Slashdot Mirror

This is one of many projects related to GriPhyN (Grid Physics Network), an organized effort by physicists to bring important data analysis tasks to the home user. Distributed data analysis for LIGO is just one of the many projects that comprise GriPhyN; others include data analysis for the Sloan Digital Sky Survey and (I believe) the Large Hadron Collider, which is nearing completion at CERN in Geneva, Switzerland. SETI@HOME definitely caught the eye of physicists who, until recently, had been stymied by the lack of funds for supercomputers. While Linux clusters have gone a long way in addressing their needs, they quickly realized that the really data intensive applications such as LIGO, LHC, and SDSS would require something more. I'm excited that I might finally be able to change my screensaver to something other than SETI@Home!

I've seen a ton of questions asking what Grid computing is. The most common one being how does it differ from parallel/distributed computing?

First off, I highly suggest reading The anatomy of the Grid by Ian Foster et. al. It provides a pretty good overview into this whole Grid thing.

But for the lazy, here's a little bit. The Grid is more than parallel computing. Typically with parallel/distributed computing the problem or resources are static or both. Grid allows both of these to change. In a nutshell, Grid computing means not having to worry about where the compute resources are. Just start a calculation and it gets done. Just like how you don't worry where your power comes from, you just plug in.

The core of the Grid is virtual organizations. Under a VO, I could get together with a few friends and pool our resources. We could set up a registry and some factories (I'm speaking OGSA here, but whatever) and create some certificates. Then, we could submit jobs to the Grid and not have to worry about the resources that they're running on.

GSI provides some really nifty security features (based on X.509 I believe). Basically you provide a mapping that allows other authorized users to run commands on your computer. When you're on the Grid you create a proxy for your certificate that is passed to the process that you run on this other computer. Then if that computer needs more resources, it can create another proxy certificate and delegate to another server.

Also, Grid computing is more than just computing. There is data storage and instrumentation sharing also. You might want to check out PPDG, GriPhyN and TeraGrid for examples of these systems.

If you're interested in playing with the GRID, you can go download Globus Toolkit 3.0 Alpha or the Java CoG Kit which is a pure Java implementation of Globus 2.x (it's much easier to install than the regular Globus 2.2.x).

Hi there. Nice work on USITE Crerar, then. I've been in there a number of times since the lab was installed. Impressive Seti numbers, too.

Since you ask, I work for Dr. Foster under the rubrick of the Computation Institute on campus. Where on campus that is currently is in a state of flux between Ryerson and the RI.

I think that the USITE computers will continue running Seti for the time being, as we are not suffering from a lack of heterogenous resources, but a lack of software to pull it all together. We've currently got I think twelve grid sites lined up to accept Sloan cluster finding work, not counting the WorldGrid, which we should (hopefully) be able to tie into as well before long. iVDGL might be of interest to you on that front. My group, the Grid Physics Network, is less concerned with these management issues, than with taking the existing tools and wrapping them around real physics problems.

On the other hand, I do know people who wouldn't mind if those lab computers got set up with some Condor daemons, which would be a lot more useful than some single-purpose cluster finding code, anyway, and more in keeping with the grid computing style of doing things. I don't know how much demand there is for WinNT pools, though.

You can read about it at: www.griphyn.org

Buried on the web site is the original proposal they made, and it gives you some idea of the amount of data we're working with.

Some approximate statistics from the paper:

SDSS gets data at 8MB/s, 10TB/year.
LIGO will get data at 10MB/s, 250TB/year.
CMS will get data at 100MB/s, 5 Petabytes per year.

Work has already been done with simulated data for CMS, and a demo of virtual data (may be pre-calculated, or calculated on demand) for CMS was shown at the Supercomputing 2001 conference last week. They used Condor clusters from a few different sites. I'm not sure which sites made it into the final demo, but it may have included U. Florida, Argonne, and U. Wisconin.

Not true - CERN is neither responsible for GriPhyN (which is an NSF project), nor the invention of the data grid concept. The experiments at CERN (and other places) "merely" drive the need for something like GriPhyN.

That said, there is a European project similar in scope to GriPhyN, which CERN is a part of: the DataGrid Project.

For a book about Grids, you can look for "The Grid: Blueprint for a New Computing Infrastructure", edited by Ian Foster and Carl Kesselman. Both are on the GriPhyN project: Dr. Foster is a principle investigator on GriPhyN, and Dr. Kesselman is one of the Senior Personnel.

Damn. I know I'm gonna get (-1, Redundant) on my original article, but I stupidly hit the wrong button and Submitted it instead of Previewing it. Nevertheless, here's what I wanted to write after thinking it through:

You could also read the original project summary for the Grid Physics Network.

Although the site linked by the story (or click here for your One-Link (tm) GriPhyN info, in case you're too lazy to check the article out) has more new info that the original one, and it's more easily understood.
(Grrr. It's Preview, not Submit. It's Preview, not Submit).

You can read the project summary for the Grid Physics Network.

Here's what they're proposing to use the GriPhyN for: The four physics experiments are about to enter a new era of exploration of the fundamental forces of nature and the structure of the universe. The CMS and ATLAS experiments at the Large Hadron Collider will search for the origins of mass and probe matter at the smallest length scales; LIGO (Laser Interferometer Gravitational-wave Observatory) will detect the gravitational waves of pulsars, supernovae and in-spiraling binary stars; and SDSS (Sloan Digital Sky Survey) will carry out an automated sky survey enabling systematic studies of stars, galaxies, nebulae, and large-scale structure.

How about exploring the fundamental forces of nature and structure of the universe? That's what the GriPhyN project (Grid Physics Network) will be supporting. Some other related grid projects and forums can be found on the Links page.

Another good description of GriPhyN, maintained by one of the principle investigators, is here.

How about exploring the fundamental forces of nature and structure of the universe? That's what the GriPhyN project (Grid Physics Network) will be supporting. Some other related grid projects and forums can be found on the Links page.

Another good description of GriPhyN, maintained by one of the principle investigators, is here.