Results 1 -
6 of
6
PULC: ParaStation User-Level Communication. Design and Overview
- In Parallel and Distributed Processing
, 1998
"... . PULC is a user-level communication library for workstation clusters. PULC provides a multi-user, multi-programming communication library for user level communication on top of high-speed communication hardware. In this paper, we describe the design of the communication subsystem, a first implement ..."
Abstract
-
Cited by 6 (1 self)
- Add to MetaCart
(Show Context)
. PULC is a user-level communication library for workstation clusters. PULC provides a multi-user, multi-programming communication library for user level communication on top of high-speed communication hardware. In this paper, we describe the design of the communication subsystem, a first implementation on top of the ParaStation communication card, and benchmark results of this first implementation. PULC removes the operating system from the communication path and offers a multi-process environment with user-space communication. Additionally, we have moved some operating system functionality to the user level to provide higher efficiency and flexibility. Message demultiplexing, protocol processing, hardware interfacing, and mutual exclusion of critical sections are all implemented in user-level. PULC offers the programmer multiple interfaces including TCP user-level sockets, MPI [CGH94], PVM [BDG + 93], and Active Messages [CCHvE96]. Throughput and latency are close to the hardware ...
Using Workstations as Building Blocks for Parallel Computing
- of the International Conference on Parallel and Distributed Processing, Techniques and Applications (PDPTA'96
, 1999
"... The key to efficient parallel computing on workstations clusters is a communication subsystem that removes the operating system from the communication path and eliminates all unnecessary protocol overhead. At the same time, protection and a stable multi-user, multiprogrammed environment cannot be sa ..."
Abstract
-
Cited by 1 (0 self)
- Add to MetaCart
The key to efficient parallel computing on workstations clusters is a communication subsystem that removes the operating system from the communication path and eliminates all unnecessary protocol overhead. At the same time, protection and a stable multi-user, multiprogrammed environment cannot be sacrificed. We have developed a communication subsystem, called ParaStation2, which fulfills these requirements. Its one-way latency is 14:5¯s to 18¯s (depending on the hardware platform) and throughput is 65 to 90 MByte/s, which compares well with other approaches. We were able to achieve an application performance of 5.3 GFLOP running a matrix multiplication on 8 DEC Alpha machines (21164A, 500 MHz). ParaStation2 offers standard programming interfaces, including PVM, MPI, Unix sockets, Java sockets, and Java RMI. These interfaces allow parallel applications to be ported to ParaStation2 with minimal effort. The system is implemented on a variety of platforms, including DEC Alpha workstations ...
The ParaStation Project
"... . ParaStation is a high speed communication subsystem which enables a cluster of workstations or PCs to offer performance comparable to a dedicated parallel machine. Each node is still usable as a regular workstation or PC. Speedups of 7.4 on a 8 node cluster have been achieved even on communication ..."
Abstract
-
Cited by 1 (0 self)
- Add to MetaCart
(Show Context)
. ParaStation is a high speed communication subsystem which enables a cluster of workstations or PCs to offer performance comparable to a dedicated parallel machine. Each node is still usable as a regular workstation or PC. Speedups of 7.4 on a 8 node cluster have been achieved even on communication intensive programs. The ParaStation software supports a real multitasking environment and combines the advantages of workstation clusters and of parallel machines. 1 Introduction The basic concept of the ParaStation approach is to build a scalable and efficient parallel processor from off-the-shelf workstations and PCs. In ParaStation this is done with a second interconnection network dedicated to parallel computation together with well-known programming interfaces such as UNIX sockets and PVM. By taking advantage of mass production, the resulting system is considerably less expensive than traditional parallel machines. The ParaStation system [8] is based on the retargeted MPP network of T...
ParaStation User Level Communication
"... This paper describes the design of the communication subsystem, a first implementation on top of the ParaStation communication adapter, and benchmark results of this first implementation ..."
Abstract
- Add to MetaCart
(Show Context)
This paper describes the design of the communication subsystem, a first implementation on top of the ParaStation communication adapter, and benchmark results of this first implementation
The ParaStation2 Cluster Environment
"... The key to efficient parallel computing on workstations clusters is a communication subsystem that removes the operating system from the communication path and eliminates all unnecessary protocol overhead. But the key to effective and widespread cluster computing in general is to provide a cluster e ..."
Abstract
- Add to MetaCart
(Show Context)
The key to efficient parallel computing on workstations clusters is a communication subsystem that removes the operating system from the communication path and eliminates all unnecessary protocol overhead. But the key to effective and widespread cluster computing in general is to provide a cluster environment which offers high performance and well known programming interfaces as well as an integrated management and administration of the cluster that makes it suitable for a widespread community. The ParaStation2 cluster environment fulfills these requirements. Its one-way latency for all programming inter-faces is about ¡£¢¥¤§¦
High Performance Parallel Computing with ParaStation3
"... Since 1995 Workstation clusters coupled by high-speed interconnection networks offer a promising direction for high performance computing because they are cost-effective and they closely track technology progress. In contrast to supercomputers and parallel machines, clustered workstations rely on st ..."
Abstract
- Add to MetaCart
Since 1995 Workstation clusters coupled by high-speed interconnection networks offer a promising direction for high performance computing because they are cost-effective and they closely track technology progress. In contrast to supercomputers and parallel machines, clustered workstations rely on stan-
