# Burla

## Burla

- [Home](https://docs.burla.dev/readme.md)
- [Getting Started](https://docs.burla.dev/get-started.md)
- [API Reference](https://docs.burla.dev/api-reference.md)
- [CLI Reference](https://docs.burla.dev/cli-reference.md)
- [About](https://docs.burla.dev/about.md)
- [You should never have to guess how much CPU or RAM you need.](https://docs.burla.dev/blog/dynamic-hardware.md): In fact, it's much better if you don't!
- [Vulnerability Disclosure Program](https://docs.burla.dev/security.md)
- [Managed Service Privacy Policy](https://docs.burla.dev/privacy-policy.md): Effective Date: 1/16/26
- [Software as a Service Agreement](https://docs.burla.dev/software-as-a-service-agreement.md): Effective Date: 1/16/26
- [Privacy Policy](https://docs.burla.dev/website-privacy-policy.md): Effective Date: 1/16/26
- [Query 2.4TBs of Parquet files in 76s](https://docs.burla.dev/featured-examples/process-2.4tb-of-parquet-files-in-76s.md): With <30 lines of Python.
- [Ranking 1.7M Airbnbs by TV location](https://docs.burla.dev/featured-examples/airbnb-burla.md)
- [Ranking 572M Amazon reviews](https://docs.burla.dev/featured-examples/amazon-review-distiller.md)
- [Clustering 2.7M arXiv abstracts](https://docs.burla.dev/featured-examples/arxiv-fossils.md)
- [Genomic alignment pipeline (Illumina)](https://docs.burla.dev/featured-examples/multi-stage-genomic-pipeline.md)
- [Start Here: Core Patterns](https://docs.burla.dev/all-examples/basic-examples.md)
- [Read/Write Files to Cloud Storage](https://docs.burla.dev/all-examples/basic-examples/read-and-write-gcs-files.md): Write and read files in GCS through /workspace/shared.
- [Use custom Docker images & GPUs](https://docs.burla.dev/all-examples/basic-examples/use-custom-docker-images-and-gpus.md): Run Burla workers with custom images, native tools, and GPUs.
- [Run jobs in the background](https://docs.burla.dev/all-examples/basic-examples/run-python-in-the-background.md): Detached job example.
- [Limit parallelism for APIs or databases](https://docs.burla.dev/all-examples/basic-examples/limit-parallelism-for-apis-databases-and-websites.md): Keep Burla jobs inside external service limits.
- [Combine many results/files into one](https://docs.burla.dev/all-examples/basic-examples/combine-many-results-files-into-one-map-reduce.md): A beginner-friendly map-reduce pattern for combining many outputs into one file.
- [Decide how to split your work](https://docs.burla.dev/all-examples/basic-examples/choose-how-to-split-your-work.md): Pick the input unit for a Burla job.
- [Process thousands of files](https://docs.burla.dev/all-examples/basic-examples/process-thousands-of-files-quickly.md)
- [Process one giant file](https://docs.burla.dev/all-examples/basic-examples/process-one-giant-file-quickly.md)
- [Process database rows](https://docs.burla.dev/all-examples/basic-examples/process-data-in-your-database-quickly.md)
- [ML, Embeddings & Search](https://docs.burla.dev/all-examples/ml-embeddings-and-search.md)
- [Embed 50K Wikipedia articles](https://docs.burla.dev/all-examples/ml-embeddings-and-search/gpu-embedding-demo.md)
- [Tune XGBoost on 1,000 CPUs](https://docs.burla.dev/all-examples/ml-embeddings-and-search/parallel-hyperparameter-tuning.md)
- [Run batch LLM inference](https://docs.burla.dev/all-examples/ml-embeddings-and-search/ml-inference-batch.md)
- [Search 192K artworks with CLIP](https://docs.burla.dev/all-examples/ml-embeddings-and-search/met-weirdest-art.md)
- [Large-Scale Data Processing](https://docs.burla.dev/all-examples/data-processing-examples.md)
- [Scan 2.76B NYC taxi trips](https://docs.burla.dev/all-examples/data-processing-examples/nyc-ghost-neighborhoods.md)
- [Map geotagged Flickr photos](https://docs.burla.dev/all-examples/data-processing-examples/world-photo-index.md)
- [Summarize 1M GitHub READMEs](https://docs.burla.dev/all-examples/data-processing-examples/github-repo-summarizer.md)
- [Audit 5,000 Parquet files](https://docs.burla.dev/all-examples/data-processing-examples/parquet-parallel.md)
- [Parallelize pandas apply](https://docs.burla.dev/all-examples/data-processing-examples/pandas-apply-parallel.md)
- [Production Data Workflows](https://docs.burla.dev/all-examples/production-data-jobs.md)
- [ETL 10K S3 files to Postgres](https://docs.burla.dev/all-examples/production-data-jobs/python-etl-no-airflow.md)
- [Resize an image corpus](https://docs.burla.dev/all-examples/production-data-jobs/image-dataset-resize.md)
- [Run a 2M-user API backfill](https://docs.burla.dev/all-examples/production-data-jobs/rate-limited-api-requests.md)
- [Scrape 1M web pages](https://docs.burla.dev/all-examples/production-data-jobs/parallel-web-scraping.md)
- [Run 1B option simulations](https://docs.burla.dev/all-examples/production-data-jobs/monte-carlo-simulation.md)
- [Scientific & Geospatial Computing](https://docs.burla.dev/all-examples/scientific-and-geospatial-work.md)
- [Align every FASTQ sample](https://docs.burla.dev/all-examples/scientific-and-geospatial-work/bioinformatics-alignment.md)
- [Find NOAA's rainiest day](https://docs.burla.dev/all-examples/scientific-and-geospatial-work/ghcn-rainiest-day.md)
- [NDVI for 2K Sentinel tiles](https://docs.burla.dev/all-examples/scientific-and-geospatial-work/gdal-raster-processing.md)