كليدواژه :
Scheduling algorithm , Distributed system , Sand pile algorithm , Load balancing
چكيده فارسي :
Scientific computing requires huge number of resources such as CPU, Network and Datacenters to deliver results, and this request is being added, and needs have high performance or has high throughput capabilities. Parallelism, elasticity, load balancing, utilazation are important factors, those have influence on HPC[5]. Scientific computing includes huge number of bags, that are intered to distributed systems then we called it Bag of tasks . Scheduling is the most important factor for performance improvement. Load balancing, Utilization, must be considerate in scheduling algorithm. In this research expect Hierarchical Sandpile scheduling algorithm is good self-improvement strategy to improve performance [3]