As of July 14, 2006
On July 14th, 2006 IBM published a new leading 10TB TPC-H benchmark result. This latest result with DB2 was run on a cluster of 16 8way p5 575s and delivered 180,108 QphH@10000GB with a price performance of On July 14th, 2006 IBM published a new leading 10TB TPC-H benchmark result. This latest result with DB2 was run on a cluster of 16 8way p5 575s and delivered 180,108 QphH@10000GB with a price performance of $47/QphH (system available Aug 30, 2006).
This result is 67% better than the #2 Oracle/Sun result (and the DB2 result was on 11% fewer cores). The DB2/This result is 67% better than the #2 Oracle/Sun result (and the DB2 result was on 11% fewer cores). The DB2/pSeries price performance was also 12% better.
Just as interesting is that DB2 running on a smaller number of CPUs delivered a load time for the 10TB database of just 2 hours and 44minutes where as the Oracle/Sun system took 18 hours 13 minutes to load the same amount of data (note load time includes Just as interesting is that DB2 running on a smaller number of CPUs delivered a load time for the 10TB database of just 2 hours and 44minutes where as the Oracle/Sun system took 18 hours 13 minutes to load the same amount of data (note load time includes creating of indexes and gathering stats as well).
About the benchmark
TPC-H™—Consists of a suite of business oriented ad-hoc queries and concurrent data modifications. This benchmark illustrates decision support systems that examine large volumes of data, execute queries with a high degree of complexity, and give answers to critical business questions. It measures performance at a particular scale - in this case 10 TB.
Required TPC information
DB2 V8.2 on 16 8way p5 575; 180,108 QphH@10000GB, $47/QphH@10000GB; Availability 08/30/2006
vs.
Oracle 10g on 144way Sun E25K; 108,099 QphH@10000GB, $53.80/QphH@10000GB; Availability 11/29/2005
TPC-H, QphH are trademarks of the Transaction Processing Performance Council. Visit www.tpc.org for full details