Learn More
We provide the first optimal algorithms in terms of the number of input/outputs (I/Os) required between internal memory and multiple secondary storage devices for the problems of sorting, FFT, matrix transposition, standard matrix multiplication, and related problems. Our two-level memory model is new and gives a realistic treatmentof parallel block(More)
Most le systems attempt to predict which disk blocks will be needed in the near future and prefetch them into memoryy this technique can improve a p-plication throughput as much as 50%. But why? The reasons include that the disk cache comes into play, the device driver amortizes the xed cost of an I/O operation over a larger amount of data, total disk seek(More)
In this paper we introduce parallel versions of two hierarchical memory models and give optimal algorithms in these models for sorting, FFT, and matrix multiplication. In our parallel models, there areP memory hierarchies operating simultaneously; communication among the hierarchies takes place at a base memory level. Our optimal sorting algorithm is(More)
Today, c a c hing web proxies use general-purpose le systems to store web objects. Proxies, e.g., Squid or Apache, when running on a UNIX system, typically use the standard UNIX le system (UFS) for this purpose. UFS was designed for research and engineering environments, which h a ve diierent c harac-teristics from that of a caching web proxy. Some of the(More)
The viability of storage outsourcing is critically dependent on the access performance of remote storage. We study this issue by measuring the behavior of a broad variety of I/O-intensive benchmarks as they access remote storage over an IP network. We measure the effect of network latencies that correspond to distances ranging from a local neighborhood to(More)
We sugg=t a practical and economical way to generate random bits using a computer disk drive * a source of ran-domn-. It requirw no additiond hardware (given a system with a disk), and no user involvement. As a concrete example of performance, on a Sun Wtra-1 with a Seagate Cheetah disk, it generatw bits at a rate of either 5 bits per minute or 577 bits per(More)
Modern disk drives read-ahead data and reorder incoming requests in a workload-dependent fashion. This improves their performance, but makes simple analytical models of them inadequate for performance prediction, capacity planning, workload balancing, and so on. To address this problem we have developed a new analytic model for disk drives that do readahead(More)
We provide optimal algorithms for sorting , FFT, matrix transposition, standard matrix multiplication , and related problems in terms of the number of input/outputs (I/Os) required between internal memory and secondary storage. Our two-level memory model is new and gives a realistic treatment of parallel block transfer, in which during a single I/O each of(More)
We provide the rst optimal algorithms in terms of the number of inputtoutputs IIOs required between internal memory and multiple secondary storage devices for the problems of sorting, FFT, matrix transposition, standard matrix multiplication, and related problems. Our two-level memory model is new and gives a realistic treatment o f p arallel block(More)