Server I/O networks past, present, and future
NICELI '03 Proceedings of the ACM SIGCOMM workshop on Network-I/O convergence: experience, lessons, implications
Designing Efficient Java Communications on Clusters
IPDPS '05 Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Workshop 5 - Volume 06
Exploiting NIC architectural support for enhancing IP-based protocols on high-performance networks
Journal of Parallel and Distributed Computing - Special issue: Design and performance of networks for super-, cluster-, and grid-computing: Part II
The Journal of Supercomputing
Java Fast Sockets: Enabling high-speed Java communications on high performance clusters
Computer Communications
Computer Networks: The International Journal of Computer and Telecommunications Networking
Cable-TV-based home video streaming system: practice and experience
IEEE Network: The Magazine of Global Internetworking
IPDPS'06 Proceedings of the 20th international conference on Parallel and distributed processing
Non-blocking java communications support on clusters
EuroPVM/MPI'06 Proceedings of the 13th European PVM/MPI User's Group conference on Recent advances in parallel virtual machine and message passing interface
Lightweight messages: true zero-copy communication for commodity gigabit ethernet
EUC'06 Proceedings of the 2006 international conference on Emerging Directions in Embedded and Ubiquitous Computing
NetSlices: scalable multi-core packet processing in user-space
Proceedings of the eighth ACM/IEEE symposium on Architectures for networking and communications systems
Hi-index | 0.00 |
While a number of User-Level Protocols have been developed to reduce the gap between the performance capabilities of the physical network and the performance actually available, applications that have already been developed on kernel based protocols such as TCP have largely been ignored. There is a need to make these existing TCP applications take advantage of the modern user-level protocols such as EMP or VIA which feature both low-latency and high bandwidth. In this paper, we have designed, implemented and evaluated a scheme to support such applications written using the sockets API to run over EMP withoutany changes to the application itself. Using this scheme, we are able to achieve a latency of 28.5 µs for the Datagram sockets and 37 µs for Data Streaming sockets compared to a latency of 120 µs obtained by TCP for 4-byte messages. This scheme attains a peak bandwidth of around 840 Mbps. Both the latency and the throughput numbersare close to those achievable by EMP. The ftp application shows twice as much benefit on our sockets interface while the web server application shows up to six times perfor-manceenhancement as compared to TCP. To the best of our knowledge, this is the first such design and implementation for Gigabit Ethernet.