Exploiting Task-Level Concurrency in a Programmable Network Interface
dc.citation.bibtexName | inproceedings | en_US |
dc.citation.conferenceName | ACM SIGPLAN Symposium on Principles and Practices of Parallel Programming (PPoPP) | en_US |
dc.citation.location | San Diego, CA | en_US |
dc.contributor.author | Kim, Hyong-youb | en_US |
dc.contributor.author | Pai, Vijay S. | en_US |
dc.contributor.author | Rixner, Scott | en_US |
dc.date.accessioned | 2007-10-31T00:50:22Z | en_US |
dc.date.available | 2007-10-31T00:50:22Z | en_US |
dc.date.issued | 2003-06-20 | en_US |
dc.date.modified | 2003-06-14 | en_US |
dc.date.note | 2003-02-19 | en_US |
dc.date.submitted | 2003-06-20 | en_US |
dc.description | Conference Paper | en_US |
dc.description.abstract | Programmable network interfaces provide the potential to extend the functionality of network services but lead to instruction processing overheads when compared to application-specific network interfaces. This paper aims to offset those performance disadvantages by exploiting task-level concurrency in the workload to parallelize the network interface firmware for a programmable controller with two processors. By carefully partitioning the handler procedures that process various events related to the progress of a packet, the system can minimize sharing, achieve load balance, and efficiently utilize on-chip storage. Compared to the uniprocessor firmware released by the manufacturer, the parallelized network interface firmware increases throughput by 65% for bidirectional UDP traffic of maximum-sized packets, 157% for bidirectional UDP traffic of minimum-sized packets, and 32-107% for real network services. This parallelization results in performance within 10-20% of a modern ASIC-based network interface for real network services. | en_US |
dc.description.sponsorship | National Science Foundation | en_US |
dc.identifier.citation | H. Kim, V. S. Pai and S. Rixner, "Exploiting Task-Level Concurrency in a Programmable Network Interface," 2003. | en_US |
dc.identifier.uri | https://hdl.handle.net/1911/20031 | en_US |
dc.language.iso | eng | en_US |
dc.subject | parallel programming | en_US |
dc.subject | embedded systems | en_US |
dc.subject | network interfaces | en_US |
dc.subject | memory performance | en_US |
dc.subject.keyword | parallel programming | en_US |
dc.subject.keyword | embedded systems | en_US |
dc.subject.keyword | network interfaces | en_US |
dc.subject.keyword | memory performance | en_US |
dc.title | Exploiting Task-Level Concurrency in a Programmable Network Interface | en_US |
dc.type | Conference paper | en_US |
dc.type.dcmi | Text | en_US |