Final report Page: 2 of 23
This text is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to UNT Digital Library by the UNT Libraries Government Documents Department.
Extracted Text
The following text was automatically extracted from the image on this page using optical character recognition software:
the rapid deployment of high-end community services. In this proposal, we will focus on scalable middle-
ware in support of resource management and reliability. We also propose a system architecture that inte-
grates the middleware components. Our middleware and system architecture will be designed to
accommodate and integrate middleware solutions for security and user interface' developed by other
groups. We will produce middleware that can be leveraged by community services running in clusters,
supercomputers, and in Grids. One of the novel aspects of our approach is that the tension between
resource sharing for the "common good" and resource monopolization for the "individual good" is signifi-
cantly reduced. To increase the impact of this project, the middleware will be integrated into a widely used
implementation of the Message-Passing Interface (MPI), MPICH from Argonne National Laboratory, and
the Condor system from the University of Wisconsin. The middleware will be evaluated by applying it to
high-end network services of interest to DOE.
Reliability
Security -rchitecture
Community
Service
Resource User
Management Interface
Figure 1: Community Service Framework. This project will address bold pieces. Middleware "hooks" (dotted lines)
will be provided to third party solutions for security (e.g. Kerberos) and user interface (e.g. GridRPC, web services).
1.1 High-End Network Services: Models and Middleware
We plan to investigate two "application classes" very common to high performance computing: data
parallel and distributed (Figure 2). This focus allows the project to instantly impact a large class of applica-
tions, but the middleware developed will still be very useful to other application classes. Complete support
for other application classes is our ultimate objective. To test our middleware, we plan to develop services
in each class.
Parallel services are implemented by parallel programs that require communication between the con-
stituent processors. Examples of parallel services include numeric solvers, N-body simulators, parallel
CFD, etc. Distributed services are similar to parallel services, but do not require communication among the
distributed processors. Examples include stochastic simulation (e.g. monte-carlo), parameter studies, etc.
Each processor is assigned a set of independent tasks on demand from a master node, and communicate
results back to it. In this project, we plan to develop several parallel and distributed network services to test1. By user interface, we refer to the low-level interface API, e.g. web service, RPC, etc, as opposed to a GUI level.
Upcoming Pages
Here’s what’s next.
Search Inside
This text can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Text.
Weissman, Jon B. Final report, text, April 30, 2006; United States. (https://digital.library.unt.edu/ark:/67531/metadc835934/m1/2/: accessed April 23, 2024), University of North Texas Libraries, UNT Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.