Ian Foster

Off to Africa

I leave today for South Africa, where I will participate in the 4th IFIP School in Software Technology. A week's vacation in the bush west of Johannesburg, then a week at Gordon's Bay near Capetown. It should be a fascinating trip.

Sun Blackbox video

A nice video (1 minute into Greg Papadopolous' talk) on Sun's "cluster in a shipping container" product, BlackBox, which I wrote about a while back.

Grid data mining tool based on GT4

The European Data Mining Grid project announced this week the release of their grid-enabled data mining software. This Apache-licensed, Globus-based software "is a general purpose  software facilitating user-friendly, Grid-based data mining."

It sounds pretty interesting. I hope someone will try it and let me know how it goes.

Webinar on "supercharging your cluster with Globus"

My colleagues Carl Kesselman and Steve Tuecke are presenting a Web-based seminar (aka Webinar) on March 20 on "supercharging your cluster with Univa Globus." Univa Globus is the commercially supported version of Globus--the same  open source software, but with someone to call if things go wrong.

The abstract sounds a little dull, but Carl and Steve are both excellent speakers and have interesting things to say, so this should be worth listening to:

Warren Washington on climate change

I attended a talk by the distinguished climate modeler Warren Washington on Thursday: "Climate Modeling of the 20th and 21st Centuries." He spoke on the state of the art in climate modeling, the evidence for warming, and the likely impacts of future warming. The scientific consensus is that we the planet has warmed 0.7C since the beginning of the 20th Century. January 2007 was the warmest January in recorded history. It's easy to see why people are so worried.

The talk also featured a long Q&A. One Q: how well can models predict catastrophic change? A: Not very well, as relevant physics (e.g., Antarctic ice sheet collapse, methane hydrate emissions) are not well understood. From a climate change skeptic: might observed warming not be due to some other factor? (As: Physics of greenhouse greenhouse forcing well understood, no other mechanism known, rate of change unprecedented in record.)

Warren Washington co-authored An Introduction to Three-Dimensional Climate Modeling, from which I learned much of what I know of geophysical dynamics. He was also chair of the National Science Board for a while. A great scientist.

Dev.Globus Announces First Escalation and Welcomes Six New Incubator Projects

Some good news from dev.globus--a first incubator project, the GridWay Metascheduler project, has completed incubation and is now a full Globus project. In addition, six new projects have joined the dev.globus incubation process, bringing the total number of projects to twenty-two.

A more detailed description from an announcement prepared by Jennifer Schopf, dev.globus Incubator Management Project (IMP) chair:

Globus at sea

Italian Globus enthusiast Raffaele Montella sent me this picture of the sailboat Sarima V  that he races out of Naples. Apparently he uses grid to good effect in his racing, using an online Globus-based forecasting service to obtain up-to-date weather forecasts prior to each race. Being a sailor myself, I can only applaud (and feel jealous).

He has a paper on this work in the upcoming Grid and Pervasive Computing conference, to be held in Paris May 2-4. Title: "Development of a GT4-based Resource Broker Service: an application to on-demand weather and marine forecasting." I'll post a pointer to the paper once it is online.

Swift takes wing ...

[Update: see also a later post.]

We unveiled this week the first release of Swift, a system for the specification, execution, and management of applications comprising many tasks coupled by disk-resident datasets. Such applications are common when analyzing large quantities of data, performing parameter studies, and/or executing ensemble simulations. (The word "workflow" is often used for such applications, but it doesn't sound right to me) The open source Swift software combines:

• A simple scripting language, SwiftScript, for the concise, high-level specification of such computations (without regard to data layout or location), and
• An execution engine for the rapid and reliable dispatch of many tasks to many processors, whether on parallel computers, campus grids, or multi-site grids.

Swift users in the physical, biological, and social sciences; the humanities; computer science; and education have achieved multiple-order-of-magnitude savings (!) in program development and execution time, relative to approaches based on shell scripts and other ad hoc technologies.

Swift builds on work performed with National Science Foundation's Grid Physics Network (GriPhyN) project on the Virtual Data System (VDS). Work on another VDS component, Pegasus, continues at USC/ISI.

The Swift team comprises Yong Zhao (imminent PhD, already interviewing), Mike Wilde, Mihael Hatigan, Tibi Stef-Praun, Ben Clifford, and Nika Nefedova. Gregor von Laszewski architected (and Mihael Hatigan built) the Karajan runtime system on which Swift is built. Globus services are used to access remote computers and to move data.

DDM Contributed to dev.globus

The Univa guys announced that they have contributed their Data Distribution Manager software to dev.globus as a new open source incubator project.

The following extract from email posted by Steve Tuecke to the DDM/dev.globus mail list summarizes some key features:

Seeking GRAM4 Success Stories

I wrote recently about the nice work that has been done on GRAM4 (aka WS GRAM) the new incarnation of the Globus "Grid Resource Allocation and Management" service.

Stuart Martin posted the following message yesterday, asking for input on what people are doing with GRAM4: