Blogroll

Syndicate

February 2006

Sun Mon Tue Wed Thu Fri Sat
      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        

New Blog

Subscribe w/Newsgator

  • Subscribe in NewsGator Online

Google Search


« Morning - 11/13/03 | Main | Noon - 11/14/03 »

Nov 14, 2003

Grids

I wrote a skeptical piece about grid computing recently, and in general I would reaffirm my contention that grid computing is not something you will see outside of marketing pieces for the near future. Long term it does have an important place, but it's going to take a significant effort by the application providers to embrace it, rearchitect, and some cases completely rethink how they deliver apps.

Yesterday I received a really interesting email from one of our research guys here at Labs that epitomizes how good things can happen when someone looks at a problem and innovates a solution using new technology.

Here's an excerpt:

"Over the last couple of months our group has been analyzing the R/3 source code in order to extract meaningful patterns that can be reused in future products, to work on eliminating duplication and redundancy, and to expose more value in existing software while reducing its overall cost of ownership.

With more then 70 million lines of application-only code and more than a million program units, R/3 is one of the most complex products currently available, and as we started writing tools to analyze this source code, it became clear that even with a $10 million computer it would take us at least 7-8 months to run initial statistical analysis of the data. A much more cost efficient approach would be to take advantage of the spare CPU cycles available on the SAP Labs campus; thus, we came up with the idea of building a grid of software agents here in Palo Alto."

I think what is really cool about this is how they have developed a very simple solution to solve a very complex problem. This is a really good example of using grid technology for computational analysis, much like what Seti@home is doing. Clearly, there are many applications within the enterprise that will benefit from this technology.

* EDIT: I forgot to mention that aside from thinking of a way to do this, this team actually developed the software and it now runs on ordinary desktop computers all over our campus in Palo Alto, which is well over 1,500 desktops. The software is a very elegant desktop process that has no interface, it just quietly goes about it's business of requesting work, performing analysis and returning the result, all the while doing this as a low priority process using only spare CPU cycles.

Comments

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment