In this work we describe our results on identifying the most important overheads in the network subsystem of massively parallel processors (MPPs), specially the Intel Paragon. We show how poor implementation techniques currently prevent performance of applications using TCP/IP protocols to com-municate between supercomputers connected by high performance networks from achieving data rates close to the capacity of the medium. We identify some of the possible solutions to the problem, and discuss what can be expected in future systems. In particular, we present our re-sults on evaluation of some of those solutions, including a user-level protocol implementation which scales with the number of concurrent connections and performance superior to...
High-end supercomputers are increasingly built out of commodity components, and lack tight integrati...
With processor speeds no longer doubling every 18–24 months owing to the exponential increase in pow...
Clusters of workstations are a popular platform for high-performance computing. For many parallel ap...
In this paper we review network related performance issues for current Massively Parallel Processors...
Parallel processing has been proposed as a means of improving network protocol throughput. Several d...
Typical MPI applications work in phases of computa-tion and communication, and messages are exchange...
This paper surveys parallelism in communication subsystems. Amodel that describes parallelism in ter...
This paper presents early experiences with using high-speed ATM interfaces to connect multiple Intel...
High-performance computing applications were once limited to isolated supercomputers. In the past fe...
Presently, massively parallel processors (MPPs) are available only in a few commercial models. A seq...
© 2008 ACM, Inc.Most modern parallel computers are clusters using Myrinet or Ethernet communication ...
Summarization: To meet the demand for higher performance, flexibility, and economy in today's state-...
The original publication can be found at www.springerlink.comThis paper gives an overview of two rel...
High-end supercomputers are increasingly built out of commodity components, and lack tight integrati...
This paper presents early experiences with using high-speed ATM interfaces to connect multiple Intel...
High-end supercomputers are increasingly built out of commodity components, and lack tight integrati...
With processor speeds no longer doubling every 18–24 months owing to the exponential increase in pow...
Clusters of workstations are a popular platform for high-performance computing. For many parallel ap...
In this paper we review network related performance issues for current Massively Parallel Processors...
Parallel processing has been proposed as a means of improving network protocol throughput. Several d...
Typical MPI applications work in phases of computa-tion and communication, and messages are exchange...
This paper surveys parallelism in communication subsystems. Amodel that describes parallelism in ter...
This paper presents early experiences with using high-speed ATM interfaces to connect multiple Intel...
High-performance computing applications were once limited to isolated supercomputers. In the past fe...
Presently, massively parallel processors (MPPs) are available only in a few commercial models. A seq...
© 2008 ACM, Inc.Most modern parallel computers are clusters using Myrinet or Ethernet communication ...
Summarization: To meet the demand for higher performance, flexibility, and economy in today's state-...
The original publication can be found at www.springerlink.comThis paper gives an overview of two rel...
High-end supercomputers are increasingly built out of commodity components, and lack tight integrati...
This paper presents early experiences with using high-speed ATM interfaces to connect multiple Intel...
High-end supercomputers are increasingly built out of commodity components, and lack tight integrati...
With processor speeds no longer doubling every 18–24 months owing to the exponential increase in pow...
Clusters of workstations are a popular platform for high-performance computing. For many parallel ap...