A high-throughput adaptive computing infrastructure for bioinformatics research

To support our company´s bioinformatics research and product development we developed the Celera computing analysis pipeline infrastructure (CCAPI). The N-tiered architecture, which includes a Web based user interface, an Oracle database, and a network of hundreds of CPUs, allows us to perform complex and compute-intensive data processing and analysis on genomic content in a cost-effective way. Developed using Perl, Java/JSP, Oracle´s AQ and Platform LSF, CCAPI creates an adaptive and scalable computing framework that supports task decomposition and parallelization, optimizes computing resource allocation, facilitates the integration of proprietary and third-party applications, and supports automatic error detection and recovery.