WIDE Paper-List in 2007 Implementation and Evaluation of Moderate Parallelism in the BIND9 DNS Server wide-paper-jinmei-bind9-parallelism-00.txt WIDE Project: http://www.wide.ad.jp/ If you have any comments on this document, please contact to ad@wide.ad.jp. ^L Title: Implementation and Evaluation of Moderate Parallelism in the BIND9 DNS Server Author(s): Tatuya Jinmei and Paul Vixie Date: 2007-01-07 Filename: wide-paper-jinmei-bind9-parallelism-00.txt % - 査読ありのコンファレンス・ワークショップ等 @inproceedings{ Jinmei:USENIX2006, author = "Tatuya Jinmei and Paul Vixie", title = "Implementation and Evaluation of Moderate Parallelism in the BIND9 DNS Server" booktitle = "Proceedings of 2006 USENIX Annual Technical Conference", pages = "115--128", month = "June", year = 2006 } %url: http://www.usenix.org/events/usenix06/tech/jinmei.html %x-wide-wgname: dns-wg %keywords: DNS, BIND9, Threads, Parallelism %references: Tatuya Yuji Sekiya, et.al., Cooking in Sillicon Valley %summary_ja: %summary: | % Suboptimal performance of the ISC BIND9 DNS server with multiple % threads is a well known problem. This paper explores practical % approaches addressing this long-standing issue. First, intensive % profiling identifies major bottlenecks occurring due to overheads for % thread synchronization. These bottlenecks are then eliminated by % giving separate work areas with a large memory pool to threads, % introducing faster operations on reference counters, and implementing % efficient reader-writer locks. Whereas some of the solutions % developed depend on atomic operations specific to hardware % architecture, which are less portable, the resulting implementation % still supports the same platforms as before through abstract APIs. % The improved implementation scales well with up to four processors % whether it is operating as an authoritative-only DNS server, with or % without dynamic updates, or as a caching DNS server. It also reduces % the memory footprint for large DNS databases. Acceptance of this new % sever will also have a positive side effect in that BIND9, and its new % features such as DNSSEC, should get wider acceptance. The direct % result has other ramifications: first, the better performance at the % application level reveals a kernel bottleneck in FreeBSD; also, while % the results described here are based on our experience with BIND9, the % techniques should be applicable to other thread-based applications. %misc: