{"id":44542412,"url":"https://github.com/purestorage-openconnect/lakebench-k8s","last_synced_at":"2026-04-05T20:02:16.539Z","repository":{"id":338167748,"uuid":"1156821042","full_name":"PureStorage-OpenConnect/lakebench-k8s","owner":"PureStorage-OpenConnect","description":"Deploy, benchmark, and compare lakehouse architectures on Kubernetes with a single YAML file. Lakebench provisions Spark, Iceberg, and Trino, generates synthetic data at any scale, runs medallion pipelines (Bronze→Silver→Gold), and scores performance with an 8-query benchmark across pluggable catalogs, table formats, and query engines. ","archived":false,"fork":false,"pushed_at":"2026-03-27T21:11:48.000Z","size":1203,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-03-28T03:44:44.225Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/PureStorage-OpenConnect.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":"docs/supported-components.md","governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-02-13T04:54:28.000Z","updated_at":"2026-03-27T21:09:30.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/PureStorage-OpenConnect/lakebench-k8s","commit_stats":null,"previous_names":["purestorage-openconnect/lakebench-k8s"],"tags_count":15,"template":false,"template_full_name":null,"purl":"pkg:github/PureStorage-OpenConnect/lakebench-k8s","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/PureStorage-OpenConnect%2Flakebench-k8s","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/PureStorage-OpenConnect%2Flakebench-k8s/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/PureStorage-OpenConnect%2Flakebench-k8s/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/PureStorage-OpenConnect%2Flakebench-k8s/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/PureStorage-OpenConnect","download_url":"https://codeload.github.com/PureStorage-OpenConnect/lakebench-k8s/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/PureStorage-OpenConnect%2Flakebench-k8s/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31448216,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-05T15:22:31.103Z","status":"ssl_error","status_checked_at":"2026-04-05T15:22:00.205Z","response_time":75,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"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":"2026-02-13T19:01:51.632Z","updated_at":"2026-04-05T20:02:16.528Z","avatar_url":"https://github.com/PureStorage-OpenConnect.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Lakebench\n\n[![Python 3.10+](https://img.shields.io/badge/python-3.10%20|%203.11%20|%203.12%20|%203.13-blue)](https://www.python.org/downloads/)\n[![License](https://img.shields.io/badge/license-Apache%202.0-green)](LICENSE)\n\n**A/B testing for lakehouse architectures on Kubernetes.**\n\nDeploy a complete lakehouse stack from a single YAML, run a medallion pipeline\nat any scale, and get a scorecard you can compare across configurations.\n\n\u003c!-- TODO: Add terminal recording / screenshot of `lakebench run` output here --\u003e\n\n## Why Lakebench?\n\n- **Compare stacks.** Swap catalogs (Hive, Polaris), query engines (Trino,\n  Spark Thrift, DuckDB), and table formats -- same data, same queries,\n  different architecture. Side-by-side scorecard comparison.\n- **Test at scale.** Run the same workload at 10 GB, 100 GB, and 1 TB to find\n  where throughput plateaus or resources saturate on your hardware.\n- **Measure freshness.** Sustained mode streams data through the pipeline and\n  benchmarks query performance under sustained ingest load.\n\n## Quick Start\n\n```bash\npip install lakebench-k8s\n```\n\n\u003e Pre-built binaries (no Python required) are available on\n\u003e [GitHub Releases](https://github.com/PureStorage-OpenConnect/lakebench-k8s/releases).\n\n```bash\npip install lakebench-k8s\nlakebench init                           # quick setup (4 questions)\nlakebench run lakebench.yaml --generate  # deploy + generate + pipeline + benchmark\nlakebench results                        # view scorecard\nlakebench destroy lakebench.yaml         # tear down everything\n```\n\nMinimum config -- 4 lines:\n\n```yaml\n# lakebench.yaml\nendpoint: http://s3.example.com:80\naccess_key: YOUR_KEY\nsecret_key: YOUR_SECRET\nscale: 10                          # 1 = ~10 GB, 10 = ~100 GB, 100 = ~1 TB\n```\n\nName is auto-generated. Recipe defaults to `hive-iceberg-spark-trino`.\nOverride anything with flat fields or nested YAML:\n\n```yaml\n# lakebench.yaml (with overrides)\nname: flashblade-polaris\nrecipe: polaris-iceberg-spark-trino\nendpoint: http://10.21.227.93:80\naccess_key: ${S3_ACCESS_KEY}       # env var substitution\nsecret_key: ${S3_SECRET_KEY}\nscale: 50\nmode: batch\nspark_image: apache/spark:4.1.1-python3\n```\n\nEleven recipes are available -- see [Recipes](https://github.com/PureStorage-OpenConnect/lakebench-k8s/blob/main/docs/recipes.md)\nfor the full list.\n\nCompare two configurations side-by-side:\n\n```bash\nlakebench compare config-hive.yaml config-polaris.yaml\n```\n\nFor all recipes, see [`examples/`](examples/) or run `lakebench init --advanced`\nfor the full interactive wizard.\n\n## What You Get\n\nAfter `lakebench run` completes, the terminal prints a scorecard:\n\n```\n ─ Pipeline Complete ──────────────────────────────\n  bronze-verify         142.0 s\n  silver-build          891.0 s\n  gold-finalize         234.0 s\n  benchmark              87.0 s\n\n  Scores\n    Time to Value:        1354.0 s\n    Throughput:           0.782 GB/s\n    Efficiency:           3.41 GB/core-hr\n    Scale:                100.0% verified\n    QpH:                  2847.3\n\n  Full report: lakebench report\n ──────────────────────────────────────────────────\n```\n\n`lakebench report` generates an HTML report with per-query latencies,\nbottleneck analysis, and optional platform metrics (CPU, memory, S3 I/O per\npod).\n\n## How It Works\n\n```\n                    ┌──────────────────────────────────┐\n                    │         lakebench.yaml           │\n                    └────────────┬─────────────────────┘\n                                 │\n                    ┌────────────▼─────────────────────┐\n                    │   deploy (Kubernetes namespace,   │\n                    │   S3 secrets, PostgreSQL, catalog, │\n                    │   query engine, observability)     │\n                    └────────────┬─────────────────────┘\n                                 │\n     Raw Parquet ──► Bronze (validate) ──► Silver (enrich) ──► Gold (aggregate)\n         S3              Spark                Spark               Spark\n                                                                    │\n                                                        ┌───────────▼──────────┐\n                                                        │  8-query benchmark   │\n                                                        │  (Trino / DuckDB /   │\n                                                        │   Spark Thrift)      │\n                                                        └──────────────────────┘\n```\n\n## Prerequisites\n\n- `kubectl` and `helm` on PATH\n- Kubernetes 1.26+ (minimum 8 CPU / 32 GB RAM for scale 1)\n- S3-compatible object storage (FlashBlade, MinIO, AWS S3, etc.)\n- [Kubeflow Spark Operator 2.4.0+](https://github.com/kubeflow/spark-operator)\n  (or set `spark.operator.install: true`)\n- [Stackable Hive Operator](https://docs.stackable.tech/home/stable/hive/) for\n  Hive recipes (not needed for Polaris)\n\n## Commands\n\n| Command | Description |\n|---------|-------------|\n| `init` | Generate a starter config file |\n| `validate` | Check config and cluster connectivity |\n| `info` | Show deployment configuration summary |\n| `deploy` | Deploy all infrastructure components |\n| `generate` | Generate synthetic data at the configured scale |\n| `run` | Execute the medallion pipeline and benchmark |\n| `benchmark` | Run the 8-query benchmark standalone |\n| `query` | Execute ad-hoc SQL against the active engine |\n| `status` | Show deployment status |\n| `report` | Generate HTML scorecard report |\n| `recommend` | Recommend cluster sizing for a scale factor |\n| `destroy` | Tear down all deployed resources |\n\nSee [CLI Reference](https://github.com/PureStorage-OpenConnect/lakebench-k8s/blob/main/docs/cli-reference.md)\nfor flags and options.\n\n## Component Versions\n\n| Component | Version |\n|-----------|---------|\n| Apache Spark | 3.5.4, 4.0.2, 4.1.1 |\n| Spark Operator | 2.4.0 (Kubeflow) |\n| Apache Iceberg | 1.10.1 |\n| Delta Lake | 4.0.0 |\n| Hive Metastore | 3.1.3 (Stackable 25.7.0) |\n| Apache Polaris | 1.3.0-incubating |\n| Trino | 479 |\n| DuckDB | bundled (Python 3.11) |\n| PostgreSQL | 16, 17, 18 |\n\nAll versions are overridable in the YAML config. See\n[Supported Components](https://github.com/PureStorage-OpenConnect/lakebench-k8s/blob/main/docs/supported-components.md).\n\n## Documentation\n\n- [Getting Started](https://github.com/PureStorage-OpenConnect/lakebench-k8s/blob/main/docs/getting-started.md) -- prerequisites, install, first run\n- [Configuration](https://github.com/PureStorage-OpenConnect/lakebench-k8s/blob/main/docs/configuration.md) -- full YAML reference\n- [Recipes](https://github.com/PureStorage-OpenConnect/lakebench-k8s/blob/main/docs/recipes.md) -- catalog + format + engine combinations\n- [Compatibility Matrix](https://github.com/PureStorage-OpenConnect/lakebench-k8s/blob/main/docs/compatibility-matrix.md) -- Spark, Iceberg, and Delta version support\n- [Running Pipelines](https://github.com/PureStorage-OpenConnect/lakebench-k8s/blob/main/docs/running-pipelines.md) -- batch and sustained modes\n- [Benchmarking](https://github.com/PureStorage-OpenConnect/lakebench-k8s/blob/main/docs/benchmarking.md) -- scorecard and query benchmark\n- [Architecture](https://github.com/PureStorage-OpenConnect/lakebench-k8s/blob/main/docs/architecture.md) -- system design\n- [Troubleshooting](https://github.com/PureStorage-OpenConnect/lakebench-k8s/blob/main/docs/troubleshooting.md) -- common errors and fixes\n\n## License\n\nApache 2.0\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpurestorage-openconnect%2Flakebench-k8s","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpurestorage-openconnect%2Flakebench-k8s","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpurestorage-openconnect%2Flakebench-k8s/lists"}