3 citations found. Retrieving documents...
BBN, Butterfly Parallel Processor Overview, BBN Laboratories, Cambridge, Massachusetts, June 1985.

 Home/Search   Document Not in Database   Summary   Related Articles   Check  

This paper is cited in the following contexts:
An Interface Between Object-Oriented Systems - Crowl (1987)   (Correct)

....DCR 8320136. We thank the Xerox Corporation University Grants Program for providing equipment used in the preparation of this paper. 1 Introduction The Chrysalis operating system for the Butterfly Parallel Processor presents an objectoriented programming environment based on shared memory [BBN, 1985a; BBN, 1985b] However, because of Chrysalis s low level orientation and its use of type unsafe features of the C programming language [Kernighan and Ritchie, 1978] programs using the environment are difficult to program and highly error prone. Using C as the primary programming language for the ....

BBN, Butterfly Parallel Processor Overview, BBN Laboratories, Cambridge, Massachusetts, June 1985.


Performance Benefits and Limitations of Large NUMA.. - Sevcik, Zhou (1992)   (5 citations)  (Correct)

....since each ring segment adds one cycle to the round trip delay. As may be observed in Table 3, the access latencies in Hector form an approximate arithmetic progression, and the NUMAness of Hector is relatively mild compared to other hierarchical NUMA systems, such as DASH [4] and Butterfly [5]. In this section, we assume a configuration of six processors per station, five stations per local ring, and four local rings per global ring, for a total of one hundred twenty processors. This is the same structure as was assumed in section 2. Resource Utilization ( f 0 0.2 0.4 0.6 0.8 1 ....

Butterfly Parallel Processor Overview. Tech. Report No. 6148, BBN Laboratories, Cambridge, Mass. (1988).


Bridge: A High-Performance File System for Parallel Processors - Dibble (1988)   (35 citations)  Self-citation (Parallel)   (Correct)

....blocks are assigned to different physical nodes. Following a bit of background information, we introduce the notion of interleaving in section 3 and (in section 4) describe its realization in an experimental file system known as Bridge. Our prototype runs on a BBN Butterfly parallel processor [1], but the ideas on which it is based are equally applicable to a large number of other parallel architectures and to locally distributed collections of conventional machines. For the most critical file operations, early analytical and experimental results indicate that Bridge will deliver good ....

....functional layers. The top layer consists of the Bridge Server and a group of special purpose programs we call tools. The middle layer consists of local file systems on individual nodes. The lowest layer manages physical storage devices. Our implementation runs under the Chrysalis operating system [1] on the BBN Butterfly Parallel Processor [14] The components of the file system are implemented as user processes; no changes to the operating system were required. All components communicate via message passing. Messages are implemented on the Butterfly with atomic queues and buffers in shared ....

"Butterfly TM parallel processor overview," Tech. Rep. 6149, Version 2, BBN Laboratories, June 1986.

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