sql cache method

[gsdcypher]

I am trying to use the best method for my site. In the description below, how many products/categories would justify medium and how many large?


None. SQL queries are not cached at all. If you have very few products/categories you might actually find this gives the best speed for your site.

Database. SQL queries are cached to a database table. Sounds strange but this might provide a speed increase for sites with medium numbers of products/categories.

File. SQL Queries are cached to your server's hard disk. For this to work you must ensure that the directory where queries are cached to is writeable by the web server. This method is probably most suitable for sites with a large number of products/categories.

I will have approx 30 cats and 375 - 400 products

thanks.

[DrByte]

1. Where did you get those descriptions from?

2. What hosting company are you with?

3. If your mysql server is tuned poorly and runs on the slow side, don't use database-based caching.

4. If your webserver itself is on the slow side, use No caching, because the file-based caching will just create unnecessary extra load.


In short ... trial and error is the best approach ... there's no way I can tell you whether one is better than the other without knowing exactly how your webserver and database server are configured, tuned, and operating.

When in doubt, choose None.

[gsdcypher]

Thanks DrByte.

Those descriptions are from the Zen Cart installation process. I was just trying figure out what constitutes med and large.

I am upgrading from 1.3.7 to 1.3.8a and moving over to the zen recommended blue host. So if you have any tips for blue host configuration I would appreciate it.

I chose file during the installation process... How would I change that to database?

Thanks.

[DrByte]

I chose file during the installation process... How would I change that to database?

You can edit your configure.php files and change the SQL_CACHE_METHOD setting there. Be sure to do both.

I'm inclined to suggest "none" or "database" over "file", simply due to garbage-collection algorithms.