Guest blog by Martin Whittaker, VP of Engineering
HP servers offer outstanding performance running Oracle software in every environment, from industry standard to mission critical, from scale-out to scale-up, from ProLiant to Integrity. But customers can’t make critical comparisons on price and performance because Oracle won’t let them. They refuse to publish benchmark results with HP servers, even though they have audited benchmark results that are ready for publication.
HP Integrity servers running HP-UX power the world’s most demanding customer workloads. With the release in 2010 of Intel’s current Itanium processor (code-named “Tukwila”), HP set new standards for flexibility, resiliency, and performance, setting a new world record for single system business intelligence performance with HP’s Superdome 2 server(1). HP Integrity servers more than doubled their performance from the previous generation. This performance benchmark used an Oracle database. As it turns out, this was one of the last benchmark results published on an HP server using Oracle’s database software.
As HP prepares to incorporate the next version of Intel’s Itanium processor (code-named “Poulson”), and again looking to more than double performance for customers, Oracle’s reluctance to release benchmark results with HP servers is a sign of weakness. Maybe it’s because Oracle is not competitive with its own hardware and can only tout results that can’t be compared to HP. HP Integrity servers with HP-UX and Oracle database software outperform Oracle’s own servers running Oracle.(2) The last time Oracle agreed to publish a benchmark result with the DL980 G7 was in 2010 and it continues to be the leading result in its category (3). The DL980 G7 still delivers more than a 7x price/performance advantage over Oracle (4) and we’re ready to release another update to the DL980 G7 further improving its performance.
Oracle no longer supports publishing benchmarks with HP servers. We wonder why?
Follow this link for more information on how HP puts Customers first.
VP of Engineering
(1) Source: based on Superdome 2 1 TB TPC-H non-clustered result published 4/26/10. http://www.tpc.org/tpch/results/tpch_result_detail
(2) Source: based on Superdome 2 1 TB TPC-H non-clustered result published 4/26/10. http://www.tpc.org/tpch/results/tpch_result_detail
(3) Source: based on DL980 G7 SPECjAppServer2004 result published 9/20/10: http://www.spec.org/jAppServer2004/results/res2010
(4)Source: based on DL980 G7 3TB TPC-H non-clustered result published 6/21/10. http://www.tpc.org/tpch/results/tpch_result_detail