NFV Compute Acceleration Evaluation and APIs

Document Type Expired Internet-Draft (individual)
Last updated 2015-12-31 (latest revision 2015-06-29)
Stream (None)
Intended RFC status (None)
Expired & archived
plain text pdf html bibtex
Stream Stream state (No stream defined)
Consensus Boilerplate Unknown
RFC Editor Note (None)
IESG IESG state Expired
Telechat date
Responsible AD (None)
Send notices to (None)

This Internet-Draft is no longer active. A copy of the expired Internet-Draft can be found at


Network functions are being virtualized and moved to industry standard servers. Steady growth of traffic volume requires more compute power to process the network functions. Network packet based architecture provides a lot of scope for parallel processing. Generic parallel processing can be done in common multicore platforms like GPUs, coprocessors like Intel Xeon Phi[6][7] and Intel[7]/AMD[10] multicore CPUs. In this draft to check the feasibility and to exploit this parallel processing capability, multi string matching is taken as the sample network function for URL filtering. Aho-Corasick algorithm has been made use for multi pattern matching. Implementation utilizes OpenCL [3] to support many common platforms[7][10][11]. A list of optimizations is done, the application is tested on Nvidia Tesla K10 GPUs. A common API for NFV Compute Acceleration has been proposed.


Bose Perumal (
Wenjing Chu (
Ramki Krishnan (

(Note: The e-mail addresses provided for the authors of this Internet-Draft may no longer be valid.)