From 31046c5f570fd752d8bd3fde3d2bbd6210520904 Mon Sep 17 00:00:00 2001 From: =?utf8?q?Ale=C5=A1=20K=C5=99enek?= Date: Mon, 20 Oct 2008 11:31:44 +0000 Subject: [PATCH] finished deployment scenarios --- org.glite.lb.doc/src/LBAG-Introduction.tex | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-) diff --git a/org.glite.lb.doc/src/LBAG-Introduction.tex b/org.glite.lb.doc/src/LBAG-Introduction.tex index bbdfacd..736657d 100644 --- a/org.glite.lb.doc/src/LBAG-Introduction.tex +++ b/org.glite.lb.doc/src/LBAG-Introduction.tex @@ -33,11 +33,12 @@ and delivering the events to \LB servers where users can query for them \subsection{Deployment scenarios} -\TODO{salvet} +%\TODO{salvet} -\TODO{jakou zatez ktery snese} +%\TODO{jakou zatez ktery snese} \subsubsection{Standalone \LB server} +\label{deploy-stand} This is the recommended standard production deployment. @@ -48,7 +49,7 @@ from the WMS, not affecting its performance directly. In this setup the full reported performance is achieved, currently up to several hundreds thousands jobs per day, with the goal -of one million, see~\cite{lbtp}). +of one million, see~\cite{lbtp}. Further performance can be gained with clustering $M$ WMS's and $N$ \LB's while configuring all WMS's to distribute jobs to the \LB's uniformly. @@ -59,6 +60,7 @@ and number of evaluated notifications, and it must be determined empirically for each specific installation. \subsubsection{Hybrid \LB server-proxy} +\label{deploy-hybrid} \LB server runs on the WMS node, in combined server-proxy mode, serving both user queries and supporting WMS. @@ -76,6 +78,15 @@ The functionality is available for \LBnew only. \subsubsection{\LB server on WMS node} -% dedictvi pre-proxy , chybny deployment +\textbf{This setup is obsolete and very inefficient, hence discouraged.} -% Highly obsolete and inefficient \dots +Ancient \LB versions, where \LB proxy was not available yet, +used to be frequently installed with \LB server on the WMS machine. +With the introduction of \LB proxy it makes little sense anymore +but, unfortunately, this setup still persists at some sites. +It's consequence is doubling both CPU and disk load, yielding observable +performance degradation. + +Large production sites should consider standalone \LB server (Sect.~\ref{deploy-stand}) +instead, while for smaller sites new hybrid setup (Sect.~\ref{deploy-hybrid}) may +be more appropriate. -- 1.8.2.3