Date: Thu, 4 May 2017 03:51:13 +0000 (UTC) From: Benjamin Kaduk <bjk@FreeBSD.org> To: doc-committers@freebsd.org, svn-doc-all@freebsd.org, svn-doc-head@freebsd.org Subject: svn commit: r50229 - head/en_US.ISO8859-1/htdocs/news/status Message-ID: <201705040351.v443pDBH036836@repo.freebsd.org>
next in thread | raw e-mail | index | archive | help
Author: bjk Date: Thu May 4 03:51:12 2017 New Revision: 50229 URL: https://svnweb.freebsd.org/changeset/doc/50229 Log: Some corrections to the 2017Q1 pNFS entry I had mixed up the File Layout and Flex File Layout; only the latter allows NFSv3 servers to act as data servers. Also lowercase Read/Write/Data/MetaData to match the prevailing style. Discussed with: rmacklem Modified: head/en_US.ISO8859-1/htdocs/news/status/report-2017-01-2017-03.xml Modified: head/en_US.ISO8859-1/htdocs/news/status/report-2017-01-2017-03.xml ============================================================================== --- head/en_US.ISO8859-1/htdocs/news/status/report-2017-01-2017-03.xml Wed May 3 20:44:44 2017 (r50228) +++ head/en_US.ISO8859-1/htdocs/news/status/report-2017-01-2017-03.xml Thu May 4 03:51:12 2017 (r50229) @@ -261,20 +261,21 @@ <body> <p>Parallel NFS (pNFS) is an extension to the NFSv4 protocol that allows for file accesses within a single logical mount to be - performed against multiple NFS file servers, "in parallel". - Thus, a pNFS server separats the MetaData operations from the - Data operations (Read/Write/Setattr of size N), letting - existing NFSv3 servers be used for the bulk data storage - while still taking advantage of NFSv4 protocol enhancements, - among other things.</p> + performed against multiple file servers, with the potential + for data access to occur in parallel. The pNFS + "layout" specifies how the division occurs, with + metadata operations occuring against the main server, and + bulk data operations (read/write/setattr/etc.) occuring via + a layout-specific scheme between the client and data + servers.</p> <p>My first attempt at a pNFS server using GlusterFS was a dud. It worked, but performance was so poor that it was not usable. This attempt that I call Plan B, only uses &os;, - with one &os; server handling the MetaData operations and K - &os; servers configured to serve Data. An NFSv4.1 client + with one &os; server handling the metadata operations and K + &os; servers configured to serve data. An NFSv4.1 client that supports the pNFS File Layout will be able to - Read/Write to the Data servers directly, spreading out the + read/write to the data servers directly, spreading out the RPC load and allowing growth beyond that of what a single &os; NFS server could achieve.</p>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201705040351.v443pDBH036836>