Ricardo Bianchini. Alewife Systems Memo #43. Application Performance on the Alewife Multiprocessor, September 1994. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
QuickStep: A System for Performance Monitoring and Debugging.. - Mitra (1995) (Correct)
....system. The main goal of the system is to aid in identifying performance bottlenecks and tuning programs to reduce the effect of such bottlenecks. Furthermore, it can also be used to analyze effects of optimization of application code. We use MP3D and SOR two large parallel applications [3] to demonstrate appilcations of Quickstep. 5.1 Case Study 1: MP3D In this chapter, we use MP3D an application from the SPLASH suite to demonstrate how QuickStep provides useful insight into program behavior as a result of optimization. Mp3d simulates the interactions between particles ....
Ricardo Bianchini. Alewife Systems Memo #43. Application Performance on the Alewife Multiprocessor, September 1994.
QuickStep: A System for Performance Monitoring and Debugging.. - Mitra (1995) (Correct)
....system. The main goal of the system is to aid in identifying performance bottlenecks and tuning programs to reduce the effect of such bottlenecks. Furthermore, it can also be used to analyze effects of optimization of application code. We use MP3D and SOR two large parallel applications [3] to demonstrate appilcations of Quickstep. 5.1 Case Study 1: MP3D 5.1.1 Description In this chapter, we use MP3D an application from the SPLASH suite to demonstrate how QuickStep provides useful insight into program behavior as a result of optimization. Mp3d simulates the interactions ....
Ricardo Bianchini. Alewife Systems Memo #43. Application Performance on the Alewife Multiprocessor, September 1994.
Processor Mechanisms for Software Shared Memory - Carter (Correct)
....transform (FFT) and a multigrid solver. The FFT program was chosen because of its importance in signal processing applications, while the multigrid solver is typical of scientific applications. The C language source code for these applications was provided by the Alewife group at M.I.T. [4], and compiled on the mmcc compiler for the M Machine. Porting these applications to the M Machine required that a number of assembly language routines be written to perform tasks such as forking parallel processes, barriers, and fetch and add. These routines were then linked in with the compiled ....
Ricardo Bianchini. Application performance on the Alewife multiprocessor. Alewife Systems Memo #43, Laboratory for Computer Science, Massachusetts Institute of Technology, 1994.
The MIT Alewife Machine: Architecture and Performance - Agarwal, Bianchini, Chaiken, .. (1995) (173 citations) Self-citation (Bianchini) (Correct)
....subsection compares the performance of an application written in a message passing style on Alewife and on the Thinking Machines CM 5 multiprocessor. The last two subsections present details of two application case studies, MP3D and MICCG3D. For a complete description of the results, refer to [10]. 4.1 Shared memory performance Shared memory applications perform well on Alewife, proving the viability of both the software extended coherence mechanism and the flat communication substrate provided by the mesh network. Table 7 summarizes the main characteristics of the applications evaluated ....
Ricardo Bianchini. Application Performance on the Alewife Multiprocessor. ALEWIFE Memo No. 43, Laboratory for Computer Science, Massachusetts Institute of Technology, August 1994.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC