v0.2

What's Changed

• ansible to create k8s cluster on equinix by @rootfs in #1
• auto shutdown k8s clusters by @rootfs in #2
• add k8s workflow by @rootfs in #3
• fix env by @rootfs in #4
• fix ssh key path by @rootfs in #5
• fixmetal download by @rootfs in #6
• sudo to chmod by @rootfs in #7
• prepare metal config by @rootfs in #8
• use pre-created metal config by @rootfs in #9
• allow user to set termination time by @rootfs in #10
• Adding Prom by @jtaleric in #12
• Adding First Perf test by @jtaleric in #13
• Wait for monitoring pods to be running by @jtaleric in #20
• Need to wait on statefulset by @jtaleric in #21
• integrate with kepler action as BM self host runner provider by @SamYuan1990 in #23
• set the kepler service unit file so to expose idle power by @rootfs in #25
• k8s action: monitor kepler cpu utilization by @rootfs in #28
• kepler stress test results by @rootfs in #29
• fix path by @rootfs in #30
• refactor by @rootfs in #31
• set git push by @rootfs in #32
• use checkout action by @rootfs in #33
• fix relative path by @rootfs in #34
• reformat output by @rootfs in #35
• add markdown files to create github pages by @rootfs in #36
• add nightly run schedule by @rootfs in #38
• don't overwrite validator queries by @rootfs in #39
• publish validation results by @rootfs in #40
• fix path by @rootfs in #41
• don't set global git config on self hosted runner by @rootfs in #42
• fix bug for kepler-action integration by @SamYuan1990 in #27
• fix schedule by @rootfs in #43
• support schedule trigger by @rootfs in #47
• switch to kepler action main by @SamYuan1990 in #44
• turn on logging by @rootfs in #48
• update commit message by @rootfs in #49
• add vm cpu pinning by @rootfs in #50
• try to integrate with kepler action by @SamYuan1990 in #46
• cancel schedule for action by @SamYuan1990 in #51
• fix(validator): adapt to changes to validator by @sthaha in #52
• update with latest validation config by @rootfs in #53
• chore: Fix typo in validation by @dave-tucker in #54
• make the vm use the entire bare metal instance by @rootfs in #55
• use default stressor by @rootfs in #56
• disable component power and bpf metrics in stress tests by @rootfs in #57
• Integrate Model Trainer by @KaiyiLiu1234 in #58
• feat(trainer): Update Trainer by @KaiyiLiu1234 in #64
• as kepler action support Tekton update code accordingly by @SamYuan1990 in #61
• fix(model_server_test): Add placeholder for model server test to main by @KaiyiLiu1234 in #68
• fix: 0 metrics found error do to large scrape-config by @sthaha in #71
• rename job=node to job=metal by @sthaha in #73
• chore: track cpu usage of both kepler and kube-apiserver in stress tests by @rootfs in #74
• fix markdown table rendering issue by @rootfs in #77
• slim down markdown table by @rootfs in #78
• fix broken table by @rootfs in #79
• try to fix table rendering issue again by @rootfs in #80
• try to fix table rendering issue again by removing offending characters by @rootfs in #81
• add stress test result regression checker bot by @rootfs in #82
• don't update report in kepler-action by @rootfs in #84
• tune the stress detection prompt by @rootfs in #85
• add validation regression checker by @rootfs in #86
• add validation regression checker by @rootfs in #87
• tune prompt and switch to gpt-4-turbo by @rootfs in #88
• bot: ignore idle mode validation regression by @rootfs in #91
• chore(metal): show verbose output by @sthaha in #106
• fix(metal): adapt to change to report path by @sthaha in #107
• feat(model_server): Incorporate Model Server by @KaiyiLiu1234 in #110
• feat(model_server): Add Separate Workflow for Model Server Implementation by @KaiyiLiu1234 in #112
• feat(model_server): Add Model Server to Metal CI by @KaiyiLiu1234 in #111
• chore(model_server): Remove Schedule and Git Push for Testing by @KaiyiLiu1234 in #118
• chore: add a k8s provisioning flow by @rootfs in #121
• chore: create kepler namespace first before setting prometheus role by @rootfs in #122
• fix(model_server): change workflow time to 11pm UTC by @KaiyiLiu1234 in #124
• chore(model_server): Change workflow name by @KaiyiLiu1234 in #126
• chore: add kepler validation diagrams by @rootfs in #142
• chore: don't re-setup python by @rootfs in #143
• chore: when rendering the comparison chart, reset the timestamp so that charts are not stretched by @rootfs in #146
• chore: decouple kepler and open metrics by @rootfs in #147
• chore: fine tune the prompt in regression detection by @rootfs in #148
• feat(model_trainer): Add Trainer Ansible Playbook and Workflow by @KaiyiLiu1234 in #131
• fix(model_server): Properly Activated new models by @KaiyiLiu1234 in #151
• fix(model_server): add model_server ansible vars by @KaiyiLiu1234 in #152
• chore(model_server): Update Model Server Version to 0.7.11-2 by @KaiyiLiu1234 in #156
• feat(model_e2e): Added Model e2e by @KaiyiLiu1234 in #163
• chore: add metal sweep action by @rootfs in #171
• WIP added node-density-churn by @shashank-boyapally in #170
• Node density churn new action name by @shashank-boyapally in #172
• chore: migrate to official equinix metal runner action by @vprashar2929 in #165
• Node density churn by @shashank-boyapally in #177
• chore: Reduce the vm resource allocation by @vprashar2929 in #178
• chore(rapl): Add a step to capture RAPL power domain availability by @vimalk78 in #179
• feat(model_e2e): Train and Validate Models on Equinix Server by @KaiyiLiu1234 in #182
• Update e2eSeparateServers with Changes from Main by @KaiyiLiu1234 in #185
• feat(model_e2e): Add Separate Server Workflow e2e by @KaiyiLiu1234 in #186
• chore(equinix_runner): Replaced Create-runner and Cleanup with reusable workflows by @KaiyiLiu1234 in #190
• chore(metal/flow): keep reports and artifacts its own dir by @sthaha in #198
• chore(ci): Update workflows for consistent artifact handling by @vprashar2929 in #199
• chore(push_validations): Push Validation Results to Github Repo by @KaiyiLiu1234 in #206
• chore(rename_workflows): Renamed workflows to be more descriptive by @KaiyiLiu1234 in #209
• chore(push_validations): Move all Validation Data to Repo by @KaiyiLiu1234 in #210
• feat(pyproject): for analytics by @sthaha in #202
• chore: Add playbook to install node-exporter by @vimalk78 in #217
• chore: enable selected collectors in node-exporter by @vimalk78 in #224
• changed commands by @shashank-boyapally in #203
• chore(ci): update metal delete action to use sustainable-computing-io fork by @vprashar2929 in #222
• fix: resolve node-exporter permission error for energy_uj access by @vprashar2929 in #229
• Merge main with overallErrorMetrics by @KaiyiLiu1234 in #234
• Add node config to model server validations workflow by @KaiyiLiu1234 in #235
• chore: selectively enable sweep by @rootfs in #236
• chore: use isolated cpu for the vm by @rootfs in #240
• Add Node Exporter to e2e by @KaiyiLiu1234 in #241
• indexing metrics by @shashank-boyapally in #237
• fix(node_config): Add hatch run to trainer by @KaiyiLiu1234 in #246
• fix(daily_validations): Enable selection of subset of error metrics by @KaiyiLiu1234 in #249
• chore: fix sweep workflow error by @rootfs in #250
• Update equinix_metal_sweep.yml by @rootfs in #252
• update actions by @shashank-boyapally in #255
• chore(ansible): refactor variable structure by moving host-specific and shared vars by @vprashar2929 in #256
• chore(ansible): refactor kvm_vm role by @vprashar2929 in #258
• feat: Add support for capturing Prometheus snapshot by @vprashar2929 in #266
• docs: add steps for using Prometheus Snapshot by @vprashar2929 in #277
• chore: add poly model in analytics by @rootfs in #280
• fix(setup_runner): Added setup runner acton in Validation Scripts by @KaiyiLiu1234 in #276
• update analytics vm-cpu-time-predict-vm-power by @sunya-ch in #281
• feat(ansible): Introduce role for Kepler installation by @vprashar2929 in #270
• [CI] add date validation before create new github issue to avoid dupl… by @SamYuan1990 in #265
• fix(model_server): Resolve git error by @KaiyiLiu1234 in #282
• [CI] Hot fix for actions/upload-artifact and actions/download-artifact by @SamYuan1990 in #285
• [CI] add Dependency Bot settings by @SamYuan1990 in #284
• chore(ci): use latest equinix metal-runner-action by @vprashar2929 in #283
• add error handle for action steps by @SamYuan1990 in #288
• feat: add timestamp to validator directory by @vprashar2929 in #289
• feat(train_logs): Upload Training Logs by @KaiyiLiu1234 in #294
• increasing timeout so that churn can continue by @shashank-boyapally in #291
• chore: not expose estimated idle power by @rootfs in #295
• chore(ci): ensure workflow fails on errors by @vprashar2929 in #296
• feat: Add support for deploying Kepler using Compose by @vprashar2929 in #290
• feat(vm_metrics): Add vm metrics training process by @KaiyiLiu1234 in #297
• fix(switch_to_main): Switched model server branch to main repo by @KaiyiLiu1234 in #301
• fix(disable_daily_vals): Disable daily validations by @KaiyiLiu1234 in #302
• fix(vm_metrics): Add --vm-train tag by @KaiyiLiu1234 in #303
• chore: use vm-train option to train models by @rootfs in #305
• chore: use the latest model server image tag by @rootfs in #306
• chore: ignore model server start error by @rootfs in #307
• Update equinix_k8s_flow_churncheck.yml by @rootfs in #308
• chore(model-server): Use hatch to deploy Model Server by @KaiyiLiu1234 in #309
• chore: make the model server image tag a param by @rootfs in #310
• chore: add model train and validate report by @rootfs in #311
• chore(deps): bump the github-actions group across 1 directory with 2 updates by @dependabot in #287
• chore(validate_model_defaults): Added Polynomial and Xgboost to default list of target models by @KaiyiLiu1234 in #312
• chore: add aws metal e2e by @rootfs in #314
• fix(exclude-sweapper-process): Exclude swapper process on metal kepler by @KaiyiLiu1234 in #316
• chore: add GHA to create aws ec2 ami with centos stream 9 and nvidia driver by @rootfs in #317
• Creating AMI with NVIDIA driver by @rootfs in #319
• chore: install nvidia dgcm in ami by @rootfs in #321
• chore: use the right dcgm output by @rootfs in #322
• feat: add GPU operation by @rootfs in #320
• [Feat][Chore] Pr level protect by @SamYuan1990 in #325
• don't re-install crio if it is already there by @rootfs in #327
• chore: pre install crio and pytorch images to save test image by @rootfs in #328
• fix(enable_ssh_file_access): Remove enforce by @KaiyiLiu1234 in #324
• [Nit] code refactor by @SamYuan1990 in #313
• feat(aws): Incorporate AWS fully into Metal CI by @KaiyiLiu1234 in #315
• feat(prometheus_snapshot): Upload prometheus snapshot for trainer workflows by @KaiyiLiu1234 in #330
• feat(upload-train-logs): Upload train logs to repo by @KaiyiLiu1234 in #333
• fix(add-train-logs) Add .log extension to train logs by @KaiyiLiu1234 in #334
• chore: set validator runtime by @rootfs in #332
• fix(upload-train-logs): Add correct destination filepath for log upload by @KaiyiLiu1234 in #336
• fix(duplicate-env): Remove duplicate env keyword by @KaiyiLiu1234 in #337
• chore: add support for kubeburner in PR check by @SamYuan1990 in #329
• chore(deps): Bump actions/checkout from 3 to 4 in the github-actions group by @dependabot in #338
• [CI]: using artifacts to keep result, a part of #339 by @SamYuan1990 in #341
• [fix]: fix checkout code after config kube error by @SamYuan1990 in #342
• [fix]: add checkout action in steps and add more log when kepler not … by @SamYuan1990 in #343
• [fix]: attempt to use config-dir to fix config folder check issue by @SamYuan1990 in #344
• feat(ansible): add Kepler config file support by @vprashar2929 in #345
• [fix]: try with kubeconfig with try best by @SamYuan1990 in #346
• [fix]: ref latest pip model document try to set up httpx client for o… by @SamYuan1990 in #347
• [fix]: add missing env for estimator service by @SamYuan1990 in #349

New Contributors

• @jtaleric made their first contribution in #12
• @SamYuan1990 made their first contribution in #23
• @sthaha made their first contribution in #52
• @dave-tucker made their first contribution in #54
• @shashank-boyapally made their first contribution in #170
• @vprashar2929 made their first contribution in #165
• @sunya-ch made their first contribution in #281
• @dependabot made their first contribution in #287

Full Changelog: v0.1...v0.2