It is often thought that the terms management and scheduling are synomnous. This is not the case. Scheduling is in fact just one out of five important aspects of cluster workload management. This paper discusses the role of scheduling within the wider area of workload management, and illustrates how scheduling is interwoven with the other components. In addition, we show how to apply these principles to real world challenges of building and managing clusters.

We use the term "Grid" to refer to a software system that provides uniform and location independent access to geographically and organizationally dispersed, heterogeneous resources that are persistent and supported. While, in general, Grids will provide the infrastructure to support a wide range of services in the scientific environment (e.g. collaboration and remote instrument control) in this paper we focus on services for high performance computing and data handling. We describe the services and architecture of NASA’s Information Power Grid ("IPG") - an early example of a large-scale Grid - and some of the issues that have come up in its implementation.

Information Power Grid (IPG) is the name of NASA's project to build a fully distributed computing and data management environment-a Grid. The IPG project has near, medium, and long-term goals that represent a continuum of engineering, development, and research topics. The overall goal is to provide the NASA scientific and engineering communities a substantial increase in their ability to solve problems that depend on use of large-scale and/or dispersed resources: aggregated computing, diverse data archives, laboratory instruments and engineering test facilities, and human collaborators. The approach involves infrastructure and services than can locate, aggregate, integrate, and manage resources from across the NASA enterprise. An important aspect of IPG is to produce a common view of these resources, and at the same time provide for distributed management and local control. In addition to addressing the overall goal of enhanced science and engineering, there is a potential important side effect. With a large collection of resources that have common use interfaces and a common management approach, the potential exists for a considerable pool of computing capability that could relatively easily, for example, be called on in extraordinary situations such as crisis response.

"Parallel I/O" is the support of a single parallel application run on many nodes; application data is distributed among the nodes, and is read or written to a single logical file, itself spread across nodes and disks. Parallel I/O is a mapping problem from the data layout in node memory to the file layout on disks. Since the mapping can be quite complicated and involve significant data movement, optimizing the mapping is critical for performance. We discuss our general model of the problem, describe four Collective Buffering algorithms we designed, and report experiments testing their performance on an Intel Paragon and an IBM SP2 both housed at NASA Ames Research Center. Our experiments show improvements of up to two order of magnitude over standard techniques and the potential to deliver peak performance with minimal hardware support.

Current processor allocation techniques for highly parallel systems are typically restricted to contiguous allocation strategies for which performance suffers significantly due to the inherent problem of fragmentation. As a result, message-passing systems have yet to achieve the high utilization levels exhibited by traditional vector supercomputers. We are investigating processor allocation algorithms which lift the restriction on contiguity of processors in order to address the problem of fragmentation. Three noncontiguous processor allocation strategies-paging allocation, random allocation, and the Multiple Buddy Strategy (MBS)-are proposed and studied in this paper. Simulations compare the performance of the noncontiguous strategies with that of several well-known contiguous algorithms. We show that noncontiguous allocation algorithms perform better overall than the contiguous ones, even when message-passing contention is considered. We also present the results of experiments on an Intel Paragon XP/S-15 with 208 nodes that show noncontiguous allocation is feasible with current technologies.

Current processor allocation techniques for highly parallel systems have thus far been restricted to contiguous allocation strategies for which performance suffers significantly due to the inherent problem of fragmentation. We are investigating processor allocation algorithms which lift the restriction on contiguity of processors in order to address the problem of fragmentation. Three non-contiguous processor allocation strategies: naive, random and the multiple buddy strategy (MBS) are proposed and studied in this paper. Simulations compare the performance of the non-contiguous strategies with that of several well-known contiguous algorithms. We show that non-contiguous allocation algorithms perform better overall than the contiguous ones, even when message-passing contention is considered. We also present the results of experiments on an Intel Paragon XP/S-15 with 208 nodes that show non-contiguous allocation is feasible with current technologies.>