User guide

• Overview
  • Parsl and Concurrency
  • Parsl and Execution
  • Parsl and Communication
    • Parameter Passing
    • File Passing
    • Futures
  • The Parsl Environment
    • Memory environment
    • File system environment
    • Service Environment
    • Environment Summary
• Apps
  • Python Apps
    • Special Keyword Arguments
    • Returns
    • Limitations
  • Bash Apps
    • Special Keywords
    • Returns
    • Limitations
• Futures
  • AppFutures
  • DataFutures
• Passing Python objects
• Staging data files
  • Parsl files
  • Staging providers
    • NoOpFileStaging for Local/Shared File Systems
    • FTP, HTTP, HTTPS: separate task staging
    • FTP, HTTP, HTTPS: in-task staging
    • Globus
      • Globus Configuration
      • Globus Authorization
    • rsync
      • rsync Configuration
      • rsync Authorization
• Execution
  • Execution providers
  • Executors
  • Launchers
  • Blocks
  • Elasticity
    • Parallelism
    • Configuration
  • Multi-executor
• Error handling
  • Exceptions
  • Retries
  • Lazy fail
  • Retry handlers
• Memoization and checkpointing
  • App caching
    • App equivalence
    • Invocation equivalence
    • Ignoring arguments
    • Caveats
  • Checkpointing
    • Creating a checkpoint
    • Resuming from a checkpoint
• Configuration
  • How to Configure
  • Heterogeneous Resources
  • Accelerators
  • Multi-Threaded Applications
  • Ad-Hoc Clusters
  • Amazon Web Services
  • ASPIRE 1 (NSCC)
  • Blue Waters (NCSA)
  • Illinois Campus Cluster (UIUC)
  • Bridges (PSC)
  • CC-IN2P3
  • CCL (Notre Dame, with Work Queue)
  • Comet (SDSC)
  • Cooley (ALCF)
  • Cori (NERSC)
  • Frontera (TACC)
  • Kubernetes Clusters
  • Midway (RCC, UChicago)
  • Open Science Grid
  • Polaris (ALCF)
  • Stampede2 (TACC)
  • Summit (ORNL)
  • Theta (ALCF)
  • TOSS3 (LLNL)
  • Further help
• Monitoring
  • Monitoring configuration
  • Visualization
    • Workflows Page
    • Workflow Summary
• Example parallel patterns
  • Bag of Tasks
  • Sequential workflows
  • Parallel workflows
  • Parallel workflows with loops
  • MapReduce
• Structuring Parsl programs
• Join Apps
  • Usage
  • What/why/how can you do with a join app
  • Terminology
• Performance and Scalability
  • Scalability
    • Strong scaling
    • Weak scaling
  • Throughput
  • Summary
• Usage statistics collection
  • Why are we doing this?
  • Opt-In
  • What is sent?
  • How is the data sent?
  • When is the data sent?
  • What will the data be used for?
  • Feedback
• Plugins
  • Executors
  • Providers, Launchers and Channels
  • File staging
  • Memoization/checkpointing