What are you looking for ?
Advertise with us
RAIDON

Securities Technology Analysis Center Report: kdb+ 3.6 With Dell EMC PowerScale F200 Under Stac-M3

Benchmarks on stack involving kdb+ 3.6 database system with F200 all-flash scale-out NAS cluster

STAC (Securities Technology Analysis Center LLC ) recently performed STAC-M3 Benchmarks on a stack involving the kdb+ 3.6 database system with a Dell EMC PowerScale F200 all-flash scale-out NAS cluster.

Dell Powerscale appliance F200

Dell Powerscale Appliance F200r

The results are available (registration required).

STAC-M3 is the set of standard enterprise tick-analytics benchmarks for database software/hardware stacks that manage large time series of market data (‘tick data’). In this project, it ran the baseline benchmark suite (code named Antuco).

The stack under test was Kx’s kdb+ 3.6 DBMS distributed across 9 PowerEdge R640 servers sharing a 3-node PowerScale F200 all-flash scale-out NAS cluster. Dell EMC (Dell Technologies, Inc.), chose to highlight that this solution involving 9 database servers accessing networked flash storage was/had:

  • Faster than a solution using Dell EMC’s larger flash storage appliance (SUT ID KDB190430) in 2 of 17 mean-response time STAC-M3 Antuco benchmarks, including:

      • 77% faster in the single-user VWAB operation (STAC-M3.v1.1T.VWAB-D.TIME)

      • 33% faster in the 10-user market snapshot operation (STAC-volume curve (STAC-M3.?1.10T.VOLCURV.TIME)

  • 7.7x the speed of a solution involving a parallel file system with 14 database servers and 18 storage servers (KDB200401) in the single-user NBBO operation (STAC-M3.β1.1T.NBBO.TIME )

  • Highest storage efficiency of any publicly reported solution involving kdb+ 3.6 (least storage used for the same database size)

      • 166% for this SUT vs 149% for all others (STAC-M3.v1.1.STORAGE.EFF)

Details are in the STAC report (registration required). Premium subscribers have access to the code used in this project and the micro-detailed configuration information for the solution.

Read also:
R&D: Stack Usage Analysis for Efficient Wear Leveling in Non-Volatile Main Memory Systems
Approach allows to save up to 85% of write overhead for wear-leveling in experiments.
October 2, 2020 | Press Release
Stac Report: kdb+ 4.0 With Intel Persistent Memory Pool Under Stac-M3
Solution sets records across scale and baseline benchmarks.
September 11, 2020 | Press Release

 

Articles_bottom
ExaGrid
AIC
ATTOtarget="_blank"
OPEN-E