T

Nikhil, Rishiyur S.; Papadopoulos, Gregory M.; Arvind, .

doi:10.1145/139669.139715

Cited by 161 publications

(3 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Traditionally, these multiprocessor systems, e.g., the JMachine [8], *T [16], etc., have opted for a single primitive that unifies remote transfer of data and control, in contrast to our approach.…”

Section: Discussionmentioning

confidence: 99%

Separating data and control transfer in distributed operating systems

Thekkath¹,

Levy

Lazowska

1994

SIGOPS Oper. Syst. Rev.

View full text Add to dashboard Cite

Advances in processor architecture and technology have resulted in workstations in the 100+ MIPS range. As well, newer local-area networks such as ATM promise a ten-to hundred-fold increase in throughput, much reduced latency, greater scalability, and greatly increased reliability, when compared to current LANs such as Ethernet.We believe that these new network and processor technologies will permit tighter coupling of distributed systems at the hardware level, and that distributed systems software should be designed to benefit from that tighter coupling.In this paper, we propose an alternative way of structuring distributed systems that takes advantage of a communication model based on remote network access (reads and writes) to protected memory segments.A key feature of the new structure, directly supported by the communication model, is the separation of data transfer and contd transfer.This is in contrast to the structure of traditional distributed systems, which are typically organized using message passing or remote procedure call (RPC). In RPC-style systems, data and contiol are inextricably linked-all RPCS must transfer both data and control, even if the control transfer is unnecessary.We have implemented our model on DECstation hardware connected by an ATM network. We demonstrate how separating data transfer and control transfer can eliminate unnecessary control transfers and facilitate tighter coupling of the client and server. This has the potential to increase performance and reduce server load, which supports scaling in the face of an increasing number of clients. For example, for a small set of file server operations, our analysis shows a 50'%0decrease in server load when we switched from a communications mechanism requirhg both control transfer and data transfer, to an alternative structure based on pure data transfer.tAUthor's currentaddress

show abstract

Section: Discussionmentioning

confidence: 99%

Separating data and control transfer in distributed operating systems

Thekkath¹,

Levy

Lazowska

1994

SIGOPS Oper. Syst. Rev.

View full text Add to dashboard Cite

show abstract

“…A number of multithreaded architecture designs have been reported in the literature [2,4,7,10,11,16,18]. The performance of multithreaded multiprocessor systems have been studied through an analytical approach by Agarwal [1], Alkalaj and Bopanna [3], Boothe and Ranade [5], Nemawarkar and Gao [14], Nemawarkar, et al [15], and Saavedra-Barrera, et al [17].…”

Section: Related Workmentioning

confidence: 99%

An executable analytical performance evaluation approach for early performance prediction

Jacquet

Janot

Leung

et al.

Proceedings International Parallel and Distributed Processing Symposium

View full text Add to dashboard Cite

Percolation has recently been proposed as a key component of an advanced program execution model for future generation high-end machines featuring adaptive data/code transformation and movement for effective latency tolerance. An early evaluation of the performance effect of percolation is very important in the design space exploration of future generations of supercomputers. In this paper, we develop an executable analytical performance model of a high performance multithreaded architecture that supports percolation. A novel feature of our approach is modeling interactions between software (program) and hardware (architecture) components. We solve the analytical model using a queuing simulation tool enriched with synchronization. The proposed approach is effective and facilitates obtaining performance trends quickly. Our results indicate that percolation brings in significant performance gains (by a factor of 2.7 to 11). Further, our results reveal that percolation and multithreading can complement each other.

show abstract

“…This lias been done in the Thinking Machines CM-5 [40] and i n the *?' project [39]. However, problems arise when a group of processors is reallocated from one job to another bec.ause messages left in the network may arrive a t a target processor unprepared to accept them.…”

Section: Fine-grain Communicationmentioning

confidence: 99%

Semantic Models For Parallel Computers

Dennis¹

A Workshop on Future Directions of Massively Parallel Processing, the New Frontiers

View full text Add to dashboard Cite

past designs and experience.As a design tool for computer systems, a semantic model defines an formal interface (contract) between a user (an application program) and a n implementor (the computer system designer). The semantic model must incorporate mechanisms for dealing the issues that m u s t be addressed if the pro"&ctzve benefits of designing parallel computers o n the basis of a sound and complete se computation are to be achieve view the weakchitectures for massively paralt out how these are illustrative design of multiprocessor sysm o n informal semantic models that are used as the basis f o r programming purallel c o r n p q i p n d the ways in which they are unsatisfactory. * y poant out the requirements to support the modular construction of software and note 2h.e implications these principles have f o r a satasfactory general semantic model.

show abstract

T

Cited by 161 publications

References 20 publications

Separating data and control transfer in distributed operating systems

Separating data and control transfer in distributed operating systems

An executable analytical performance evaluation approach for early performance prediction

Semantic Models For Parallel Computers

Contact Info

Product

Resources

About