Re: Squid memory footprint (was: MemPools rewrite)

From: Henrik Nordstrom <hno@dont-contact.us>
Date: Fri, 03 Nov 2000 00:03:15 +0100

Chemolli Francesco (USI) wrote:

> I don't know if and how it could be done,

In the reiserfs-raw approach it was done by not having a swap.state file
or "permanent" StoreEntry structures. Instead the objects were hashed
directly onto the filesystem meta structure tree (sort of the inode
layer in ext2/UFS, but with a much larger and dynamic namespace).

> Besides, we could save up tons of startup time for swap.state
> parsing and checks: just mark the file clean, compute an MD5
> checksum and store it somewhere This could allow for
> simultaneous access by two processes (synchronous double-checking).
> No heap fragmentation problems either, reallocs are easy,
> disk space is cheap.

And if you don't even have the file then the start up is even quicker.
Simply pick up where you was, doing what you did.

/Henrik
Received on Thu Nov 02 2000 - 16:42:05 MST

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:12:54 MST