The virtual computing environment

Rousselle, P.; Tymann, Paul; Hariri, Salim; Fox, Geoffrey

doi:10.1109/hpdc.1994.340265

Cited by 4 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are other metasystems [17][30] [43][50] and heterogeneous parallel computing [7][10] [18][46] projects underway. Our work differs from other heterogeneous parallel processing systems in the scope.…”

Section: Related Workmentioning

confidence: 99%

Campus-Wide Computing : Early Results Using Legion At the University of Virginia

Grimshaw

Nguyen-Tuong

Lewis

et al. 1997

The International Journal of Supercomputer Applications and Hig

View full text Add to dashboard Cite

The Legion project at the University of Virginia is an attempt to provide system services that provide the illusion of a single virtual machine to users, a virtual machine that provides both improved response time via parallel execution and greater throughput. Legion is targeted towards both workstation clusters and towards larger, wide-area, assemblies of workstations, supercomputers, and parallel supercomputers. Rather than construct Legion from scratch we are extending an existing object-oriented parallel processing system by aggressively incorporating lessons learned over twenty years by the heterogeneous distributed systems community. The campus-wide virtual computer is an early Legion prototype. In this paper we present challenges that had to be overcome to realize a working CWVC, as well as performance on a production biochemistry application.

show abstract

Section: Related Workmentioning

confidence: 99%

Campus-Wide Computing : Early Results Using Legion At the University of Virginia

Grimshaw

Nguyen-Tuong

Lewis

et al. 1997

The International Journal of Supercomputer Applications and Hig

View full text Add to dashboard Cite

show abstract

“…Building these memories bv hand can be an enormouslv difficult knowledge engineering task. ~e c e n t work has shown that large case bases can be automatically generated with A1 techniques (3) and accessed extremely efficiently by parallel inferencing techniques (4). 2) Hybnd knowledge and databases.…”

mentioning

confidence: 99%

Enterprise-Wide Computing

Grimshaw¹

1994

Science

View full text Add to dashboard Cite

pressing industrial and scientific problems.Recently, the use of high-performance computing techniques has led to the development of methods for building and accessing VLKBs with the faster processing and large memories enabled by modem supercomputers. Whereas entering the knowledge base for CYC has taken tens of person-years, these new techniques permit the automatic generation of VLKBs in much shorter times. In addition, new accessing u techniques provide searches that are several orders of magnitude faster than serial algorithms for matching complex patterns on relativelv unstructured data. Although u these techniques may not obviate the need for CYC-like wroiects (after all, common . , sense is hard to find, even among people), they open many intriguing possibilities. Some examples are mentioned here: 1) Large case-based svstems. Instead of . " solving problems from scratch, systems can solve new problems by analogy to previous solutions (2). Such "case-based"' reasoning requires very large memories of previous problem solutions. Building these memories bv hand can be an enormouslv difficult knowledge engineering task. ~e c e n t work has shown that large case bases can be automatically generated with A1 techniques (3) and accessed extremely efficiently by parallel inferencing techniques (4).2) Hybnd knowledge and databases. Many large corporate and scientific databases can be used to create A1 knowledge bases. First, specific information about the domain of discourse is used to encode knowledge about the underlying characteristics and functions of objects in that domain. Following this, traditional database queries are used to create a knowledge base relating swecific instances from the database to the more generic A1 knowledge. The resulting hybrid knowledge and database can be used to combine searching and inferencing, with supercompbting techniques again providing efficient pattern-matching capabilities that are difficult to encode and inefficient to run in the unaugmented database. This technique is particularly important in applications where old data must be explored in novel ways to see if recentlv discovered patterns werk previously existent in the database (examples include epidemiology and pharmaceutical research). 3) Software agents. A recent innovationin A1 technology is the creation of intelligent agents to help users explore complex unstructured information, such as that in the millions of documents distributed across the Internet.Although not yet a commercial technology, software agents are expected to become an important mechanism in providing access and navigation aids to the laree amounts of information . 3 stored in so-called cyberspace. Current In-ternet agents, for example, provide knowledge-based interface tools for making the net more user friendly (5) and use A1 techniques to help filter out vast amounts of irrelevant information (6). A recently started project in my laboratory, for example, focuses on the use of parallel inferencing techniques to provide a basis for creating agents t...

show abstract

VS3: The Virtual Storage Service-Based System

Xie

Zhao

et al. 2009

2009 Fourth International Conference on Internet Computing for Science and Engineering

View full text Add to dashboard Cite

The virtual computing environment

Cited by 4 publications

References 6 publications

Campus-Wide Computing : Early Results Using Legion At the University of Virginia

Campus-Wide Computing : Early Results Using Legion At the University of Virginia

Enterprise-Wide Computing

VS3: The Virtual Storage Service-Based System

Contact Info

Product

Resources

About