Last modified 09:10:00, 9 Dec 1996 - 330.64453125K - PostScript
Last modified 00:00:00, 30 Jul 1993 - 74.4K - compressed PostScript
A "general" paper about the evolution of the PVM
system (not much content). Appeared in Compcon'93.
Last modified 00:00:00, 23 Feb 1993 - 65.7K - compressed PostScript
Overview paper on PVM & HeNCE, appeared
in Computers in Physics April'93.
Last modified 00:00:00, 30 Jul 1993 - 36.8K - compressed PostScript
A "white-paper" & experiences from a test implementation
for general purpose concurrent computing primitives
Appeared in ICCI 93.
Last modified 00:00:00, 27 Jan 1993 - 69.9K - compressed PostScript
Shared-object interfaces to PVM, from an MS thesis,
appeared in Intl. Parallel Proc Symposium, Apr'93.
Last modified 00:00:00, 18 May 1994 - 114.4013671875K - compressed PostScript
PVM performance and analysis of overheads using textbook
problems and suggested techniques for improvement.
Last modified 19:36:00, 22 Nov 1994 - 43.1K - compressed PostScript
Viewgraphs of a talk on threads and parallel I/O in PVM
presented at the European PVM Users Group meeting, Rome 10/94.
Last modified 00:00:00, 30 Jul 1993 - 171.5546875K - compressed PostScript
Expanded version of Eclipse Gordon Bell paper
Submitted to JPDC.
Last modified 00:00:00, 30 Jul 1993 - 62.4K - compressed PostScript
The Eclipse paper that won the 92 Gordon Bell prize.
Appeared in Supercomputing 92.
Last modified 20:36:00, 6 Jan 1995 - 4.5K - (La)TeX
Template TeX document for IEEE papers?
Last modified 15:02:00, 21 Jul 1998 - 215.7734375K - PostScript
Last modified 18:57:00, 2 Feb 1995 - 143.5107421875K - compressed PostScript
Description of recent and upcoming PVM developments. Invited
paper to appear in Intl. Journal of Supercomputer Apps.
Last modified 00:00:00, 6 May 1994 - 93.2K - compressed PostScript
PVM performance results for the NAS parallel benchmarks on
Enet, FDDI, Gswitch, and new fast communication schemes.
(to appear in JPDC). A slightly expanded version of this,
in NASA Ames tech report format is in nasapvm1.ps.Z.
- Performance of the NAS Parallel Benchmarks on PVM Based
Last modified 00:00:00, 6 May 1994 - 96.6K - compressed PostScript
Abstract: The NAS parallel benchmarks are a set of applications that embody
the key computational and data-movement characteristics of typical
processing in computational aerodynamics. Five of these, the kernel
benchmarks, have been implemented on the PVM system, a software
system for network-based concurrent computing, with a view to
determining the efficacy of networked environments for
high-performance computational aerodynamics applications, and to
experimentally investigate enhancements to the software
infrastructure that optimize communication performance in such
environments. We present results of porting and executing the NPB
kernels in three different cluster environments using low- to
medium-powered workstations on Ethernet and two types of FDDI
networks. Our results indicate that mediocre to very good
performance could be obtained despite the communications intensive
nature of the applications. In most cases, we were able to achieve
performance levels within an order of magnitude of a Cray Y/MP-1 on
8-workstation clusters via optimizations to the PVM infrastructure
alone, i.e. with little or no algorithmic modifications. However,
our results also indicate that further improvements are possible,
and that network based computing has the potential to be a viable
technology for high-performance scientific computing.
Authors: S. White;
A. Alundy and
Last modified 00:00:00, 10 Nov 1993 - 79.4K - compressed PostScript
Vugraphs from an Oct'93 talk on the NAS parallel benchmarks
on PVM, and a new improved communication scheme.
Last modified 00:00:00, 20 Jan 1993 - 116.6025390625K - compressed PostScript
The "experiences" paper on PVM - about XPVM, PICL
ports, and real applications. CP&E Jun'92.
- OMIS - On-line Monitoring Interface Specification
Last modified 17:59:00, 15 Feb 1996 - 126.525390625K - gzipped PostScript
Abstract: The On-line Monitoring Interface Specification (OMIS) aims at
defining an open interface for connecting on-line software
development tools to parallel programs running in a
distributed environment. Interactive tools like debuggers and
performance analyzers and automatic tools like load balancers
are typical representatives of the considered class of tools.
The current situation is characterized by the fact that tools
either follow the off-line paradigm by only having access to
trace data and not to the running program or else they are
on-line oriented but suffer from the following deficiencies:
they do not support interoperability in the sense that
different tools can be used simultaneously - not even tools
from the same developer. Furthermore, no unified environment
exists where the same tools can be used for parallel programs
running on different target architectures. A reason for this
situation can be found in a lack of systematic development of
monitoring systems, i.e. systems which provide a tool with
necessary runtime information about the application programs
and make it possible to even manipulate the program run. The
goal of the OMIS project is to specify an interface which is
appropriate for a large set of different tools. Having an
agreed on on-line monitoring interface facilitates the
development of tools in the way that tool implementation and
monitoring system implementation are now decoupled. Bringing
n tools to m systems (consisting of hardware, operating
system, programming libraries etc.) will be reduced in
complexity from n m to n + m. In addition, it will eventually
be possible to simultaneously use tools of different
developers and to compose unified tool environments. The
research group at LRR-TUM will implement an OMIS compliant
monitoring system for the PVM programming model running on a
network of workstations. Several interactive and automatic
tools will be connected to this concrete system. The present
document defines the goals of the OMIS project and list
necessary requirements for such a monitoring system. We will
describe the system model OMIS is primarily intended for and
give an outline of available services of the interface. A
special section will give details on how to extend OMIS, as
this is an indispensable feature for future tool development.
We would appreciate to get feedback on the design of OMIS. If
you would like to see special issues incorporated into this
specification document you are invited to contact the
Authors: Thomas Ludwig (email@example.com), Lehrstuhl fur Rechnertechnik und
Rechnerorganisation, Institut fur Informatik (LRR-TUM),
Technische Universitat Munchen, D-80290 Munchen, Germany; Tel: +49-89-2105-2042; FAX: +49-89-2105-8232;
Roland Wismuller (firstname.lastname@example.org);
Vaidy Sunderam (email@example.com), Mathematics & Computer Science, Emory
University, Atlanta, Georgia 30322, USA; Tel: +1-404-727-5926; FAX: +1-404-727-5611 and
Arndt Bode (firstname.lastname@example.org).
Last modified 00:00:00, 17 Sep 1993 - 66.5K - compressed PostScript
Invited paper at Parco'93 in Grenoble. General PVM & Eclipse
paper, with some new performance results and future
directions in IO & threads.
Last modified 18:22:00, 22 Mar 1995 - 108.8857421875K - compressed PostScript
ICDCS'95 paper on parallel visualization.
Authors: John Stasko and
Last modified 00:00:00, 4 Nov 1993 - 72.6K - compressed PostScript
General PVM paper to appear in David Walker/Rolf Hempel's
special issue of Parallel Computing on Message Passing Models.
Last modified 19:36:00, 19 Oct 1995 - 87.7K - compressed PostScript
Comparison of parallel vs. distributed I/O, issues in
distributed I/O mechanisms over general purpose networks,
Pious file model, interface, implementation and
experiences. To appear in FGCS 95.
Last modified 00:00:00, 19 Jul 1994 - 54.3K - compressed PostScript
PIOUS - the parallel I/O system - architecture, design.
Last modified 00:00:00, 19 Jul 1994 - 76.0K - compressed PostScript
PIOUS - the parallel I/O system - interfaces,
Last modified 00:00:00, 14 Oct 1993 - 16.5K - compressed tar
Not a paper - but source code for a few textbook examples,
but only tested on 3.1 ...
Last modified 21:11:00, 24 Feb 1995 - 217.7666015625K - compressed tar
The NAS parallel benchmarks (all 8) ported to PVM 3.3. This is
only an alpha version, so use at your own risk.
- PVM versions of 5 NAS Parallel Benchmarks Kernels
Last modified 16:45:00, 27 Nov 1995 - 64.3K - compressed tar
Benchmarks contained: pvmep, pvmcg, pvmmg, pvmmfr, pvmmis.
Author: Anders Alund
Last modified 00:00:00, 23 Feb 1993 - 122.8837890625K - compressed tar
PVM performance figures (PostScript)
Last modified 11:54:00, 22 Jul 1998 - 135.9453125K - PostScript
Last modified 12:38:30, 18 Aug 1998 - 184.0703125K - PostScript
Last modified 00:00:00, 28 Feb 1993 - 36.1K - compressed tar
Viewgraphs for a brief talk on PVM.
Last modified 00:00:00, 20 Jan 1993 - 78.7K - compressed PostScript
The "original" PVM paper, appeared in CP&E Dec'90.
Last modified 00:00:00, 19 Mar 1993 - 274.6337890625K - compressed tar
Viewgraphs for a long, sometimes tedious, talk
or tutorial on PVM.
Last modified 19:08:00, 31 Oct 1994 - 2.8K - text
HCW 95 referee report form.
Last modified 00:00:00, 22 Mar 1993 - 119.3564453125K - compressed PostScript
One of the early conference papers on HeNCE
Appeared in Supercomputing 1991.
Last modified 23:48:00, 21 Jul 1996 - 297.6083984375K - PostScript
Last modified 00:00:00, 22 Mar 1993 - 77.4K - compressed PostScript
A brief paper on PVM with some narrative about
large computational problems. Appeared in
Siam Parallel Proc. Conference, 1991.
Last modified 00:00:00, 29 Jul 1993 - 142.8291015625K - compressed PostScript
Third-party critique. Supposed to be supplemented by
a 200-page evaluation report.
Authors: Tony Sjkellum and
- Multiparadigm Distributed Computing with TPVM
Last modified 07:47:00, 14 Mar 1996 - 122.9248046875K - compressed PostScript
Abstract: Distributed concurrent computing based on lightweight
processes can potentially address performance and
functionality limits in heterogeneous systems. The TPVM
framework, based on the notion of "exportable services", is
an extension to the PVM message passing system, but uses
threads as units of computing, scheduling, and
parallelism. TPVM facilitates and supports three different
distributed concurrent programming paradigms: (a) the
traditional, task based, explicit message passing model; (b)
a data-driven instantiation model that enables
straightforward specification of computation based on data
dependencies; and (c) a partial shared-address space model
via remote memory access, with naming and typing of
distributed data areas. The latter models offer significantly
different computing paradigms for network-based computing,
while maintaining a close resemblance to, and building upon,
the conventional PVM infrastructure in the interest of
compatibility and ease of transition. The TPVM system
comprises three basic modules: a library interface that
provides access to thread-based distributed concurrent
computing facilities, a portable thread interface module
which abstracts the required threads-related services, and a
thread server module which performs scheduling and system
data management. System implementation as well as
applications experiences have been very encouraging,
indicating the viability of the proposed models, the
feasibility of portable and efficient threads systems for
distributed computing, and the performance improvements that
result from multithreaded concurrent computing.
Authors: Adam Ferrari (email@example.com), Department of Computer Science,University of
Virginia, Charlottesville, VA 22903, USA and
V. S. Sunderam (firstname.lastname@example.org), Department of Computer Science,University of
Virginia, Charlottesville, VA 22903, USA.
Last modified 00:00:00, 19 Sep 1994 - 867.6962890625K - compressed tar
A preliminary, even pre-alpha, version of the TPVM threads-based
PVM subsystem. Caveat receptor!
Last modified 00:00:00, 8 Jun 1993 - 26.3K - compressed PostScript
Viewgraphs on a recent talk, about futures
including I/O & groups.
Last modified 09:00:30, 20 Oct 1998 - 8.7M - binary