This paper describes a new interconnect standard, the Scalable Coherent Interface (SCI, ANSI/IEEE Std 1596-1992), and new developments in parallel computer architecture that become possible by this technology. For instance, using SCI, shared-memory multiprocessors can be built from standard desktop machines at low costs. The technical features and applications of SCI, novel architectural opportunities it opens, and its advantages in terms of low costs and high performance, are described. In addition, a project at our institution (LRR-TUM) is outlined that sets up, and develops software for, an SCI-based distributed shared memory multiprocessor built from standard desktop computers. Keywords: Scalable Coherent Interface (SCI), PVM, D...
We are currently investigating two differentapproaches to scalable shared memory: Munin, a distribut...
In this paper we discuss our preliminary experience with implementing active messages and Split-C on...
In the past decade there has been a dramatic shift from mainframe or ‘host-centric ’ computing to a ...
The Scalable Coherent Interface (SCI) project (IEEE P1596) found a way to avoid the limits that are ...
Due to their excellent cost/performance ratio, clusters of PCs can be attractive high-performance co...
In this paper we wish to show some of the possible ways in which small SCI-rings can be connected to...
In this paper we discuss our experience with highperformance cluster computing using Scalable Cohere...
Recently several network technologies which support user-level communication between processes using...
The Scalable Coherent Interface (SCI) defines a high-speed interconnect that provides for a distribu...
A cluster is a group of independent compute nodes which are coupled by an interconnection network. T...
This paper describes work in progress to develop a standard for interoperability among high-performa...
This paper describes work in progress to develop a stan-dard for interoperability among high-perform...
A Local Area Network (LAN) can be built using the ANSI/IEEE Standard 1592-1992 Scalable Coherent Int...
The arrival multi-core processors or chip multiprocessors (CMP) operated with symmetrical multiproce...
current approaches to distributed sh ared-memo y computing, these authors find that the reduced cost...
We are currently investigating two differentapproaches to scalable shared memory: Munin, a distribut...
In this paper we discuss our preliminary experience with implementing active messages and Split-C on...
In the past decade there has been a dramatic shift from mainframe or ‘host-centric ’ computing to a ...
The Scalable Coherent Interface (SCI) project (IEEE P1596) found a way to avoid the limits that are ...
Due to their excellent cost/performance ratio, clusters of PCs can be attractive high-performance co...
In this paper we wish to show some of the possible ways in which small SCI-rings can be connected to...
In this paper we discuss our experience with highperformance cluster computing using Scalable Cohere...
Recently several network technologies which support user-level communication between processes using...
The Scalable Coherent Interface (SCI) defines a high-speed interconnect that provides for a distribu...
A cluster is a group of independent compute nodes which are coupled by an interconnection network. T...
This paper describes work in progress to develop a standard for interoperability among high-performa...
This paper describes work in progress to develop a stan-dard for interoperability among high-perform...
A Local Area Network (LAN) can be built using the ANSI/IEEE Standard 1592-1992 Scalable Coherent Int...
The arrival multi-core processors or chip multiprocessors (CMP) operated with symmetrical multiproce...
current approaches to distributed sh ared-memo y computing, these authors find that the reduced cost...
We are currently investigating two differentapproaches to scalable shared memory: Munin, a distribut...
In this paper we discuss our preliminary experience with implementing active messages and Split-C on...
In the past decade there has been a dramatic shift from mainframe or ‘host-centric ’ computing to a ...