{"id":13667039,"url":"https://github.com/sslab-gatech/vbench","last_synced_at":"2025-04-06T02:30:35.422Z","repository":{"id":34714069,"uuid":"38690783","full_name":"sslab-gatech/vbench","owner":"sslab-gatech","description":"VBench: Multicore Scalability Application Benchmark","archived":false,"fork":false,"pushed_at":"2019-08-19T16:43:06.000Z","size":17168,"stargazers_count":9,"open_issues_count":1,"forks_count":5,"subscribers_count":22,"default_branch":"master","last_synced_at":"2025-03-21T15:45:28.694Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/sslab-gatech.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2015-07-07T13:52:04.000Z","updated_at":"2024-12-18T02:04:18.000Z","dependencies_parsed_at":"2022-09-14T18:40:37.253Z","dependency_job_id":null,"html_url":"https://github.com/sslab-gatech/vbench","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sslab-gatech%2Fvbench","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sslab-gatech%2Fvbench/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sslab-gatech%2Fvbench/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sslab-gatech%2Fvbench/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/sslab-gatech","download_url":"https://codeload.github.com/sslab-gatech/vbench/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247425769,"owners_count":20937011,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-08-02T07:00:26.579Z","updated_at":"2025-04-06T02:30:35.112Z","avatar_url":"https://github.com/sslab-gatech.png","language":"Python","funding_links":[],"categories":["Benchmarks"],"sub_categories":["Mobile Computing"],"readme":"### VBench\n\nVBENCH is a fork of MOSBENCH but allows running benchmarks inside\nVM as well.\n\n## Supported benchmarks:\n\n- Currently, we have included the following benchmarks for evaluation:\n  - Gmake: Fork intensive, but spends most of the time in the userspace.\n  - Exim: Crazily stresses `mm`, `fs`, and `process` sub-systems.\n  - Psearchy: Stresses the interrupt service, used to stress the writer side of the `mmap_rwsem`.\n  - Metis: Stresses the reader side of the `mmap_sem`.\n  - Hist: Stresses the reader side of the `mmap_sem`.\n\n## Setup environment\n\n- First compile the relevant benchmarks as follows:\n  - Exim: `make all-exim`\n  - Psearchy: `make all-psearchy`\n  - Metis: `make all-metis`\n  - Hist: `make all-hist`\n\n- Configure the following variables in the `config.py`:\n```python\n    ROOT = \"~/vbench\" # Provide the relative path (this is also used in the VM)\n    CPU_CNTS = 160 # Total number of cores (cores * sockets)\n    CORES_PER_SOCKET = 10\n    CPU_STEP = 10 # Numbers of cores to increase until CPU_CNTS in each run\n```\n\n- Modify the hostname in the hosts.py. You can set it to `127.0.0.1` and disable sanity check in `mparts/host.py`\n\n- Create multiple `tmpfs` partitions with the following command:\n```bash\n  $ sudo ./mkmounts tmpfs-separate # sudo is mandatory\n```\n\n- After this, you can use `config.py` to execute benchmarks.\n```bash\n $ ./config.py -h\n\nOptions:\n  -h, --help            show this help message and exit\n  -d, --dryrun          dry run (default: sanityRun = False)\n  -t TRIALS, --trials=TRIALS\n                        trials (default: TRY = 1)\n  -c CORES, --cores=CORES\n                        number of cores for dryrun, default = 80\n  -n, --profile-nos     get numbers using perf stat\n  -r, --perf-record     record all functions\n  --kvm-record          perf record for guest, only works with -r option\n  --guest-only-record   only records the guest, no host side recording\n  -p, --pin             pin all the tasks in qemu process\n  -l, --lockstats       get lock stats\n  -m, --multivm         run two VMs simultaneously\n  --kvm-stat            get the kvm stats if possible (only possible for PVM)\n  --trace-kvm           get the results using trace-cmd\n```\n\n### Executing a benchmark\n- To run a benchmark, here `exim`, across all cores, execute the following:\n```bash\n $ ./config.py exim # runs the exim benchmark from core 1 to core 160.\n```\n\n- After running the benchmark, results are displayed on the terminal\n  and are stored in the results folder, which can be used for plotting\n  by using the `./graph` script.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsslab-gatech%2Fvbench","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsslab-gatech%2Fvbench","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsslab-gatech%2Fvbench/lists"}