Skip to main content

High Throughput Computing week

We;ve just finished a week (well, four days) of talks, tutorials and discussion about High Throughput Computing. The event was opened by Miron Livny, leader of the Condor team, who gave an excellent introduction - the key point is that HTC is about the number of tasks that can be completed in a given time, whereas "traditional" High Performance Computing is about how much computing power can be brought to bear at a given time. As Miron puts it, Floating Operations per Year is not necessarily 60*60*24*7*52 Floating Operations per Second (FLOPS).

We've hosted events by the Condor team in the past, but for HTC week we extended our range. In particular, John Powers and Dan Ciruli of Digipede flew over from the Bay Area to tell us about their product. A day of hands-on tutorials allowed delegates to compare the strengths of Digipede and Condor, and the evening discussions included ways the systems could be used together.

Scheduled discussions looked at requirements for HTC in academia and in business, at Green IT, data handling and policy management. As one outcome of the discussions, we're looking to capture HTC design patterns, publish them on the web and incorporate them into training materials. On the academic side, we are planning to write a report to explain the policy issues to university heads of research computing.

There is talk of running this event again next year. We would like to extend the range of participants again, e.g. by looking to UnivaUD, DataSynapse or Platform. For more vendors to attend, we will need more commercial users, and vice versa, so we need to start encouraging people now. I'm wondering whether a small exhibition area might be useful to the commercial vendors and delegates - not a major trade show (with the concomitant expense) but enough to be an effective market place.

Watch this space - and get in touch if you're interested!

Comments

Popular posts from this blog

Project Leadership

When I look back over recent projects that I've been involved in, it seems that one key to making a project successful is having someone on the team who really drives it forward: someone who is invested in the project as a whole and not just their own part in it. We (by which I mean the University's Applications division) have a well-defined project process, with defined roles, required milestones, deployment standards, and so forth.    All these are useful, but if the team doesn't have a leader, it seems a project can lose its way, perhaps not responding to changing circumstances, getting stuck on a technical problem, or not securing a needed resource in time to meet some external constraint. The leader can be any member of the team - it could be a developer, a project manager, or the sponsor, or someone in another role.  A team can include several people who are this committed to the project; it doesn't have to be a single person.   As an example, one of...

Business Model Canvas

A Business Model Canvas is a tool for mapping the core functions and capabilities of an organisation.  Compared to the Core Diagrams that I described in an earlier post , the business model canvas attempts to present more aspects of the business, starting with the value proposition – a statement of what the organisation offers to its users (in the business world, to its customers).  It shows the activities and resources, as Core Diagrams do, but also shows user relationships & channels, and also benefits and costs.  I’m not aware of any universities that have used this tool but you can find examples from elsewhere on the web. We are considering business model canvases as a tool for mapping the strategic capabilities of units at the University of Edinburgh.  Phil Taylor, our EA contractor, sketched an outline of what a business model canvas might begin to look like for HR: This is only intended to be suggestive: the real canvas would need to result from in-de...

"That's not Grid!" - A cautionary tale

I don't usually attempt humour here, but see if you like this... Once upon a time, in the far-off world of Computerland, a great guru arose and declared a vision. "I see a future when all computers will be linked together and people will run their programs without knowing which computers are running them. People do not need to know where their jobs run; they just need the results. I call this Grid Computing!" The people of Computerland were excited by the guru's vision. They went away and worked to make it happen. When they were ready, they returned to the guru and said: "Oh great guru! We have implemented your vision. We can run our programs on whichever processor is free at the time, making sure that all programs can run and making best use of all our processors. No longer does one computer sit idle while another one is overloaded. Your vision is a great success!" "No, no, no!", said the guru, "That's not Grid!" The guru exp...