A case for NOW (networks of workstation)

@inproceedings{Patterson1995ACF,
  title={A case for NOW (networks of workstation)},
  author={David A. Patterson and David E. Culler and Thomas E. Anderson},
  booktitle={PODC '95},
  year={1995}
}
In this paper, we argue that because of recent technology advances, networks of workstations (NOWs) are poised to become the primary computing infrastructure for science and engineering, from low end interactive computing to demanding sequential and parallel applications. We identify three opportunities for NOWs that will benefit end- users: dramatically improving virtual memory and file system performance by using the aggregate DRAM of a NOW as a giant cache for disk; achieving cheap, highly… 

Figures and Tables from this paper

High-performance sorting on networks of workstations

We report the performance of NOW-Sort, a collection of sorting implementations on a Network of Workstations (NOW). We find that parallel sorting on a NOW is competitive to sorting on the large-scale

A distributed device paradigm for commodity, applications

TLDR
This paper introduces a client-driven paradigm for harnessing specific system components and presenting them to the operating system as local devices (Virtual Devices) based on the Transparent Remote Execution Environment (TREx).

Hardware Support for Flexible Distributed Shared Memory

TLDR
It is found that, even with aggressive DSM hardware support, custom protocols can provide significant speedups for some applications, and the additional hardware acceleration provided by the most aggressive design avoids the need to develop more efficient custom protocols.

Constructing a Memory-Based Distributed Code Storage on Networked Diskless Embedded Systems

TLDR
In comparison with a client-server approach, the experimental results show that NFC successfully integrates memory resources in a network to significantly extend code storage and reduce file-access time of a diskless embedded system.

Large Scientific Calculations an Dedicated Clusters of Workstations

TLDR
Results show that the SCI clusters can be favorabl e in price and competitive in performance to large multi-proces sor servers, in particular when the aggregated memory demand is high.

Anonymous Remote Computing: A Paradigm for Parallel Programming on Interconnected Workstations

TLDR
The anonymous remote computing (ARC) paradigm is proposed to address the issues specific to parallel programming on workstation systems and it provides fault tolerance and load adaptability to parallel programs on workstations.

Remote Memory Access in Workstation Clusters

TLDR
It is found that the software-only model, with current remote page-fault latencies, does not provide acceptable memory-sharing performance, and experiments show that a gigabit network is necessary for good performance.

A Parallel/Distributed Platform for University Computational Infrastructure Service Provisioning *

TLDR
The proposed system named PLEIADES allows its users to transparently utilize various computer resources in order to form virtual clusters for their parallel/distributed computational needs.

Realizing the performance potential of the virtual interface architecture

TLDR
This paper examines the low-le vel performance of two VI implementations, one implemen ted in hardware, the other implemented in device driver so ftware, and reports that both VI implementations offer significa nt performance advantage relative to the corresponding UDP impleme ntation on the same hardware.

An Integrated Simulation Environment for Parallel and Distributed System Prototying

TLDR
The concepts, mechanisms, and results of an Integrated Simulation Environment (ISE) that makes possible the rapid virtual prototyping and profiling of legacy and prototype parallel process ing algorithms, architectures, and systems using a networked cluster of workstations are introduced.
...

References

SHOWING 1-10 OF 22 REFERENCES

A Case for NOW (Networks Of Workstations)

TLDR
The 100-node NOW prototype aims to demonstrate practical solutions to the challenges of efficient communication hardware and software, global coordination of multiple workstation operating systems, and enterprise-scale network file systems.

The interaction of parallel and sequential workloads on a network of workstations

TLDR
This paper examines the plausibility of using a network of workstations (NOW) for a mixture of parallel and sequential jobs, and presents a methodology for deriving an optimal delay time for recruiting idle machines for use by parallel programs.

Active Messages: A Mechanism for Integrated Communication and Computation

TLDR
It is shown that active messages are sufficient to implement the dynamically scheduled languages for which message driven machines were designed and, with this mechanism, latency tolerance becomes a programming/compiling concern.

HPAM: an active message layer for a network of hp workstations

TLDR
The HPAM layer currently runs on a network of 4 HP 9000/ 735 workstations with Medusa FDDI interface cards and achieves a round trip time for a 20 byte payload of 29 psec, an order of magnitude better than traditional software, and a maximum bandwidth of 12 MB/s.

Preemptable remote execution facilities for the V-system

TLDR
It is argued that preemptable remote execution allows idle workstations to be used a a "pool of processors" without interfering with use by their owners and without significant overhead for the normal execution of programs.

LogP: towards a realistic model of parallel computation

TLDR
A new parallel machine model, called LogP, is offered that reflects the critical technology trends underlying parallel computers and is intended to serve as a basis for developing fast, portable parallel algorithms and to offer guidelines to machine designers.

Cooperative caching: using remote client memory to improve file system performance

TLDR
It is concluded that cooperative caching can significantly improve file system read response time and that relatively simple cooperative caching algorithms are sufficient to realize most of the potential performance gain.

Efficient software-based fault isolation

TLDR
It is demonstrated that for frequently communicating modules, implementing fault isolation in software rather than hardware can substantially improve end-to-end application performance.

Computer Architecture: A Quantitative Approach

This best-selling title, considered for over a decade to be essential reading for every serious student and practitioner of computer design, has been updated throughout to address the most important