HP Publishs First Linux TPC-C Benchmarks
The first ever official TPC-C benchmark on a Linux system has been published.
This was run on a cluster of 32 HP servers with Intel Xeon CPUs, running Redhat Linux and Oracle RDBMS. The system had over 18 terabytes of storage, and cost over 2 million US dollars. Performance was higher than a similar system running on MS Windows.
Put few of those together and we can at least make some decent 64 player BF1942 servers.
To show the world that throwing hardware at a problem like Oracle contributes to the solution FAR more that the software on the box.
.NET will show up all over, not a bad thing, but this kind of technology isn't the "get you to the moon and back" kind, and we know this. I would like to see the rest of the supported Oracle platforms/OS compete, Sun/SPARC/Solaris, Itanium/Linux, PA-Superdome/HPUX, UniTrash[UniSys lol]/Linux, NEC, etc.
.NET fares against Java.
..
One would think that Oracle has been written in such a way that makes what OS it runs on moot.
Based on this, since I have become used to seeing Linux compete with and exceed Windows performance at even Window's own core technologies, e.g. SMB/CIFS, "length-measuring" (you know what I'm getting at here) with Oracle might not be the PR this kind of story is meant to attract.
Also, run this benchmark for a year, sustained. Uptime at the end of a year would be far more interesting that this peak crap - 100-150 CIOs at Fortune 500 companies might care about that, but they are probably more interested in *nix being able to stay up on the order of years and not having to reboot for anything but a kernel patch [with kernel modules even being hot-fixable].
Windows is not interesting to me. Its not penetrating the back end, even though Microsoft marketing pretends it can. Its been thrown around as a workgroup server, but cant be a serious directory or DNS server. Its polluted the world with ASP. Even
Also, Microsoft pays no dividends (MSFT), and has a monopoly (most PCs shit with some Windows OS). They rely on "doing better next time" to make the share price go up (it hasn't done very well since they completed the monopoly). They have to penetrate the back end (which I think they will fail unless they accept Windows is not suitable for an OS for everything, and that Microsoft could be selling its wares on other platforms - probably with some success - reference, Apple's only useful apps are IE and Office X - or at least a huge common denominator) to make more money than last quarter. Being in control of a monopoly and not paying dividends and losing drastic cash on XBoX and alienating users with oppressive licensing schemes and protections might be the beginning of their end. We will have to see how
Oh well, hope the guy who needs a 32 way Intel box enjoys it. I would just be ill watching big iron like that boot off a PC BIOS. SRM or Openrom would be nice
Legalize the constitution. Think for yourself question authority.
The Oracle&Linux server according to the article won't be available until May, by then Windows .NET Server 2003 will be out (out end of the year) and according to Microsoft will improve Windows/SQL Server results by 18-86% 'shattering the TPC-C benchmark' as they put it
http://www.microsoft.com/windows.netserver/evaluat ion/performance/tpcc.mspx