Solution Performance

Kdb+ is a time series database built for the Financial Services Industry to handle historical data used by 20 out of 21 top global banks. The STAC Antuco benchmark and STAC Kanaga benchmark test multiple aspects of database and HW performance on a typical customer workload. Both are IO-stressing benchmarks with various types of access (random-like, sequential like, with overlap, etc.)

Visit www.kx.com ›

"Our customers are at the forefront of high frequency sensor analytics, financial trading, streaming data surveillance, regulatory monitoring, and real time analytics. To stay competitive, they are in the chase for increased amounts of online memory in order to improve their “time to result” of time-series data analytics. Intel® Optane™ DC persistent memory provides our customers with the opportunity to achieve this by giving customers systems with a massively increased memory set. As both an in-memory and on disk database, we have always pushed against the limits of memory technology with kdb+. This recent testing with Storage over App Direct mode provided access to significantly larger sets of kdb+ in-memory data. Intel® Optane™ DC persistent memory truly enlarges the capabilities of Kx technology by taking full advantage of its unique architecture."
-Mark Sykes, CTO of First Derivatives and COO, Kx

Challenges

Quick access to historical data is absolutely critical to electronic trading to get a competitive advantage. The dataset size is often too large to be stored in DRAM. Adding additional memory is often cost prohibitive, resulting in the need to scale out and increasing HW footprint.

Solution

Higher memory capacities of Intel® Optane™ DC persistent memory (App Direct Mode (Storage over AD)) facilitates storing larger datasets and more historical data closer to the CPU, delivering significant performance improvements (see charts)1 2 3 4, compared to using SSDs (current solution).

Value

Significantly better performance – Kx customers will benefit from being able to access historical data much faster than currently available solutions.

Product and Performance Information

1

Performance results are based on testing by Intel and Kx, and auditing by STAC, on Mar 25, 2019 and may not reflect all publicly available security updates. No product or component can be absolutely secure.

 

2

Kx kdb+* v3.6 (STAC* Antuco benchmark); OS: Red Hat Enterprise Linux* 7.6 kernel 3.10.0-957.el7.x86_64. Testing by Intel and Kx, and auditing by STAC, completed on March 25, 2019.

BASELINE: Lenovo ThinkSystem SR650*: 2S Intel® Xeon® Platinum processor 8280L, 2.7GHz, 28 cores, turbo, and HT on, BIOS IVE135N-2.10, 768GB of DDR4, 12 slots / 64GB / 2933 MT/s / DDR4 LRDIMM, 2 x 8TB Intel® SSD P4600 NVMe*, Red Hat Enterprise Linux* 7.6 kernel 3.10.0-957.el7.x86_64.

NEW: Lenovo ThinkSystem SR950*?: 2S Intel® Xeon® Platinum processor 8280L, 2.7GHz, 28 cores, turbo, and HT on, BIOS IVE135N-2.10, 768GB of DDR4, 12 slots / 64GB / 2933 MT/s / DDR4 LRDIMM, 12288GB of DCPMM, 24 slots / 512GB / 2666 MT/s, 2 x 8TB Intel® SSD P4600 NVMe*, Red Hat Enterprise Linux* 7.6 kernel 3.10.0-957.el7.x86_64.

3

Kx kdb+* v3.6 (STAC* Kanaga benchmark); Workload consists of historical data for 2011-2012; OS: Red Hat Enterprise Linux* 7.6 kernel 3.10.0-957.el7.x86_64. Testing by Intel and Kx, and auditing by STAC, completed on March 25, 2019.

BASELINE: Lenovo ThinkSystem SR950*?: 4S Intel® Xeon® Platinum processor 8280L, 2.7GHz, 28 cores, turbo, and HT on, BIOS PSE121C-1.50, 1532GB of DDR4, 24 slots / 64GB / 2666 MT/s / DDR4 LRDIMM, 6 x 8TB Intel® SSD P4600 NVMe*, Red Hat Enterprise Linux* 7.6 kernel 3.10.0-957.el7.x86_64.

NEW: Lenovo ThinkSystem SR950*?: 4S Intel® Xeon® Platinum processor 8280L, 2.7GHz, 28 cores, turbo, and HT on, BIOS PSE121C-1.50, 1532GB of DDR4, 24 slots / 64GB / 2666 MT/s / DDR4 LRDIMM, 12288GB of DCPMM, 24 slots / 512GB / 2666 MT/s, 6 x 8TB Intel® SSD P4600 NVMe*, Red Hat Enterprise Linux* 7.6 kernel 3.10.0-957.el7.x86_64.

4

Software and workloads used in performance tests may have been optimized for performance only on Intel® microprocessors. Performance tests, such as SYSmark* and MobileMark*, are measured using specific computer systems, components, software, operations, and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to www.intel.com/benchmarks.