Constantin Fürst
|
b7cae18b6d
|
restructure the engine location bench, correct and update the plotter to use new total time
|
1 year ago |
Constantin Fürst
|
405166cbe8
|
add peak perf benchmark descriptors
|
1 year ago |
Constantin Fürst
|
148c4c213a
|
re-run mtsubmit tests
|
1 year ago |
Constantin Fürst
|
fb9164ae89
|
add script to disable dsa and then load a new config
|
1 year ago |
Constantin Fürst
|
9886b20112
|
remove the multiple tests for mtsubmit and ensure that the wq will always receive the same elements to make the test fair
|
1 year ago |
Constantin Fürst
|
01850cf97b
|
add new mtsubmit test results
|
1 year ago |
Constantin Fürst
|
3964da0d7a
|
use ms10 instead of ms50 as ms50 with > 2 threads will overfill the wq of size 128 (max) and cause an error unhandled in the benchmark yet
|
1 year ago |
Constantin Fürst
|
f9b00a5b32
|
modify mtsubmit to meassure both ssaw and ms50 submit methods
|
1 year ago |
Constantin Fürst
|
846fa9be43
|
re-run mtsubmit benchmarks for 4e
|
1 year ago |
Constantin Fürst
|
6f620d6461
|
re-run mtsubmit benchmarks for 1e
|
1 year ago |
Constantin Fürst
|
59853ffc92
|
remove the mtsubmit performance plotter which was based on a false assumption, modify the mtsubmit cost plotter for the new task set size
|
1 year ago |
Constantin Fürst
|
d72e28adb8
|
make mtsubmit a multisubmit
|
1 year ago |
Constantin Fürst
|
8f68f41ba0
|
position legend for optimal submit method in bottom right to make the entire plot readable
|
1 year ago |
Constantin Fürst
|
8ba84587be
|
add plotter for submit method with logscale to show optimal
|
1 year ago |
Constantin Fürst
|
4d4cc389c6
|
recreate the plots for mtsubmit with the new benchmark results
|
1 year ago |
Constantin Fürst
|
10b1f9018c
|
add plotted results from the new scripts
|
1 year ago |
Constantin Fürst
|
c27514890e
|
finish the move to save entire results and not condensed average in the plotter scripts
|
1 year ago |
Constantin Fürst
|
9e28ca5a27
|
re-run mtsubmit benchmarks
|
1 year ago |
Constantin Fürst
|
ee9119ea46
|
remove the 1gib benchmark results
|
1 year ago |
Constantin Fürst
|
573e312a69
|
update the submit method benchmark plotter
|
1 year ago |
Constantin Fürst
|
9c5b573beb
|
re-run benchmarks with current configuration
|
1 year ago |
Constantin Fürst
|
21a81c5aa0
|
do not pretty print the json files produced to save space
|
1 year ago |
Constantin Fürst
|
151ddd79d5
|
set repetition count of tests to 1000 again and replace the 1 GiB tests with 32 MiB
|
1 year ago |
Constantin Fürst
|
b7133346d8
|
add benchmark results
|
1 year ago |
Constantin Fürst
|
b44b52b600
|
meassure nanoseconds instead of microseconds
|
1 year ago |
Constantin Fürst
|
21194b7fde
|
fix the benchmarker script by testing for the correct amount of parameters and obtaining the test name correctly
|
1 year ago |
Constantin Fürst
|
e386370144
|
add dsa configuration files to repository
|
1 year ago |
Constantin Fürst
|
d3a09181ce
|
move the existing results into a new folder and modify them to be only benchmark descriptors
|
1 year ago |
Constantin Fürst
|
167370dbf9
|
prepare submitmethod benchmark plotter for the new result type
|
1 year ago |
Constantin Fürst
|
fb3cabb6e6
|
change naming of the json entries for timing, these are now not averages anymore but vectors
|
1 year ago |
Constantin Fürst
|
e6df656845
|
use vector for timing - this update got lost somehow
|
1 year ago |
Constantin Fürst
|
5b889b87e1
|
fix include naming
|
1 year ago |
Constantin Fürst
|
d0631f18e0
|
write script to re-run existing benchmarks in a given folder
|
1 year ago |
Constantin Fürst
|
bc8c4f8ab3
|
restructure of directory layout
|
1 year ago |
Constantin Fürst
|
8065dd4345
|
collect entire duration vector and dont condense the information down
|
1 year ago |
Constantin Fürst
|
d77230dece
|
use throughput as unit for benchmarks
|
1 year ago |
Constantin Fürst
|
b1213c1673
|
regenerate the mtsubmit plot with the new data
|
1 year ago |
Constantin Fürst
|
1ba3669aad
|
redo the benchmarks for mtsubmit that seem out-of-line
|
1 year ago |
Constantin Fürst
|
1405ef38fe
|
properly set the titles of the graphs and add parameter so they are not cut off
|
1 year ago |
Constantin Fürst
|
e129638a87
|
add benchmark plotting scripts using seaborn to display the results, the plots are also added
|
1 year ago |
Constantin Fürst
|
7c751aa2ce
|
format benchmark-plan.md so that it is legible in rendered version too
|
1 year ago |
Constantin Fürst
|
9fa15d9045
|
add benchmark-findings.md which is a scratchpad for things that stand out to me with the benchmark results
|
1 year ago |
Constantin Fürst
|
8705cd55ff
|
add benchmark results for submit, mtsubmit and crosscopy
|
1 year ago |
Constantin Fürst
|
151cadf3c7
|
formulate the benchmark descriptions more concisely, in benchmark-plan.md
|
1 year ago |
Constantin Fürst
|
854cd6916c
|
refrain from using a benchmarking script as this is too inflexible and a huge amount of work to create, manual mode it is ..., also create more concrete plan for benchmarks and the conclusions to draw from them in bechmark-plan.md (new)
|
1 year ago |
Constantin Fürst
|
071ed9be18
|
add missing read of iteration count to json task description reader
|
1 year ago |
Constantin Fürst
|
8c5a061343
|
remove buggy option for multiple sizes
|
1 year ago |
Constantin Fürst
|
897e363f44
|
fix forgotten add to vector in AddToTaskVector of ReadTaskData
|
1 year ago |
Constantin Fürst
|
1bfb1f316c
|
ignore first five runs to reduce influence of warmup, add non-batch-descriptor batch loop for testing, calculate standard deviation for all three messurements
|
1 year ago |
Constantin Fürst
|
9265ce6f65
|
add empty benchmarker-script which will orchestrate all benchmarks and print the results in the future, add comments about planned tests to this script and remove the benchmark notes markdown file which is now obsolete
|
1 year ago |