Enter your email above, then click "Sign Up" to join the STAC mail list and (optionally) register to access materials on the site. Click for terms.
New systems show generational improvement and the benefit of Big Memory
2-socket Ivy Bridge was 1.5x faster in the end-to-end Greeks benchmark than previously tested 2-socket Sandy Bridge.
September 27 - Results revealed, and Intel presents four steps to improve Monte Carlo performance.
System using kdb+ 3.1 set a record for NBBO performance and was at least 2x faster in 14 of the 17 response-time benchmarks than an identical system using kdb+ 2.8.
At 8x market rate, mean tick-to-trade latency was 5.2 microseconds with a standard deviation of 1.2 microseconds.
6.1 usec mean latency at market rate, with 1.4 usec jitter
Audited STAC-M3 Benchmark of Kx Systems kdb+ 2.8 with Intel SSD-based storage beats spindle-based benchmarks while using much less space