A Hierarchical Shared Memory Cluster Architecture with Load Balancing and Fault Tolerance

Abstract

Recently a great deal of attention has been paid to the design of hierarchical shared memory cluster system. Cluster computing has made hierarchical computing systems increasingly common as target environment for large-scale scientific computations. This paper proposes hierarchical shared memory cluster architecture with load balancing and fault tolerance. Hierarchies of shared memory and caches structure the architecture. The hierarchical load balancing approach focuses on reducing the redistribution cost. The fault tolerant model is adopted to build highly available clusters in hierarchical shared memory clusters. Performance analysis and results reveal that hierarchical shared memory clusters performs much better creating a reliable hierarchical network cluster system with high scalability.

A Comparative Analysis of Performance of Shared Memory Cluster Computing Interconnection Systems

“…In this development the hierarchical shared memory cluster system prevails due to its hardware simplicity and cost-effectiveness. Several shared memory cluster systems with large number of processors have been designed using hierarchy of processors in clusters [7,8]. The hierarchical shared memory cluster system efficiently supports the hierarchy of memories using a common memory.…”

“…The basic motivation is to compare the performance of shared memory cluster [4], distributed shared memory cluster [9], hierarchical shared memory cluster [7], and the virtual distributed shared memory cluster computing interconnection systems [14]. In this subsection, the above said four types of shared memory clusters are compared on three common platforms: (i) cluster architecture, (ii) load balancing, and (iii) fault tolerance.…”

“…Thus communication on the fly eliminates many data transactions and decreases the overall interprocessors and memory traffic. The hierarchical shared memory cluster computing interconnection system architecture (HSMC) uses hierarchical memory that reduces global and local intra-and intercluster communication [7]. The hierarchical shared memory cluster architecture is shown in Figure 2.…”

“…The architecture of the cluster systems should be flexible enough to handle any component or subcomponent that is required to make the system. The fault tolerance in the clusters provides error information in real time and makes it possible to detect which part or component needs repair and accordingly initiates a service call to identify the parts needed for maintenance [5][6][7][8][9][10][11].…”

“…With the increasing performance needs, clusters provide additional mechanism to take advantage of aggregate performance of the participating nodes through different types of cluster architectures. The various important classes of shared memory clusters studied in the literature are shared memory clusters [4], hierarchical shared memory clusters [7], distributed shared memory clusters [9], and virtual distributed shared memory clusters [14].…”