This paper analyzes and evaluates some novel latency hiding features of the KSR1 multiprocessor: prefetch and poststore instructions and automatic updates. As a case study, we analyze the performance of an iterative sparse solver which generates irregular communications. We show that automatic updates signiicantly reduce the amount of communication. Although prefetch and poststore instructions ...

Data-management applications are focused around so-called CRUD actions that create, read, update, and delete data from persistent storage. These operations are the building blocks for numerous applications, for example dynamic websites where users create accounts, store and update information, and receive customized views based on their stored data. Typically, the application’s data is required...

We present an optimal on-line algorithm for the List Update Problem when the request sequence has some particular structure.

Recent research 6] has suggested that the branch history register need not contain the outcomes of the most recent branches in order for the Two-Level Adaptive Branch Predictor to work well. From this result, it is tempting to conclude that the branch history register need not be speculatively updated. This paper revis-its this work and explains when the most recent branch outcomes can be omitt...

We report results on the π-γ transition form factor obtained within the hard scattering approach including transverse momentum effects and Sudakov corrections. The results clearly favor distribution amplitudes close to the asymp-totic form, ∼ x 1 x 2 , and disfavor distribution amplitudes which are strongly concentrated in the end-point regions. This observation is backed by information on the ...

Optimum oo-line algorithms for the list update problem are investigated. The list update problem involves implementing a dictionary of items as a linear list. Several characterizations of optimum algorithms are given; these lead to optimum algorithm which runs in time 2 n (n ? 1)!m, where n is the length of the list and m is the number of requests. The previous best algorithm, an adaptation of ...

When establishing and analyzing model-parameter dependent mixed formulations, it is common to consider required ellipticity and inf-sup conditions for the continuous and discrete problems. However, in the modeling of some important categories of problems, like in the analysis of plates and shells, the ellipticity condition usually considered does not naturally hold, and the inf-sup condition ca...

WEI-MIN WU, JACK CARLEY, JIAN LUO, MATTHEW A GINDER-VOGEL, ERICK 4 CARDENAS, MARY BETH LEIGH, CHAICHI HWANG, SHELLY D. KELLY, CHAUNGMIN 5 RUAN, LIYOU WU, JOY VAN NOSTRAND, TERRY GENTRY , KENNETH LOWE, TONIA 6 MEHLHORN, SUE CARROLL, WENSUI LOU, MATTHEW W. FIELDS, BAOHUA GU, 7 DAVID WATSON, KENNETH M. KEMNER, TERENCE MARSH, JAMES TIEDJE, JIZHONG 8 ZHOU, SCOTT FENDORF, PETER K. KITANIDIS, PHILIP M...

This paper describes the initial results of our research within the context of Serenoa project, where we have developed the first version of an authoring tool, which has been developed as a plugin for the widely used Eclipse IDE in the industry and academia. This authoring tool supports all the CRUD (Create, Retrieve, Update, and Delete) operations for the adaptation rules and front-end UIs dev...

We derive an algorithm in the spirit of Rogers and Davis, Burstein that leads to upper bounds for stochastic control problems. Our bounds complement lower biased estimates recently obtained in Guyon, Henry-Labordere. We evaluate our estimates in two numerical examples motivated from mathematical finance. Joint work with Christian Litterer and Zhenjie Ren. ∗Speaker †Corresponding author: pierre....