Getting started
Introduction
Archil Documentation
Archil provides infinite, shareable disks for servers in the AWS and GCP cloud that radically simplify how teams
work with large data sets, at up to 90% lower cost. Archil disks are ideal for providing instant access to massive,
unstructured data sets stored in services like Amazon S3, GCS, or Cloudflare R2.
With an Archil disk, all of your data is available on all of your servers, instantly.
CLI Reference
Learn about Archil’s command line interface
Architecture
Learn about how Archil disks are built for resiliency and performance
Common Use Cases
Stateful AI AgentsEnable AI agents to maintain state and work with large datasets by providing fast, shared access to training data and model artifacts stored in S3.
RAG with Unstructured DataBuild retrieval augmented generation systems that can efficiently process and retrieve from large unstructured datasets like PDFs, images, and audio files.
Stateless Jupyter NotebooksRun Jupyter notebooks without local state by storing notebooks and data in S3, while maintaining interactive performance through Archil's caching.
Web HostingHost content management systems like WordPress with automatic synchronization to S3, enabling scalable and resilient web applications.
Satellite Image ProcessingProcess large satellite imagery datasets with high-performance access to object storage, enabling efficient parallel processing across multiple servers.
GenomicsAnalyze genomic data with fast access to reference genomes and sequencing data stored in object storage, supporting parallel processing pipelines.
Financial ServicesProcess market data and financial records with POSIX-compliant storage that maintains an audit trail in object storage while providing high-performance access.