# save results from 5 iterations text example above # Azure Blob Storage or Google Cloud Storage It can also save artifacts (HTML, JSON and text results) to S3, # save.sh saves results to CSV, MySQL, PostgreSQL, BigQuery or via HTTP run.sh -output ~/geekbench-testing/$i done # run 10 test iterations using a specific output directoryįor i in do mkdir -p ~/geekbench-testing/$i. run.sh -geekbench_dir /usr/local/Geekbench -upload # run with Geekbench installed in /usr/local/Geekbench and upload results to public Geekbench browser run.sh -meta_compute_service_id aws:ec2 -meta_instance_id c3.xlarge -meta_region us-east-1 -meta_test_id aws-0914 # run 1 test iteration with some metadata Removed using the save.sh -remove parameterĬlaim_url: Ĭollectd_rrd: Ĭrypto_multicore_score: Ĭrypto_score: įloating_point_multicore_score: įloating_point_score: Indexing used is documented in save/schema/geekbench3.json. Columns withoutĭescriptions are documented as runtime parameters above. Indexed MySQL/PostgreSQL columns are identified by *. The following columns are included in CSV files/tables generated by save.sh. Geekbench.txt Text formatted geekbench results Geekbench.json JSON formatted geekbench results Geekbench.html HTML formatted geekbench report This benchmark generates the following artifacts:Ĭollectd-rrd.zip collectd RRD files (see -collectd_rrd) Php-cli Test automation scripts (/usr/bin/php) This benchmark has the following dependencies: Geekbench results browser (includes system * upload Upload results upon completion to the public The current working directory will be used * output The output directory to use for writing test data Using the environment variable bm_test_id * meta_storage_config Storage configuration descriptor. Specified using the environment variable bm_run_id * meta_run_id An optional benchmark run identifiers. MayĪlso be specified using the environment variable * meta_resource_id An optional benchmark resource identifiers. May also be specified using the environment * meta_region The compute service region this test pertains to. * meta_provider_id The id of the cloud provider this test pertains * meta_provider The name of the cloud provider this test pertains It will be taken from the first line of /etc/issue * meta_os Operating system descriptor - if not specified, * meta_memory Memory descriptor - if not specified, the system * meta_instance_id The compute service instance type this test pertains Using the 'model name' attribute in /proc/cpuinfo * meta_cpu CPU descriptor - if not specified, it will be set Using the environment variable bm_compute_service_id * meta_compute_service_id The id of the compute service this test pertains * meta_compute_service The name of the compute service this test pertains * meta_burst If set to 1, designates testing performed in burst Registered and generate an error if it has not Test harness will check if Geekbench has been ![]() 'Geekbench-5.0.4-Linux' directory with anĮxecutable 'geekbench_x86_64' executable in it. Presence of a 'build.pulse/dist/Geekbench-5.0.4-Linux' or The directory tree from both pwd and -output for Specified, the benchmark run script will look up * geekbench_dir Directory where Geekbench is installed. * collectd_rrd_dir Location where collectd rrd files are stored. bak will be zippedĪnd saved along with other test artifacts (asĬollectd-rrd.zip). bak, and upon test completionĪny directories not ending in. ![]() To do so, when testing starts,Įxisting directories in -collectd_rrd_dir willīe renamed to. * collectd_rrd If set, collectd rrd stats will be captured from Review the 'geekbench_dir' parameter comments below Geekbench should be installed and registered before Test behavior is fixed, but you may specify the following optional metaĪttributes and installation attributes. NOTE: When cloning this repository, please use the -recurse-submodules option Must first purchase a license key, then download the Geekbench software, andįinally enter your registration key - e.g. This benchmark requires a registered version of Geekbench. More information about Geekbench is available on their website: Geekbench is a commercial benchmark developed by Primate This repository contains the setup and runtime configurations for the Geekbench
0 Comments
Leave a Reply. |