Ran across this article last night; it’s Tim Bray’s summary of the computational grid landscape as he found it in May of 2006. From the article:
…I needed something like memcached or Prevayler or Tangosol Coherence; a really fast distributed hash table storing data in RAM across a lot of machines. …
So I decided I needed to run it on one of these new-fangled “Grid” thingies and went looking for infrastructure. This article details what I found. I don’t claim it’s complete, just that it represents the results of research by a motivated non-specialist.
I’m sure many of you will take issue with the details of Tim’s findings. But it is a useful summary of a part of the Grid landscape, and if you haven’t spent much time paying attention to the nits of what goes on in this detail-obsessed community, you may find this article helpful.
He concludes with some great observations of things that have been bothering me, too, and which I plan to study further:
My feeling is that if something says it’s a service, the right way to talk to it is to get a hostname/port-number combination, call it up, send a request, and wait for a response. …My feeling is that if I’m going to be stressing out the performance of an app on a grid, I don’t want too many layers of abstraction between me and the messages.