Dynamic Load-Balancing in iSCSI Systems Based on a Feedback Control Mechanism

Multiple network cards and multiple connections in iSCSI are important ways to improve iSCSI systems performance, fault tolerance and scalability. But multiple connections technology cannot assign every task appropriately, so it cannot produce the excellent performance. This paper proposes a dynamic load balancing algorithm using a feedback control mechanism to improve the overall performance in iSCSI systems. In order to adjust the load state dynamically, this algorithm adopts a feedback mechanism to calculate multiple resources states that have been collected in this system, and then combines the results and the previous feedback. The test results indicate that with the help of this algorithm the performance of iSCSI system has been improved.

[1]  Marianne Winslett,et al.  Faster collective output through active buffering , 2002, Proceedings 16th International Parallel and Distributed Processing Symposium.

[2]  Samuel T. Chanson,et al.  Improved Strategies for Dynamic Load Balancing , 1999, IEEE Concurr..

[3]  Xiao Qin,et al.  A dynamic load balancing scheme for I/O-intensive applications in distributed systems , 2003, 2003 International Conference on Parallel Processing Workshops, 2003. Proceedings..

[4]  Jan H. M. Korst,et al.  Random duplicate storage strategies for load balancing in multimedia servers , 2000, Inf. Process. Lett..

[5]  Sanjeev Setia,et al.  Availability and utility of idle memory in workstation clusters , 1999, SIGMETRICS '99.

[6]  Sanda M. Harabagiu,et al.  Performance analysis of a distributed question/answering system , 2001, Proceedings 15th International Parallel and Distributed Processing Symposium. IPDPS 2001.

[7]  Li Xiao,et al.  Improving distributed workload performance by sharing both CPU and memory resources , 2000, Proceedings 20th IEEE International Conference on Distributed Computing Systems.

[8]  Hui Xiong,et al.  An iSCSI Design and Implementation , 2004, MSST.

[9]  Li Xiao,et al.  Dynamic load sharing with unknown memory demands in clusters , 2001, Proceedings 21st International Conference on Distributed Computing Systems.

[10]  Remzi H. Arpaci-Dusseau,et al.  Storage-Aware Caching: Revisiting Caching for Heterogeneous Storage Systems , 2002, FAST.

[11]  Xiao Qin,et al.  Dynamic Load Balancing for I/O- and Memory-Intensive Workload in clusters Using a Feedback Control Mechanism , 2003, Euro-Par.