Today DDN announced record performance on the Securities Technology Analysis Center (STAC) benchmark. Using the company’s EXAScaler storage solution, DDN set new public records for multiple workload types and sizes, including large and small workloads as well as I/O and compute-intensive workloads.
DDN has been the primary storage vendor for a number of extreme performance and scale-oriented HPC installations that have stringent uptime and reliability requirements,” said Steve Conway, IDC research vice president, high performance computing and data analysis. “The STAC results indicate that DDN’s performance and capability advantages are well-suited to the financial services industry, which has data intensive workloads and challenging uptime and reliability requirements.”
DDN storage outperformed SSD-based solutions to deliver higher performance, productivity and density at lower cost. The independent and audited STAC-M3 Benchmark results validate DDN’s leadership in providing high performance at scale for financial services organizations.
The STAC-M3 Benchmark suite is the industry standard for testing solutions that enable high-speed analytics on time series data. The STAC-tested system consisted of Kx Systems kdb+ 3.2 running DDN SFA12KX-40 based EXAScaler platform with Intel Enterprise Edition for Lustre 2.2.
As the STAC results demonstrate, financial services organizations can benefit from DDN’s EXAScaler platform as it enables a shared data model, empowers the consolidation of data and massively parallel data queries for market data analytics, and reduces data management costs – all while delivering extremely high performance.
The STAC benchmark illustrates why financial services customers in trading firms, hedge funds and investment banks are rapidly moving to DDN Storage. They are able to accelerate their data models to deliver significantly faster results, while simultaneously lowering TCO of the entire data management solution,” said Molly Rector, CMO, executive vice president product management and worldwide marketing, DDN.
Leading global financial institutions test their automated trading and risk management algorithms against years of compiled market data provided by STAC. In this study, two benchmark suites, ANTUCO and KANAGA, were executed. The STAC suite ANTUCO used a limited dataset size of 4.5TB to simulate system performance of a dataset residing mostly on non-volatile media with a broad range of read and write operations. The KANAGA suite analyzed performance on large datasets that range from 33TB to 897TB, with large numbers of concurrent requests.
The DDN EXAScaler high performance storage platform excelled in each aspect of the benchmark testing:
- Demonstrated 2.4x improvement in time sensitive and I/O intensive financial analytics (querying the maximum bid over the year for 1% of the symbols);
- Set world record performance for large scale (33TB to 897TB) KANAGA benchmarks; and
- Proved 50 percent better latency characteristics on 13 of the most I/O intensive workloads versus previous benchmarks.