Re: Comments on the centroids paper

Simon E Spero <> Tue, 14 September 1993 10:46 UTC

Received: from by IETF.CNRI.Reston.VA.US id aa00778; 14 Sep 93 6:46 EDT
Received: from CNRI.RESTON.VA.US by IETF.CNRI.Reston.VA.US id ac00757; 14 Sep 93 6:46 EDT
Received: from by CNRI.Reston.VA.US id aa03746; 14 Sep 93 4:57 EDT
Received: by (4.1/UCD2.05) id AA18782; Tue, 14 Sep 93 00:00:25 PDT
Received: from by (4.1/UCD2.05) id AA18682; Mon, 13 Sep 93 23:57:51 PDT
Received: from [] by (5.61/UCD2.05) id AA00955; Mon, 13 Sep 93 23:52:49 -0700
Received: from by (5.64/10.1) id AA04188; Tue, 14 Sep 93 02:51:53 -0400
Received: from by tipper (4.1/SMI-4.1) id AA08730; Mon, 13 Sep 93 17:53:17 EDT
Message-Id: <9309132153.AA08730@tipper>
To: Chris Weider <>
Subject: Re: Comments on the centroids paper
In-Reply-To: Your message of "Mon, 13 Sep 93 15:13:59 EDT." <>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Mon, 13 Sep 1993 17:53:17 -0400
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: Simon E Spero <>

  Joann Ordille's work on Nomenklator contained a lot of measurements on the
effect of meta-data caching, as well as ordinary caching. Meta-data caching, 
which records whether or not records of a particular type are held in a certain
subtree can be quite productive.

Full record caching can also be quite productive, but only on a local level; 
imagine the case where somebody consistently uses user-friendly naming for
electronic mail. If no caching is done, then each message will require a 
directory lookup, which is still non-trivial. If even a small cache is used,
then the lookup becomes very cheap. 

Multi-level caching on large data-sets has been shown to yield rapidly 
diminishing returns. Peter Honeyman's "Your cache ain't nothing but trash",
in the Winter '90 usenix (same one as Archie), has some results for 
experiments using the AFS filesystem.

Where MLC becomes really useful is when the cache is placed at a near side of 
a high-latency or low capacity link. The savings from even a small number of 
hits may more than cover the cost of the cache.