There are many scientific applications for which the computational load varies throughout the execution and causes uneven distribution of workload during run-time. One such class of applications is Adaptive Mesh Refinement (AMR) applications. AMR is a type of multiscale algorithm that achieves high resolution in localized regions of dynamic, multidimensional numerical simulations. A typical AMR application may require enormous computing resources, which usually cannot be satisfied by a single-processor machine, thereby requiring parallel and distributed systems. One of the key issues related to AMR is dynamic load balancing (DLB), which allows large-scale adaptive applications to run efficiently on parallel and distributed systems. In investigating DLB schemes, we first complete a detailed analysis of structured AMR (SAMR) applications, identifying the unique characteristics that impose severe challenges on DLB schemes. The results indicate that most of the available DLB schemes are not appropriate for SAMR applications due to their unique adaptive characteristics. Thus, we propose a novel dynamic load balancing scheme for SAMR applications on parallel systems (denoted as parallel DLB). It integrates a grid-splitting technique with direct grid movements, for which the objective is to reduce the parallel execution time. Further, our experiment shows that simply moving a DLB scheme designed for parallel systems to distributed systems will introduce significant overhead. Therefore, we propose a framework for dynamic load balancing on distributed systems (denoted as distributed DLB). It takes into consideration: (1) heterogeneity of processors, (2) heterogeneity of networks, (3) shared nature of networks, and (4) adaptive characteristics of the applications. For SAMR applications, the distributed DLB incorporates the proposed parallel DLB during the load balancing process. Both parallel DLB and distributed DLB were implemented in the ENZO code, a parallel implementation of SAMR in astrophysics and cosmology. Experiments show that the proposed DLB schemes can significantly improve the performance of SAMR applications on both parallel and distributed systems in terms of the total execution time and the quality of load balancing.

This book focuses on the future directions of the static scheduling and dynamic load balancing methods in parallel and distributed systems. It provides an overview and a detailed discussion of a wide range of topics from theoretical background to practical, state-of-the-art scheduling and load balancing techniques.

A distributed system consists of many heterogeneous processors with different processing power and all processors are interconnected with a communication channel. In such a system, if some processors are less loaded or idle and others are heavily loaded, the system performance will be reduced drastically. System performance can be improved by using proper load balancing [1, 4]. The aim of load balancing is to improve the performance measures and reduce the overall completion time and cost

This book illustrates various components of Distributed Computing Environment and the importance of distributed scheduling using Dynamic Load Balancing. It describes load balancing algorithms for better resource utilization, increasing throughput and improving user’s response time. Various theoretical concepts, experiments, and examples enable students to understand the process of load balancing in computing cluster and server cluster. The book is suitable for students of Advance Operating Systems, High Performance Computing, Distributed Computing in B.E., M.C.A., M. Tech. and Ph.D courses.

Load Balancing in Parallel Computers: Theory and Practice is about the essential software technique of load balancing in distributed memory message-passing parallel computers, also called multicomputers. Each processor has its own address space and has to communicate with other processors by message passing. In general, a direct, point-to-point interconnection network is used for the communications. Many commercial parallel computers are of this class, including the Intel Paragon, the Thinking Machine CM-5, and the IBM SP2. Load Balancing in Parallel Computers: Theory and Practice presents a comprehensive treatment of the subject using rigorous mathematical analyses and practical implementations. The focus is on nearest-neighbor load balancing methods in which every processor at every step is restricted to balancing its workload with its direct neighbours only. Nearest-neighbor methods are iterative in nature because a global balanced state can be reached through processors' successive local operations. Since nearest-neighbor methods have a relatively relaxed requirement for the spread of local load information across the system, they are flexible in terms of allowing one to control the balancing quality, effective for preserving communication locality, and can be easily scaled in parallel computers with a direct communication network. Load Balancing in Parallel Computers: Theory and Practice serves as an excellent reference source and may be used as a text for advanced courses on the subject.