Tibanna

Tibanna orchestrates execution of bioinformatics pipelines on Amazon Web Services (AWS) to provide scalable, reproducible cloud-based processing of CWL- and WDL-defined workflows packaged with Docker for large-scale data analysis.

Key Features:

• Workflow standard support: Supports Common Workflow Language (CWL), Workflow Description Language (WDL), and Docker-packaged workflows.
• Serverless scheduling: Implements a serverless scheduling framework that isolates and optimizes individual executions to manage computational resources per job.
• Execution and monitoring interfaces: Executes and monitors pipelines via local command-line commands or a Python API.
• Automatic cloud configuration: Automatically handles AWS cloud configuration for pipeline deployment and execution.
• Scalability: Demonstrates scalability in high-throughput environments, capable of processing terabyte-scale datasets.

Scientific Applications:

• Large-scale genomics data processing: Processing terabyte-scale datasets in projects such as the 4D Nucleome (4DN) Network.
• Reproducible cloud execution of workflows: Cloud-based execution of reproducible CWL- and WDL-defined bioinformatics workflows packaged with Docker.

Methodology:

Uses a serverless scheduling framework with per-execution isolation and optimization, supports CWL, WDL and Docker workflows, executes and monitors runs via local commands or a Python API, and performs automatic AWS cloud configuration.