| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | introduce verbosity aware subprocess.run wrapper and use it in benchmarks | Florian Fischer | 2019-12-13 | 1 | -27/+21 |
| | | |||||
| * | collect vmhwm for servers and use it for mysql, httpd and redis | Florian Fischer | 2019-12-11 | 1 | -8/+2 |
| | | |||||
| * | fmt benchmark definitions with yapf | Florian Fischer | 2019-12-11 | 1 | -35/+56 |
| | | |||||
| * | add and use write_tex_table function | Florian Fischer | 2019-10-30 | 1 | -0/+8 |
| | | |||||
| * | add lld and mysql documentation | Florian Fischer | 2019-10-21 | 1 | -5/+58 |
| | | |||||
| * | don't store sysbench version twice | Florian Fischer | 2019-10-16 | 1 | -3/+0 |
| | | |||||
| * | store mysql files persistent in build/benchmarks/mysql | Florian Fischer | 2019-10-13 | 1 | -23/+13 |
| | | |||||
| * | fix mysld server after server change | Florian Fischer | 2019-09-14 | 1 | -1/+1 |
| | | |||||
| * | explicitly collect binary versions | Florian Fischer | 2019-09-11 | 1 | -0/+5 |
| | | | | | | | | There is no generic way to retrieve the version of a binary. Not everyone followes GNU cli guidelines and supports "--version". The larson benchmark for example reads input from stdin when started with "larson --version" blocking the Benchmark.prepare() call. | ||||
| * | improve server benchmarks | Florian Fischer | 2019-09-11 | 1 | -1/+2 |
| | | | | | | | | servers are now dictionary with two mandatory keys: "name" and "cmd". Optional keys are: * "prepare_cmds": commands are run after a server is started * "shutdown_cmds": commands are run before a server is terminated | ||||
| * | require a name in Benchmark.__init__ | Florian Fischer | 2019-08-29 | 1 | -2/+2 |
| | | |||||
| * | readd missing numpy import for mysql summary | Florian Fischer | 2019-08-28 | 1 | -0/+2 |
| | | |||||
| * | revert accidentally added changes | Florian Fischer | 2019-08-28 | 1 | -2/+2 |
| | | |||||
| * | add license header comments to benchmark definitions | Florian Fischer | 2019-08-28 | 1 | -0/+19 |
| | | |||||
| * | remove verbosity argument from hooks | Florian Fischer | 2019-08-27 | 1 | -2/+3 |
| | | |||||
| * | improve benchmark code quality using pylint | Florian Fischer | 2019-08-27 | 1 | -17/+19 |
| | | | | | The description Benchmark member is replaced with class docstrings | ||||
| * | improve mysql benchmark definition | Florian Fischer | 2019-08-14 | 1 | -19/+26 |
| | | |||||
| * | fix and adjust to mysql installations on Ubuntu 18.04 | Florian Fischer | 2019-08-11 | 1 | -7/+8 |
| | | |||||
| * | don't copy benchmark.allocators again | Florian Fischer | 2019-08-10 | 1 | -6/+0 |
| | | |||||
| * | add facts to dataref export | Florian Fischer | 2019-07-31 | 1 | -2/+2 |
| | | |||||
| * | apply some flake8 hints | Florian Fischer | 2019-06-19 | 1 | -15/+12 |
| | | |||||
| * | introduce server concept to Benchmark | Florian Fischer | 2019-05-15 | 1 | -52/+9 |
| | | | | | | | | | | | | | A benchmark object can specify a list of cmds to execute as "servers" in the member attribute server_cmds. Servers are started and terminated through Popen objects. This requires the server cmds to not daemonize so the server can be terminated through the Popen object. For each started server cmd a shutdown function is registered with atexit to terminate all servers even if a exception occurs. Use the new server concept in httpd and mysql benchmarks | ||||
| * | add dataref exports | Florian Fischer | 2019-05-08 | 1 | -0/+1 |
| | | |||||
| * | improve descriptive statistics, csv-export and add general barplot function | Florian Fischer | 2019-04-13 | 1 | -49/+18 |
| | | |||||
| * | add analyze and server_benchmark feature | Florian Fischer | 2019-04-02 | 1 | -4/+4 |
| | | | | | | | | | | | | | | | | | | | | | | | --analyze uses malt to trace the benchmarks behavior. It uses the run loop but the obtained results are not stored Benchmark.server_benchmark is used if only a server is started for each allocator and clients are used to measure its performance in the run loop. If server_benchmark is set to True the cmds are run with the system default allocator. Misc changes: * The global environment is no longer changed. Instead a custom env dict is passed to suprocesses containing LD_PRELOAD. * Failing cmds no longer skip the whole benchmark instead they now skip the malfunctioning allocator. * Fix default title in plot_single_arg an analyse run are not stored | ||||
| * | move benchmark definitions into src/benchmarks | Florian Fischer | 2019-03-25 | 1 | -0/+285 |
| bench now loads all *.py files from src/benchmarks as benchmarks | |||||
 |
index : allocbench | |
| benchmark tool for POSIX memory allocators |
| aboutsummaryrefslogtreecommitdiff |