Re:Using SIMD to aggregate billions of values per second
by bluestreak on Wednesday April 08, 2020 @05:39AM
(#22811441)
these are against the latest 4.0 KDB+, after 26 March. KDB before that could not aggregate in parallel implicitly.
Re:Using SIMD to aggregate billions of values per second
by binomiq on Wednesday April 08, 2020 @09:37AM
(#22812813)
In your benchmarks, KDB's max on longs takes twice as long as sum on longs. I am not able to replicate this with the most recent version of KDB (2020.03.30).
$QHOME/l64/q -s 4
KDB+ 4.0 2020.03.30 Copyright (C) 1993-2020 Kx Systems
l64/ 4()core 516718MB <snip>
q)zz:1000000000?1000j
q)0.01*system"t do[100;max zz]"
257.52
q)0.01*system"t do[100;sum zz]"
251.95
There's a marginal difference here rather than double.For the version of KDB with the regression bug:
q)0.01*system"t do[100;max zz]"
512.21
q)0.01*system"t do[100;sum zz]"
254.24
Which starts to look more like your numbers.
Re:Using SIMD to aggregate billions of values per second
by bluestreak on Wednesday April 08, 2020 @10:13AM
(#22813097)
sorry, i think we tested with this one:
KDB+ 4.0 2020.03.17 Copyright (C) 1993-2020 Kx Systems
l64/ 12(16)core 63960MB
EXPIRE 2021.03.26
From the KDB+ 4.0 release notes:
2020.03.26 FIX fixed performance regression for max. e.g. q)x:100000000?100;system"ts:10 max x"