ISC 07

Wish I was there. Have an injured foot, recovering slowly. Not sure what I did. Aside from that, costs to fly into Dresden were huge. Looked at Frankfurt, Berlin, Prague … Ugh. Maybe next year. Microsoft PR sent me some information pointers, I invited them to post here. Hopefully they will. I want adoption numbers. Looking over the PR and thinking it through, if they were having a massive adoption, I think that would be what they would talk about. If they were having a few successes in a limited domain, thats what they would talk about. And this appears to be the case.

Looks like the financial service folks, who are probably running .NET for the most part, have no trouble using formalized clusters rather than an array of desktops. This makes sense, and would be a slam dunk IMO. The flip side of this are on the canned application side. This list does not look significantly longer than the list at launch last year. Which suggests not outright neglect on the part of ISVs but a wait-and-see attitude. On this list you have standard commercial offerings, some open-source-turned-commercial, and some long time windows desktop folks. I don’t see any bright new exciting names there. ISVs want to know, if they spend their own money to build and ship a windows cluster enabled application, that they will at minimum, break even, and far more importantly, make a growing profit. Simple economics/business. What is the ROI. In some cases, and I would argue for the vast majority on the list, the “I” was already done. It was a recompile away from WCC certification. Now put a new app on. This is where the porting expense comes from. This is something, BTW, that we have tried to talk to Microsoft about in the past, when they asked us to propose running benchmarks for them. We pointed out that there was a little matter of a port and optimization to the target first. I am not quite sure they grasp how large this barrier is, and why this is preventing more applications from getting on there. If your applications are already running on windows desktops, running on clusters may or may not be worth your while, the “I” has largely been paid for, its simply a question of marginal additional “R”. If your application runs on Linux clusters, that “I” is going to be huge, like it or not. So that “R” had better be as well. Most ISVs recognize that it won’t be. Hence Microsoft’s problem. And this brings up another point that I have tried to raise. The business case. WCC is not better, cheaper, faster. It is more expensive, less well developed/vetted, and in a fair number of cases likely slower. When we went in front of VCs in the past we were asked is our accelerator product a “gotta have”, a “would like to have”, or “nice to have”, or “yet another one”. In retrospect, our answer of “gotta have” was correct, though we weren’t believed at the time. But lets ask the same question of WCC. Is it a “gotta have”? I don’t think so. If you do a SWOT analysis of it relative to its competitors, its only strength is that it is windows. Its weaknesses are legion, its opportunities are selling to its pre-converted masses, those running clusters on desktop systems. Its threats are a better faster cheaper competitor with more mindshare which is driving the market forward. Hmmm….. Next company stockholder meeting, would someone please ask Microsoft what it is they are doing? If you are at ISC, please ask them about this. I want to know the adoption rates. I want to know their level of commitment. I want to know how they are going to lower the barriers to application porting. None of this is trivial. Disclosure: My families IRA and likely my relatives 401k’s have Microsoft stock.