Re: Squid vs. NetCache

From: John A. Lauro <jlauro@dont-contact.us>
Date: Thu, 28 May 1998 13:28:18 EDT

> Here's some sample data in our squid vs. netcache bake-off.
>
> Machine ligarius:
> Squid running on Durango II (Alpha 164LX) 533MHz, 512MB ram, 4GB cache disk
> also running other services such as ftp,nntp, etc...
>
> Machine xpress24:
> NetApp NetCache 630 533MHz Alpha, 512MB ram, 109GB cache disk providing
> no other services
>

Unfortunately, you gave squid 4x the cache disk size, and the effect
is clear in the hit ratio, grately affecting the kB/sec. It makes the
comparison very difficult, as most of the stats are a reflection of
speed for hits vs. misses... Perhaps if you only told squid to use
1GB... Even then you have differences in drive speeds...

> Preliminary heat:
> xpress24 (netcache)
> status request % kByte % sec kB/sec
> --------------------------------- -------- ------ --------- ------ ---- -------
> HIT 4423 8.98 16774 2.73 2 2.45
>
> ligarius
> status request % kByte % sec kB/sec
> --------------------------------- -------- ------ --------- ------ ---- -------
> HIT 50377 26.32 206614 15.15 1 5.22

---------------------------------------------------------------------------
John Lauro email: lauro_j@msb.flint.umich.edu
University of Michigan - Flint jlauro@umich.edu
Information Technology Services
303 E. Kearsley St. phone: (810) 762-3123
Flint, MI 48502 fax: (810) 766-6805
Received on Thu May 28 1998 - 10:39:22 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:40:29 MST