Exasol and AMD Achieve World Record Results on TPC‑H Benchmark
“Best ever” results for a 1 TB scale factor on industry standard decision support benchmark were achieved with AMD EPYC™ 7002 series processors
London, 10 September 2019 – Exasol, the analytics database, is proud to announce world record “best ever” results on industry standard decision support benchmark TPC‑H. Exasol has maintained its position as the undisputed leader – by a significant margin – for both raw performance and price-performance for over a decade across several categories. In July, the company announced outstanding results for 3 TB and 10 TB scale factors. Now, running the benchmark on AMD EPYC™ 7002 series processor-based servers, records have been broken for the 1 TB scale factor category.
“We are very excited to be working with AMD on new and more ambitious benchmarks to put both Exasol and AMD EPYC processor-based servers to the test. The results we have achieved with the new generation processors are the best ever for this category in particular,” said Mathias Golombek, CTO at Exasol.
The benchmark was completed on a scale factor of 1 Terabyte (TB) on 4 nodes and 8 nodes, demonstrating outstanding scalability and breaking records for performance and price-performance. At over 6.1 million QphH@1TB, this is the best performance ever recorded at the 1 TB scale factor. The price per performance result, also the best ever recorded for this category, is $0.06.
These results, available as of 31 July 2019, demonstrate Exasol’s speed, scalability and cost/performance which have consistently dominated the TPC‑H benchmarks since 2008. The latest results were achieved with Exasol 6.2 running on a cluster of eight HPE ProLiant DL325 Gen10 servers, each with one AMD EPYCTM 7702 64 core processor.
“We are pleased that Exasol has once more achieved remarkable benchmark results, this time using 2nd Generation AMD EPYC processor-based servers,” said Raghu Nambiar, corporate vice president and CTO, Datacenter Ecosystems & Application Engineering, AMD. “Performance demands are increasing but IT budgets and data center space are not. Built on 7nm technology, AMD EPYC™ 7002 Series processors pack more cores and more capability into the same space as the previous generation, delivering exceptional performance and scalability”
Over the years, there have been over 220 recorded TPC‑H benchmarks. Many database vendors have posted results on a vast range of hardware and at various scale factors up to 100TB. By running the same unbiased scripts, you can compare database vendor with database vendor – and you can often see how well the same database runs on different hardware.
Notes:
The full TPC‑H benchmark results are as follows and are available as of 07/31/19. Full results can be found in the link provided below.
SF 1000
Availability Date: 07/31/19
Performance: 6,145,628 QphH@1000GB
Price/Performance: 0.06 USD per QphH@1000GB
System: 8x HPE ProLiant DL325 Gen10 servers, each with:
256 GB RAM, 1x AMD EPYCTM 7702, 1x 1.6TB NVMe
Reference URL: http://www.tpc.org/3341
Claims Current of: 07/31/19
About the TPC‑H benchmarks
The TPC‑H benchmark is a performance test of the TPC, a non-profit corporation founded to define database benchmarks and to disseminate objective, verifiable performance data to the industry. The TPC‑H benchmark is a decision support benchmark; it consists of a suite of business-oriented ad-hoc queries and concurrent data modifications. The queries and the data populating the database have been chosen to have broad industry-wide relevance. 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.
About Exasol
Exasol is the analytics database. Its high-performance in-memory analytics database gives organizations the power to transform how they work with data, on-premises, in the cloud or both – and turn it into value faster, easier and more cost effectively than ever before.
To learn more about Exasol please visit www.exasol.com