summaryrefslogtreecommitdiff
path: root/VA
AgeCommit message (Expand)AuthorLines
2025-05-14VA: Change default to exp=1 (strong scaling)Birte Kristina Friesel-2/+2
2025-05-14VA: Add benchmark scriptBirte Kristina Friesel-0/+24
2025-05-14Use benchmark-specific input descriptions rather than opaque "#elements"Birte Kristina Friesel-8/+10
2025-05-13TRNS, VA: global params are not requiredBirte Kristina Friesel-5/+4
2025-05-13VA: Aspect Header: Use aspect-provided input_sizeBirte Kristina Friesel-2/+2
2025-05-13tracing: provide input_size and element_size in derived aspectBirte Kristina Friesel-2/+12
2025-05-13VA: Use common timer and timing aspect headersBirte Kristina Friesel-199/+29
2025-05-13VA: AspectC++ supportBirte Kristina Friesel-46/+206
2025-01-16VA: dos2unix; indent -linuxBirte Kristina Friesel-642/+767
2025-01-13store input size as longBirte Kristina Friesel-8/+8
2025-01-13VA: Add valgrind-ws benchmark script and baseline adjustmentsBirte Kristina Friesel-7/+49
2025-01-07BS, GEMV, TRNS, TS, VA: Add basic perf statsBirte Kristina Friesel-0/+6
2024-07-26VA hbm: reduce number of repetitionsBirte Kristina Friesel-1/+1
2024-07-25VA nmc: explicitly source SDKBirte Kristina Friesel-0/+2
2024-07-23VA: remove single-node with in/out on same nodeBirte Kristina Friesel-13/+3
2024-07-22VA baseline: configurable memcpy NUMA bindingBirte Kristina Friesel-18/+50
2024-07-22VA dimes-hetsim-nmc: variable input nodeBirte Kristina Friesel-2/+2
2024-07-17VA: do not count move_pages towards timerBirte Kristina Friesel-1/+1
2024-07-17VA hetsim nmc: nitsBirte Kristina Friesel-5/+5
2024-07-17VA memcpy: HBM supportBirte Kristina Friesel-25/+63
2024-07-17VA dimes benchmarks: re-introduce --resume; add memcpy baselineBirte Kristina Friesel-25/+31
2024-07-17VA: Remove legacy run scriptsBirte Kristina Friesel-136/+0
2024-07-17VA: Add baseline variant with memcpy overhead (always use local input data)Birte Kristina Friesel-5/+81
2024-07-15Revert "VA CPU version: log latency"Birte Kristina Friesel-2/+2
2024-07-15VA, GEMV: Add roofline benchmarksBirte Kristina Friesel-0/+25
2024-07-12VA CPU version: log latencyBirte Kristina Friesel-2/+2
2024-07-10VA: nitsBirte Kristina Friesel-13/+13
2024-07-05VA: log NUMA node of allocated rank(s)Birte Kristina Friesel-32/+52
2024-07-05VA baseline: use strong scaling by defaultBirte Kristina Friesel-1/+1
2024-07-04VA: write directly to output; no teeBirte Kristina Friesel-3/+3
2024-07-04VA: add separate benchmark scripts for NMC and HBMBirte Kristina Friesel-28/+100
2024-07-04VA: Remove legacy SDK_SINGLETHREADED macroBirte Kristina Friesel-12/+5
2024-07-04VA: Add NUMA baseline variantBirte Kristina Friesel-24/+145
2024-03-11VA: handle alloc/load/free overhead as parameters; vary them for fgbs24aBirte Kristina Friesel-18/+11
2024-02-28some fgbs24a scriptsBirte Kristina Friesel-0/+28
2024-02-28VA: n_elements_per_dpuBirte Kristina Friesel-2/+2
2023-11-27VA: revert prim benchmarks to original configurations; host-specific logsBirte Kristina Friesel-8/+61
2023-11-27VA: add dpuinfo and singlethreaded optionsBirte Kristina Friesel-6/+29
2023-11-27VA: determine number of ranksBirte Kristina Friesel-0/+3
2023-11-21VA CPU baseline: rename misleading file_size to input_sizeBirte Kristina Friesel-6/+6
2023-11-21VA basilen: … watBirte Kristina Friesel-1/+0
2023-11-21UNI, VA: fix compilation without alloc/load/free benchmarksBirte Kristina Friesel-3/+3
2023-11-20VA: more detail outputBirte Kristina Friesel-7/+33
2023-11-20VA: optionally include dpu_{alloc,load,free} overhead in benchmark outputBirte Kristina Friesel-42/+96
2023-09-21VA: remove a stray newlineBirte Kristina Friesel-1/+1
2023-07-26update config spaceBirte Friesel-5/+6
2023-06-02VA cpu: limit to DPU typeDaniel Friesel-5/+9
2023-05-31VA: Correctly calculate PIM kernel throughputDaniel Friesel-2/+2
2023-05-31VA: cleanupDaniel Friesel-7/+10
2023-05-31VA: run.sh: up to 88 threads; strong scalingDaniel Friesel-2/+9
generated by cgit v1.2.3 (git 2.39.1) at 2025-08-07 22:15:43 +0000