pointed there by my ipod touch's feed reader, Daisy Feed, and decided that the page is not well read even on the 'touch's screen, so checked it out on a laptop browser. I have been surveyed by the site owner and they asked some questions on the mobile internet, which since they didn't ask me all the questions I wanted to answer, I'll comment here.

I have struggled to embrace the internet on the phone. The screen has always been a problem and as I get used to the 'touch, which is so much better I am considering my static web site and how I present pages on the net. Much content is arranged optimally for reading on a computer hosted browser, and this is true for much of the Guardian's site as well as my own.

An example of changes I am considering include a long standing page which hosts my delicious tag cloud and a revised version optimised for the ipod touch. You can view the differences, just, by hovering or click though to view. The tag cloud is no longer just a vanity, its a quick way through to my bookmarks.

I shall be reviewing web site name structure and looking at if I can use CSS/Javascript to represent the same pages differently depending on the device. So probably best not bookmark the ipod links page, I am not sure how long it'll be there.

The mobile internet'd be a lot more useful if wireless was more ubiquitous, but I have plans to fix this. I'll use my phone, to connect the 'touch to the internet.

tags: technology guardianunlimited internet mobile

The EU's Call 4 for research projects funded by FP7

19th November - The Commission of the European Union have advertised FP7 Call 4 [Press Release]. This is the opportunity to undertake collaborative research into ICT with financial contrinbutions from the Commission. The press release talks of seven challanges, "Pervasive and trustworthy network and service infrastructures", "Cognitive systems, interaction, robotics", "Components, systems, engineering", the comma's are theirs, I want to check up on this, "Towards sustainable and personalised healthcare", "ICT for mobility, environmental stability and energy efficiency" and "ICT for independent living , inclusion and governance". The call also looks to promote research in three new areas of Future and Emerging Technologies, one of which is "Concurrent tera-device computing", you'd think we might be interested in that.

An interesting set of priorities, the Call for Proposal is on Cordis, the EU's Community Research & Development Information site.

Thursday November 13, 2008

The SuperNAP

I was invited to visit Switch Communication's Supernap facility. This must be the best datacentere in the world. It is purpose built, and designed to host new age high density computing. They set out to build a 35Kw/Rack data centre and every decision they took was to enable this goal. There is no compromise. For instance they have invented their air conditioning plant, since, so they claim, the industry leader wasn't interested in innovating for them, they pump cold air into the top, of the room, and suck the rising hot air out, leveraging the laws of physics. They have three power distribution systems, a fixed floor which supports the PDU system. If you require high density computing, these people are the people to go to.

The have a video on their home site that explains some more, and this video on Youtube, which explains even more.

I found this by querying youtube using the tag supernap.

They talk exclusively about cooling, power, and availability, and they summarise the offering this page. We i.e. Sun are quite good at fitting out data centres, but have rented space in their centre? If you can put up with US jurisdiction, its fantastic. It begs the question why anyone that's not a specialist would build another. Look at the videos.

This was posted in Jan 2008 and backdated to the time of occurrence.

tags: technology datacenter datacentre supernap video

Thursday November 06, 2008

What will the Cloud do?

I was pointed at the Eucalyptus project, an open-source software infrastructure for implementing "cloud computing" on clusters, by a colleague and decided I needed to check out Amazon first. Several colleagues have given me this advice but have the University really written an open source grid platform conforming to Amazon's EC2 APIs.

If so its a fascinating example of the speed of commoditisation. It raises the question of where's the value in building clouds? If you can't innovate above the system components where can you innovate? Its obviously pointless to copy what Google did 10 years ago and if the assembly is available in Open Source you should probably use it. The space left by Amazon for a competitive threat is that they major on Infrastructure as a Service, although of ocurse given the operating systems available you can quickly turn it into a platform. I have just checked Amazon's EC2 Page, and they now offer a database query interface to their storage solution. The space left is to offer higher levels of abstraction, specifically by offering Java, Python or Ruby space to customers, and this is what Sun's Project Caroline does. Sun also innovates at the system, silicon and software layers. IT Systems are not really commodities and sedimentation means they will continue to change, the industry still needs innovators. IT isn't done yet.

tags: technology cloudcomputing saas opensource

Billing for Clouds

When considering the some of the issues related to building private clouds, the "Usage to Billing" problem was raised and I was reminded of Emlyn Pagden's Blue Print the Utility Model - Part II. I had been consulting with a mid sized European Investment Bank, and discussed the architectural problem with them, and Emlyn. Its a while since I have read Emlyn's paper, but he took the architectural decomposition

• Measurement, what are people using
• Aggregation/Mediation - accross the whole estate
• Allocation - how many charges have they incurred
• Invoicing - give us our money

and built a reference implementation using Solaris Resource Manager and accounting functionality and some third party products. At the time, he was working for a team that wanted to sell third party software, he had no engineering resources and thus a propensity to use 3rd Party software before building significant scripted functionality. With different resources and motivations, the reference implementation might look quite different, but the paper which was based in a real prototype exposes a working solution.I suspect that not all the companies he mention either still exist, or remain in the "Systems Management" business. However the decomposition should allow easy replacement and the advances in SOA may make this easier to do.

One of the key problems that inhibit adoption of these solutions is that end-user IT departments are cost centres and financially aim to spend or underspend their budgets. Their outgoing charging tariffs are based on cost recovery and they don't care how busy what they supply is; they have to charge for what they supply. If they don't do this they make a loss, and the CIO gets fired.

Neither he, nor I experimented with testing this on a grid, and it might involve having a global /etc/projects name space across the whole cloud, but with Virtual Box testing these things becomes easier. Sadly I have picked up enough projects from this trip already, but now I need to build a grid on a laptop.

We both agreed that the invoicing function was best left to the ERP system. Private cloud builders may not need to produce an invoice since they may not be using real money, they will have to make some entries into the Financials systems either cost relief transfers or something. Also new start ups of public clouds may wish to look at Open Bravo, an open source ERP package.

tags: technology billing cloudcomputing measurement openbravo Sun Blueprintutility computing

Talking about Cloud Computing

The current technical state of systems, storage and networking and specifically the cost of broad band networking has created a tipping point. Over the last 10 years, organisations and people have been learning to build new distributed computing server complexes. It may be too late to copy the leaders, but certain design criteria and the regulatory constraints may mean that there is a slower commercial adoption cylce. The privacy, availability and response time requirements are for businesses are all different. In my mind, its commercial adoption that turns grids into clouds.

"One class of grid is where we locate one application, which has many identical parts on a distributed computing platform and we call this HPC; where we locate many copies of one application be it apache, glassfish or MySQL on a distributed computing platform we call it web 2.0 and when we locate many applications on a distributed computing platform we call it Cloud Computing"

Dave Levy

Its commerce that has the need for the Cloud, because they have usually have a large portfolio of applications, some of which behave like HPC and some of which behave like Web 2.0 and its the economics of utility that drives this. Sun's ERP solutions have leveraged our product portfolio and Moore's law to become a tiny fraction of Sun's IT estate, with the community infrastructure and the design support solutions being implemented on web 2.0 and HPC grids, now dominating Sun's internal network in terms of cycles, storage and cost.

Admittedly, there are other aspects of what makes a cloud different from the payroll bureau of thirty years ago.

Data Centres are expensive and as we are discovering in the last few years, they are best built for purpose. Building and running Data Centres also benefits from 'specialisation'. In "The Big Switch", Nicholas Carr argues that the efficiencies of the plant apply to IT. (I'm really going to have to read it). Historically, applications developers have tightly coupled their code with an operating system image, specifying the version, library installs, package cluster and patch state. This is beginning to end. Developers want to and do develop to new contracts, be it Java, Python or another run time. Also with virtualisation technology such as Virtual Box and VMware, deployers can build their utility plant and take an application appliance with an integrated OS and applications run time, this allows developers to choose whether to use modern dynamic runtimes or to tightly integrate their code with the environment.

A second driver is the amount of data coming on-line. This cornucopia of data is enabling/creating new applications, of which internet search is an obvious one. Google scans the web, but many companies and increasing social networks are scanning their storage to discover new valuable pieces of information. Internet scale also means the "clever people work elsewhere" rule of life is generating new questions. The growing number of devices attached to the internet is also discovering and delivering new digital facts. The evolution of the internet of things will make the growth in data explosive so its a good time to be introducing a new disruptive storage capability and economics. The need to analyse this massive new data source is what's driving the emergence of Hadoop and Map/Reduce. Only parallel computing is capable of getting information out of the data in any reasonable time. A fascinating proof point is documented on the NYT Blog, where Derek Gottfrid shows how he used Amazon's cloud offerings to convert the NYT's 4TB archive into .pdf using Hadoop. I'd hate to think how long it might have taken using traditional techniques.

One tendency I have observed from my work over the last year is that today building grids is now longer hard, and most dramatically Amazon and Google are turning their grids to applications hosting. A number of public sector research institutions have also been building publicly available grids for a wile, although they tend to share amongst themselves. In the public sector world at least, they have begun to address the question of grid interoperability, and everyone is looking at how to 'slice' resource chunks out of the grid for users, on demand of course.

In the commercial world the competitive positioning of various players has led to them competing with different services and different levels of abstraction. The offerings of Google's "google apps engine" vs "Amazon's EC2" are quite different. Sun believes that cloud computing offerings need to organise above the OS level now and that developers don't want to worry about the operating system, merely their run time execution environment. This is only possible because modern development and runtime environments can protect developers from both the cpu architecture and now the operating system implementation. I know that as I search for a new solution for the services I run on my Qube, I'm happy to configure the applications and their backups, but I don't want to worry about disk reliability and other system services.

Jim Baty made the comment that we're entering a Web 3.0 world which is chmod 777 for everyone.

So the economics are compelling, the state of technology is right, developers are ready to leave these decisions behind and the first movers are moving.

Can and will Sun play a role in this next stage of the maturing of IT?

This article is I hope the first of two, written from notes made during a presentation by Jim Baty, Chief Architect, Sun Global Sales and Services, Scott Matton, one of the senior architects in GSS and Lew Tucker, VP & CTO of Network.com. The article is back dated to about the time of occurrence.

technology cloudcomputing economics sunw datacentre datacenter

Wednesday November 05, 2008

Tuesday on the night of Obama's election

Last night was very quiet, I went to Kapps in Mountain View. I had left the office where a number of people were, oddly, watching the BBC web site report via a wall screen display. I had also enabled the facebook and multi-protocol chat applications on the ipod and discussed the elections with my son at home in the UK. This was pretty good as I had to stop using my laptop. The media seem to declare states as won very early, with the BBC and Guardian being earlier and more certian than the US sites. Also the polls also shut early in my british view, the west coast polling stations shut at 8:00 p.m.

McCain conceded at 20:15, I wasn't expecting it so early; in British elections, the polls don't close 'till 22:00 and the counts don't start untill the counting stations are open, which is usually at 23:00. The first results come through at just after midnight and the results aren't ususally clear 'till the early hours. I rember going to bed at 4:00 am on 1st May 1997, knowing it was good, and that Mellor and Portillo were looking for work, but the final results weren't known 'till later in the morning.

The CNN site was awesome and I used it to follow the results during the earlly evening, and I have followed the elections via realpolitics.com, the BBC page had a great feature to show the states in proportion to their electoral college votes. It looks something like this, which is much more accurate and powerfully descriptive than a geographic map.

In 1997, I had a page torn from the Guardian and ticked off results on the paper's 'Must Win' list. A pencil and paper, but nomadic solution, to go with the UK's pencil and paper voting system.

tags: politics us2008 elections "nomadic computing" travel

Monday November 03, 2008

News on the move

"Free RSS" seems to have some problems, however, I chose to load itunes on a desktop at home so will have to wait 'till I get back there to remove it, or replace it. However, Google pointed me at a thread called "the best free rss application?" at http://www.ipodtouchfans.com. Meanwhile, I am still reading my feeds at Google Reader, both on the 'touch and on my laptop.

tags: technology ipodtouch silly apps rss feedreader

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: technology future cloudcomputing hadoop filesystem grid