D. G. Feitelson and M. A. Volovic, "ParPar design document version 0.2 ". URL http://www.cs.huji.ac.il/labs/parallel/parpar.html, Jun 1997. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Exception Propagation in the ParPar System - Dror Feitelson Institute Self-citation (Feitelson) (Correct)
....Unix signals. When interaction among processes on different nodes is required, a metasignal is sent between the relevant daemons. Such metasignals are used to convey information about local signals that have been sent or should be sent. 2 The ParPar System In a nutshell, ParPar is a software MPP [4]. It is a combination of off the shelf hardware (PCs and a fast LAN) with our software, in order to emulate an MPP efficiently and inexpensively. It is not just a network of workstations [2] rather, the system is dedicated to serving parallel jobs, at the possible expense of full Unix ....
D. G. Feitelson and M. A. Volovic, "ParPar design document version 0.2 ". URL http://www.cs.huji.ac.il/labs/parallel/parpar.html, Jun 1997.
Job Scheduling in Multiprogrammed Parallel Systems - Feitelson (1997) (16 citations) Self-citation (Feitelson) (Correct)
....specify what to do in each situation. In the ParPar system, process failure is automatically propagated to other processes in the form of a SIGUSR1 signal. This allows fault tolerant applications to catch the signal and re organize the computation, whereas naive applications are terminated cleanly [207]. 3.2.3 Making Scheduling Decisions When partitioning is used without preemption, it may be the case that submitted jobs have to wait until sufficient PEs become available for them to run. The system is then faced with the question of the order in which the queued jobs should be executed. The ....
D. G. Feitelson and M. A. Volovic, "ParPar design document version 0.2 ". URL http://www.huji.ac.il/labs/parallel/parpar.html, Jun 1997.
Design, Implementation, and Impact of Multicast in the.. - Er-El, Kavas, Feitelson Self-citation (Feitelson) (Correct)
....capability of the underlying Ethernet medium. We therefore implemented a reliable multicast protocol based on UDP IP. In this paper we only describe the parts of the system that are relevant to the multicast facility. A detailed discussion of the whole system is available in the design document [7]. 3 The RDGM Library Most of the master daemon s activities involve the multicasting of messages to groups of node daemons. In the original implementation, the multicast was implemented using a loop of TCP messages. While TCP provides reliable communication, it is a rather heavy protocol, 3 and ....
D. G. Feitelson and M. A. Volovic, "ParPar design document version 0.2 ". URL http://www.cs.huji.ac.il/labs/parallel/parpar.html, Jun 1997.
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