New Challenges of Parallel Job Scheduling
The workshop on job scheduling strategies for parallel processing (JSSPP) studies the myriad aspects of managing resources on parallel and distributed computers. These studies typically focus on large-scale computing environments, where allocation and man
- PDF / 290,663 Bytes
- 23 Pages / 430 x 660 pts Page_size
- 82 Downloads / 159 Views
Abstract. The workshop on job scheduling strategies for parallel processing (JSSPP) studies the myriad aspects of managing resources on parallel and distributed computers. These studies typically focus on large-scale computing environments, where allocation and management of computing resources present numerous challenges. Traditionally, such systems consisted of massively parallel supercomputers, or more recently, large clusters of commodity processor nodes. These systems are characterized by architectures that are largely homogeneous and workloads that are dominated by both computation and communication-intensive applications. Indeed, the large majority of the articles in the first ten JSSPP workshops dealt with such systems and addressed issues such as queuing systems and supercomputer workloads. In this paper, we discuss some of the recent developments in parallel computing technologies that depart from this traditional domain of problems. In particular, we identify several recent and influential technologies that could have a significant impact on the future of research on parallel scheduling. We discuss some of the more specific research challenges that these technologies introduce to the JSSPP community, and propose to enhance the scope of future JSSPP workshops to include these topics.
1
Introduction
The last few years have brought about many radical changes in the technologies and platforms that exhibit the same salient challenges that JSSPP focuses on—all in the family of flexible allocation and management of multiple computer resources. These technologies, however, depart from the traditional supercomputer model of relatively homogeneous architectures and applications, and add new dimensions to those that are already being studied within JSSPP’s scope. This paper therefore has two main goals: (1) To present some of the major technological changes and to discuss the additional dimensions they add to the set of JSSPP challenges; and, (2) to promote and suggest research topics inspired by these dimensions in the JSSPP community. Such dimensions include, for example, reliability and resource allocation across multiple sites (Grids), workloads that are a mixture of parallel, sequential, and E. Frachtenberg and U. Schwiegelshohn (Eds.): JSSPP 2007, LNCS 4942, pp. 1–23, 2008. c Springer-Verlag Berlin Heidelberg 2008
2
E. Frachtenberg and U. Schwiegelshohn
interactive applications on multi-core desktops, and data-intensive applications on Web servers that exhibit little or no communication. Although the traditional topics of interest of JSSPP are still relevant and will likely continue to attract high quality papers to the workshop, we feel the need to introduce these new topics to JSSPP’s scope at this time, for two main reasons: 1. The field of parallel job scheduling, while still evolving, is showing signs of maturity. 2. New technologies are exhibiting many characteristics of the problems that the JSSPP community tackles. We believe that the JSSPP community’s expertise can produce meaningful contributions
Data Loading...
