VGE

VGE executes bioinformatics software pipelines on massively parallel supercomputers by distributing tasks across processors using a Message Passing Interface (MPI)–based master–worker architecture.

Key Features:

• MPI-Based Task Distribution: Executes bioinformatics workflows as Message Passing Interface (MPI) programs to enable parallel processing across large numbers of processors or cores.
• Master–Worker Execution Model: Uses a master process to reserve computational resources and assign subdivided tasks to worker processes.
• Large-Scale Job Management: Manages thousands of parallel jobs across massively parallel computing nodes.
• Task Dependency Control: Implements dependency management to coordinate execution order among workflow tasks.
• Non-Privileged Server Operation: Operates as a server program without requiring elevated system privileges.

Scientific Applications:

• Large-Scale Bioinformatics Workflows: Executes high-throughput computational pipelines on massively parallel supercomputers.
• High-Performance Genomic Data Processing: Supports distributed analysis tasks such as FASTQ data splitting and sequence alignment with BWA.
• Parallel Scientific Computing: Enables large-scale distributed computation for data-intensive bioinformatics analyses.

Methodology:

VGE converts bioinformatics pipelines into MPI-based programs, reserves computational resources, distributes divided tasks from a master process to worker processes, and manages task dependencies during parallel execution.