Increasing Performance in Commodity IP
نویسندگان
چکیده
Clusters built from commodity hardware and software have several advantages over more traditional supercomputers. Commodity clusters are cheap and ubiquitous, and so they are easier to design, program and maintain. However, even as high-speed networks reach 10Gb/s speeds and modern computer architectures reach past 1Ghz, new commodity clusters are not able to harness this power. One of the bottlenecks to high-performance computing using commodity hardware and protocols is interrupt pressure. We examine several methods used to relieve interrupt pressure in practice and propose a new solution, offloading IP fragmentation and reassembly onto a network interface card (NIC). Our proposed solution provides the performance gains of lower communication overhead and high bandwidth without sacrificing the advantages of using commodity hardware and protocols.
منابع مشابه
Experiments with Delivery of HDTV over IP Networks
We present the design, and preliminary performance results, for a system that transports uncompressed HDTV content over IP networks. Our system is motivated by the growth in use of digital video, and the ever increasing capacity of localand wide-area Internet links. We aim to demonstrate the feasibility of IP as a transport for very high quality video, and to highlight areas where performance b...
متن کاملA High Performance Parallel IP Lookup Technique Using Distributed Memory Organization and ISCB-Tree Data Structure
The IP Lookup Process is a key bottleneck in routing due to the increase in routing table size, increasing traıc and migration to IPv6 addresses. The IP address lookup involves computation of the Longest Prefix Matching (LPM), which existing solutions such as BSD Radix Tries, scale poorly when traıc in the router increases or when employed for IPv6 address lookups. In this paper, we describe a ...
متن کاملA High Performance Parallel IP Lookup Technique Using Distributed Memory Organization and ISCB-Tree Data Structure
The IP Lookup Process is a key bottleneck in routing due to the increase in routing table size, increasing traıc and migration to IPv6 addresses. The IP address lookup involves computation of the Longest Prefix Matching (LPM), which existing solutions such as BSD Radix Tries, scale poorly when traıc in the router increases or when employed for IPv6 address lookups. In this paper, we describe a ...
متن کاملHigh Performance Computing and Beowulf Clusters
Networks of personal computers (so called Beowulf systems) composed of fast PCs configured with large quantities of RAM and hard disk, and running the Linux operating system are becoming more and more attractive as cheap and efficient platforms for distributed applications. The main drawback of a standard Beowulf architecture is the poor performance of the conventional inter-process communicati...
متن کاملDistributed Shared Memory on IP Networks
Parallel programming is becoming an increasingly popular method of improving computer performance. Unfortunately, parallelized hardware remains expensive in the form of complicated symmetric multiprocessors and high-end server platforms. To provide a more available platform for parallel execution, we revisit the topic of implementing distributed shared memory on networks of commodity workstatio...
متن کامل