RE: I bet OTC is all over this article... :-P
> From: domusonline@gmail.com > Subject: Re: I bet OTC is all over this article... :-P > Date: Fri, 18 Sep 2009 23:54:59 +0100 > To: informix-list@iiug.org > Yes I stand by the comment that they will be within an order of magnitude. But that's not saying much. > ouch... you have a technical background... do you imagine that the > published results are near the reality? Well, that's the fun thing about benchmarks. Different competitors don't do head to head comparisons. Even on "comparable" boxes they'll trick it out differently. But the point is that if you had the exact same hardware and benchmarked the databases, how much different do you think you'll see? The point I'm trying to make is that if SQLServer can do 1000 TPM on a server, should I expect IDS to do a comparable amount? Or Oracle? I mean I guess and say 10-20% difference, but are you going to say that if SQLServer can do 1000 tpm, IDS can do 10,000 tpm? As much as I like IDS, I don't think so. Nor would any other one of the competitors. The key isn't which is the better database, but if I need to do 1000 TPM, how much hardware do I really need? > Using your sort of speach, I could say that I know of a customer who > picked some old (roughly the same age as their current machines) numbers > and calculated that if my current machines were able to do that, and the > new ones are able to do this (new benchmark numbers) than my new > machines sizing will be "this". > Let me rephrase what you said. If I know what can be done using IDS on machine X and the newer model has a benchmark of DB2 doing Y, I know that I should be able to see some number similar using IDS. I'm not saying that if DB2 does 25K tpm on the machine that IDS will do 25K, but it should be able to do at least 20K or maybe more than DB2 like 30K tpm. Or do you think I'm being kind and IDS coudl do 200K tpm on the same hardware? ;-) (We'll never know because we don't have any benchmarks to look at, now do we? :-P > They completley ignored the ammount of memory and disks the machines > had. They also ignored the tweaks that are made for benchmark purposes > that you cannot do in real systems etc. Hell, they even missed that the > CPU versions were not exactly the ones they were using (they were older, > and slower). > > My point (which no doubtly you'll spend time to destroy) is that this > may look good for marketing and market positions, but plase don't try to > sell us the idea that it makes sense technically... > The above is what happen when "cientists" try to transpose the benchmark > numbers into reality... > I think you missed the point. While OTC waxed on about using a newish mainframe to do some massive server consolidation, my point was that if you look at the horsepower you can get from the new generation of Intel chips, a 1U 2 socket 8 core machine can do enough to replace a rack if not more. I'm not against the fat pig, but I don't think that it would be price competitive when you consider what is happening in the Intel, AMD, Sun and even IBM's pc chip front. While OTC was talking about moving net new apps to a Linux mainframe, I looked at legacy apps on the mainframe that could be ported to Linux intel box and the savings would pay for the licenses and potentially the cost of the conversion. _________________________________________________________________ Your E-mail and More On-the-Go. Get Windows Live Hotmail Free. http://clk.atdmt.com/GBL/go/171222985/direct/01/