Thanks for your suggestions, guys. Glad to see this forum is not yet dead.
@LZAntal I'm aware of these newbie mistakes but it's not this case. Debugging mode is disabled, logging is tuned down to INFO level. Even reducing to critical makes no difference. @Mike Hostetler Great answer, thanks.Nice site, btw. I'd bet Postgres server has an optimal setup and there would be only marginal improvement if any. I did tested performance even on powerful machine with postgresql running locally. On the same hosting is already running pure Django based CMS with ten times greater amount of items and is blazingly fast. Good point is about indexes to tables. I did checked for the most important models(Product, Category, ProductImage, Price, Discount ...) that came to mind are already created. However if you look into template tags, filters and some model's methods, you'll miss optimizations like select_related or prefetch_related at places where it would be reasonable and queries are needlessly doubled or even quadrupled. I see the only way to rewrite them myself, however it looks like a quite consuming task I've currently no resources to spend. As a side note, your linked site has a fair response times, however try to add much more details in category templates like basic price, discounted price, stock amount, link for direct buy, short description, brand name, promotion info etc. and you'll see what I am talking about. I had to reduce number of displayed products from 16 to 12 per page to get some so-so results. In a database is about 7 500 items only. See what workarounds like category menu generator were already accomplished - manual caching of category tree. Memcached seems to bring some improvements after extensive requests become handled but nothing drastical. @Arek I need to recheck if our provider does use or allow to use mentioned poolers. Thanks for the tips anyway ! -- You received this message because you are subscribed to the Google Groups "Satchmo users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/satchmo-users. For more options, visit https://groups.google.com/d/optout.
