Abstract
High performance clusters are being configured specially to give data centers that require extreme performance, the processing power they need. In Cluster Computing Environment the data latency time has significant impact on the performance, when the data is accessed across clusters. Instead of processing power, memory and I/O have become the new bottleneck in achieving efficient load balance at higher performance for cluster computer systems. Initial job placement and load balancing are the key aspects affecting the performance. In this research, data access patterns, memory and CPU utilization and locality of memory are combined to consider as load metric in the load balancing aspect across cluster. A scheduling algorithm based on this metric has been proposed to dynamically balance the load in the cluster. Initial job placement for a job in the cluster considers data access patterns and for load balance aspect metric constitutes CPU, memory utilization including locality of memory. Experimental results show performance improvement to considerable levels with the implementation of the concept, specifically when the cost of data access from other clusters is higher and is proportionate to the amount of data.
Highlights
A cluster computer is a collection of computers interconnected with a High-speed network technology
The proposed initial job placement algorithm is based on data access patterns and load balancing algorithm is based on CPU utilization, CPU queue length, memory utilization
For the simulation results,we evaluate the performance of the proposed dynamic load balancing system with the with proposed new load metric
Summary
A cluster computer is a collection of computers interconnected with a High-speed network technology. Dynamic load balancing systems can be classified into initial job placement and process migration. Initial job placement improves the resource utilization of the entire system by distributing the workload on to several nodes. We can expect further improvement in performance if the initial job placement system enhances resource utilization system-wide, and in terms of each node[4]. Initial job placement in the cluster considers data access patterns to designate a node for a job. For this purpose, we have developed a new algorithm and a new load metric which contains information about both the system load and resource utilization. A dynamic load balancing algorithm is designed and implemented using the load metric and its performance is evaluated
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
