Application Runtime Research Articles

Designing computer systems with MEMS-based storage

For decades the RAM-to-disk memory hierarchy gap has plagued computer architects. An exciting new storage technology based on microelectromechanical systems (MEMS) is poised to fill a large portion of this performance gap, significantly reduce system power consumption, and enable many new applications. This paper explores the system-level implications of integrating MEMS-based storage into the memory hierarchy. Results show that standalone MEMS-based storage reduces I/O stall times by 4--74X over disks and improves overall application runtimes by 1.9--4.4X. When used as on-board caches for disks, MEMS-based storage improves I/O response time by up to 3.5X. Further, the energy consumption of MEMS-based storage is 10--54X less than that of state-of-the-art low-power disk drives. The combination of the high-level physical characteristics of MEMS-based stor-age (small footprints, high shock tolerance) and the ability to directly integrate MEMS-based storage with processing leads to such new applications as portable gigabit storage systems and ubiquitous active storage nodes.

Message-passing environments for metacomputing

In this paper, we present the three libraries PACX-MPI, PLUS, and PVMPI that provide message-passing between different high-performance computers in metacomputing environments. Each library supports the development and execution of distributed metacomputer applications. The PACX-MPI approach offers a transparent interface for the communication between two or more MPI environments. PVAMPI allows the user spawning parallel processes under the MPI environment. The PLUS protocol bridges the gap between vendor-specific (e.g., MPL, NX, and PARIX) and vendor-independent message-passing environments (e.g., PVM and MPI). Moreover, it offers the ability to create and control processes at application runtime.

Parallel FFT on ATM-based networks of workstations

In this paper, we first evaluate the performance degradation caused by unequal bandwidths on the execution of conventional parallel algorithms such as the fast Fourier transform on an ATM-based Network of Workstations. We then present a strategy based on dynamic redistribution of data points to reduce the bottlenecks caused by unequal bandwidths. We also extend this strategy to deal with processor heterogeneity. Using analysis and simulation we show that there is a considerable reduction in the runtime if the proposed redistribution strategy is adopted. The basic idea presented in this paper can also be used to improve the runtimes of parallel applications in connection-oriented environments.

An effective garbage collection strategy for parallel programming languages on large scale distributed-memory machines

This paper describes the design and implementation of a garbage collection scheme on large-scale distributed-memory computers and reports various experimental results. The collector is based on the conservative GC library by Boehm & Weiser. Each processor traces local pointers using the GC library while traversing remote pointers by exchanging "mark messages" between processors. It exhibits a promising performance---in the most space-intensive settings we tested, the total collection overhead ranges from 5% up to 15% of the application running time (excluding idle time). We not only examine basic performance figures such as the total overhead or latency of a global collection, but also demonstrate how local collection scheduling strategies affect application performance. In our collector, a local collection is scheduled either independently or synchronously . Experimental results show that the benefit of independent local collections has been overstated in the literature. Independent local collections slowed down application performance to 40%, by increasing the average communication latency. Synchronized local collections exhibit much more robust performance characteristics than independent local collections and the overhead for global synchronization is not significant. Furthermore, we show that an adaptive collection scheduler can select the appropriate local collection strategy based on the application's behavior. The collector has been used in a concurrent object-oriented language ABCL/ f and the performance is measured on a large-scale parallel computer (256 processors) using four non-trivial applications written in ABCL/ f .

A study of integrated prefetching and caching strategies

Prefetching and caching are effective techniques for improving the performance of file systems, but they have not been studied in an integrated fashion. This paper proposes four properties that optimal integrated strategies for prefetching and caching must satisfy, and then presents and studies two such integrated strategies, called aggressive and conservative. We prove that the performance of the conservative approach is within a factor of two of optimal and that the performance of the aggressive strategy is a factor significantly less than twice that of the optimal case. We have evaluated these two approaches by trace-driven simulation with a collection of file access traces. Our results show that the two integrated prefetching and caching strategies are indeed close to optimal and that these strategies can reduce the running time of applications by up to 50%.

Using versioned object-oriented data in programming languages

The problem of accessing object versions from object-oriented programming language is discussed. Different approaches to consistency checking are discussed and compared. Some extensions to object-oriented languages are proposed to make it possible to version objects both at the compilation time and the application run-time.

LIMES: a hypertext interface to a cholesterol management expert system

Hypermedia representations add superior explanation facilities and user interfaces to expert systems. LIMES (LIpid Management Expert System) is a hypermedia cholesterol management application built on the HyperLipid Advisory System created on the Macintosh II at Stanford University. HyperLipid is an expert system which was designed to administer the National Cholesterol Education Program clinical algorithm for detecting and treating hyperlipidemia. LIMES executes the HyperLipid knowledgebase within a HyperCard stack which embeds the inference engine of the expert system. The HyperCard module serves initially as a programming tool to communicate with the rule development system and ultimately as the interface to the runtime application. Predictors of the utility of this and other systems are the expertise differential between the system and end user, and the degree of risk inherent in the domain.