Comments on: Are analytic DBMS vendors overcomplicating their interconnect architectures? http://www.dbms2.com/2008/08/30/are-analytic-dbms-vendors-overcomplicating-their-interconnect-architectures/ Choices in data management and analysis Thu, 09 Feb 2012 21:52:17 +0000 hourly 1 http://wordpress.org/?v=3.0.3 By: Shawn Fox http://www.dbms2.com/2008/08/30/are-analytic-dbms-vendors-overcomplicating-their-interconnect-architectures/#comment-96812 Shawn Fox Fri, 05 Sep 2008 04:01:53 +0000 http://www.dbms2.com/?p=508#comment-96812 Netezza also uses gigabit ethernet to connect the various components of the database together. Every set of 14 SPUs connects to a SPA via gigabit ethernet. A SPU is a basically a node consisting of RAM, CPU, disk drive, and the FPGA. The SPA is a custom gigabit ethernet switch with additional functionality. The SPA then connects to a cisco gigabit ethernet switch. The Linux host is connected to the cisco switch via either a single or dual gigabit ethernet depending on the size of the s ystem. This yields 180MB/second or so throughput between host and the SPUs. Netezza also uses gigabit ethernet to connect the various components of the database together.

Every set of 14 SPUs connects to a SPA via gigabit ethernet. A SPU is a basically a node consisting of RAM, CPU, disk drive, and the FPGA. The SPA is a custom gigabit ethernet switch with additional functionality. The SPA then connects to a cisco gigabit ethernet switch. The Linux host is connected to the cisco switch via either a single or dual gigabit ethernet depending on the size of the s ystem. This yields 180MB/second or so throughput between host and the SPUs.

]]> By: Curt Monash http://www.dbms2.com/2008/08/30/are-analytic-dbms-vendors-overcomplicating-their-interconnect-architectures/#comment-96646 Curt Monash Wed, 03 Sep 2008 22:24:09 +0000 http://www.dbms2.com/?p=508#comment-96646 Nair, Thanks for the catches!! There actually was a THIRD broken link. Grrr. That's what happens when one forgets to test ... CAM Nair,

Thanks for the catches!!

There actually was a THIRD broken link. Grrr. That’s what happens when one forgets to test …

CAM

]]> By: Steve Wooledge http://www.dbms2.com/2008/08/30/are-analytic-dbms-vendors-overcomplicating-their-interconnect-architectures/#comment-96618 Steve Wooledge Wed, 03 Sep 2008 18:07:44 +0000 http://www.dbms2.com/?p=508#comment-96618 David Cheriton, the lead of the distributed systems group at Stanford University, wrote a nice post about this topic: http://www.asterdata.com/blog/index.php/2008/06/17/network-interconnect-the-greatest-challenge-in-massively-parallel-processing-mpp-systems/ David Cheriton, the lead of the distributed systems group at Stanford University, wrote a nice post about this topic: http://www.asterdata.com/blog/index.php/2008/06/17/network-interconnect-the-greatest-challenge-in-massively-parallel-processing-mpp-systems/

]]> By: Nair http://www.dbms2.com/2008/08/30/are-analytic-dbms-vendors-overcomplicating-their-interconnect-architectures/#comment-96612 Nair Wed, 03 Sep 2008 16:56:30 +0000 http://www.dbms2.com/?p=508#comment-96612 These 2 links are broken - http://www.dbms2.com/community/node/30307 http://www.dbms2.com/news/2008/082208-network-architecture.html?ts0hb= These 2 links are broken –
http://www.dbms2.com/community/node/30307
http://www.dbms2.com/news/2008/082208-network-architecture.html?ts0hb=

]]>