Squid server optimizing information:

Statistics from Sat Mar 8 01:44:11 2003 to Fri Mar 14 18:08:23 2003

Squeezer is a tool for gathering different statistical information from Squid web cache server to tune it fine. The tool is destined basically for web cache operators, but can be obtained and used for any other purpose. The software has been written in PERL and as Squid itself and PERL - it's free. But you use it at your own risk! There are several sections that are disabled by default but can be enabled by editing the script and enabling them. The problems with having them all enabled are: it slows the processing WAY down, and it makes this report very large. If you don't use siblings, there are also a few sections that can be disabled by editing the script, these are "Sibling efficiency" and "By Peer Status"

General:

All values given to you by squeezer are related to the period, limited by the times described here.

• Start date is a time of the first analyzed service request
• End date is the time of the last analyzed service request
• Total time is the length of analyzed period
• kBytes/Hour is the average amount of data (in kilobytes) transferred per hour of analyzed period
• Average Object Size is the average size (in kB) of objects requested
• Hit Rate is a percentage of cache hits to total requests
• Bandwidth Savings Total is the total amount of kBytes read from the cache
• Bandwidth Savings Percent is the percentage of kBytes read from the cache to total kBytes transfered
• Average Speed Increase shows how much faster all the traffic is on average compared to potentially all-direct traffic

Output and it's interpretation:

The output contains several profiles of information with several different positions (categories) in each. Each category is characterized with several values which are described below:

• Cached Requests is the number of requests that were answered from the cache in each category
• Cached Req/s are the number of cached requests per second of time spent answering cached requests in each category
• Cached % of Reqs is a percentage of the number of cached requests to number of total requests in each category
• Cached Reqs Graph is a graphical representation of the Cached % of Reqs


• Requests is the number of service requests classified into a category
• Req/s are the number of requests that were served of this category in the amount of time that was used serving requests for the category
• Req % is the share of current category's requests in total number of requests
• Req Graph is a graphical representation of the Req %


• Cached Bytes is a total number of bytes retrieved from the cache in each category
• Cached Bytes/s are the number of cached bytes per second of time spent answering cached requests in each category
• Cached % of Bytes is a percentage of bytes retrieved from the cache to the total number of bytes in this category
• Cached Bytes Graph is a graphical representation of the Cached % of Bytes


• kBytes is the amount of traffic classified into category (in kilobytes)
• B/s is the speed (in bytes per second) of transfer in current category (in kilobytes)
• kB % is the share of current category's traffic in total traffic from this web cache
• kB Graph is a graphical representation of the kB %


• Time is the total amount of time spent serving requests in each category
• Time % is a percentage of time spend on each category's requests to the total amount of time spent serving requests
• Time Graph is a graphical representation of the Time %


• Largest Cached Item is the size (in bytes) of the largest item in each category that was cached
• Largest Item is the size (in bytes) of the largest item requested whether it was cached or not

Cache Statistics:

This is an overview of the cache, it shows whether it is doing you any good to have this server running.

• Cached_Local is the characteristics for fetches from local web cache server; it shows you the general information about the efficiency of the local server - how much traffic goes locally instead of from outside and how much faster that is compared to the direct fetches
• Cached_Other is an aggregate category for the fetches from the cooperating web cache servers - it gives you the overview of cooperation between you and your siblings/parents
• Cached_Total is the total traffic traffic fetched from all caches, local and all peers
• Direct fetches is the characteristics for direct fetches from the origin servers
• Total is the characteristics of all the traffic coming through this web cache server - no matter who satisfied the request: cached or direct


• kB Times to direct in this category shows how much all the traffic is on an average faster compared to potentially all-direct traffic; good for reporting and cache advertising purposes ;) this value is important - it tells you whether the cache speeds up your service or not!

Total Requests:

This table is the total of all requests served by this caching server.

By Caching Server:

This shows how many requests were served for each caching server.

By Client:

This shows how many requests were served for each client accessing the server.

I manually removed MANY lines of this output, probably 500 or so to make this a manageable size.

By Peer Status:

The basic stuff showing you advantages/disadvantages of using parent and sibling relationship. Use with caution! Remember that your parents and siblings could be available via lines of different speed and quality. Using those aggregates for analysis may cause false conclusions.
The strong point of this stuff is opportunity to compare cache digests versus ICP, if there are siblings communicating both ways in similar environment.
No idea how to use that more:( Any suggestions?

By cache result codes:

This table shows the characteristics for different cache result codes. The only useful information I have found is the difference between TCP_MEM_HIT (which are objects fetched from squid's RAM buffer) against TCP_HIT (which are fetched from disk buffer). Low value for TCP_HIT displays the need for dedicate more RAM for Squid and/or rearranging the cache_dir layout - spreading several cache directories over several disk/controllers or any further disk performance improvement.

By hours:

This shows cache statistics by hour of day. This is in the local time zone for the log file.

Refresh pattern efficiency:

This shows how many requests were served for each refresh pattern.

By size of request:

This shows cache statistics by size of request in bytes.

Sibling efficiency:

This table shows you the characteristics of the fetches from individual cooperating web cache servers. This one will help you to properly set up sibling relationships. The servers are sorted by the amount of transfer - the effective siblings/parents are higher in the table. Remember about some rules

• the slowest sibling will go probably down the table - it answers slowly for ICP queries and the chance of SIBLING HIT for him is reduced; it does mean that faster siblings should have bigger Xfer % - if it is not true something goes wrong - e.g. you probably try to access large (and having good ICP hit ratio), but slow (heavily loaded) server - this will slow you down! If it is parent, you can use weight options to balance it; if it is sibling you'll probably need to copy it's data to your cache (do not use proxy only option!) or even turn it off.
• the fastest siblings can use the proxy only option to avoid the replication of the data which is available fast e.g. from the same fast network (LAN or MAN) by the some backup/additional/private link; doing that you save your disk space and database entries for something, which could be fetched at any time with small cost

By Status Code:

This table shows detail of the status returned by different requests.