Luke, -----Original Message----- From: Luke Lonergan [mailto:LLonergan@xxxxxxxxxxxxx] Sent: Tuesday, November 15, 2005 7:10 AM To: Adam Weisberg Cc: pgsql-performance@xxxxxxxxxxxxxx Subject: RE: Hardware/OS recommendations for large databases ( 5TB) Adam, > -----Original Message----- > From: pgsql-performance-owner@xxxxxxxxxxxxxx > [mailto:pgsql-performance-owner@xxxxxxxxxxxxxx] On Behalf Of Claus > Guttesen > Sent: Tuesday, November 15, 2005 12:29 AM > To: Adam Weisberg > Cc: pgsql-performance@xxxxxxxxxxxxxx > Subject: Re: Hardware/OS recommendations for large databases > ( 5TB) > > > Does anyone have recommendations for hardware and/or OS to > work with > > around 5TB datasets? > > Hardware-wise I'd say dual core opterons. One dual-core-opteron > performs better than two single-core at the same speed. Tyan makes > some boards that have four sockets, thereby giving you 8 cpu's (if you > need that many). Sun and HP also makes nice hardware although the Tyan > board is more competetive priced. > > OS wise I would choose the FreeBSD amd64 port but partititions larger > than 2 TB needs some special care, using gpt rather than disklabel > etc., tools like fsck may not be able to completely check partitions > larger than 2 TB. Linux or Solaris with either LVM or Veritas FS > sounds like candidates. I agree - you can get a very good one from www.acmemicro.com or www.rackable.com with 8x 400GB SATA disks and the new 3Ware 9550SX SATA RAID controller for about $6K with two Opteron 272 CPUs and 8GB of RAM on a Tyan 2882 motherboard. We get about 400MB/s sustained disk read performance on these (with tuning) on Linux using the xfs filesystem, which is one of the most critical factors for large databases. Note that you want to have your DBMS use all of the CPU and disk channel bandwidth you have on each query, which takes a parallel database like Bizgres MPP to achieve. Regards, - Luke The What's New FAQ for PostgreSQL 8.1 says "the buffer manager for 8.1 has been enhanced to scale almost linearly with the number of processors, leading to significant performance gains on 8-way, 16-way, dual-core, and multi-core CPU servers." Why not just use it as-is? Cheers, Adam ---------------------------(end of broadcast)--------------------------- TIP 4: Have you searched our list archives? http://archives.postgresql.org
