ACO Algorithm-Based Parallel Job Scheduling Investigation on Hadoop

Provided by: AICIT
Topic: Cloud
Format: PDF
The pursuit of Hadoop by researchers is based on its architecture advantages; however, its job scheduling shortcoming and workload unbalance are the key bottlenecks of its built-in FIFO algorithm for large amounts of small granularity jobs under cloud computing situation. After analyzing FIFO's advantages and disadvantages, this paper innovatively proposes that ACO (Ant Colony Optimization) can improve job scheduling performance dynamically with cost matrix, hormone matrix and probability matrix, and then simulates large amounts of jobs scheduling on cloud computing scene.

Find By Topic