Date: Fri, 28 Mar 2014 09:56:08 -0500 (CDT) From: Bob Friesenhahn <bfriesen@simple.dallas.tx.us> To: Joar Jegleim <joar.jegleim@gmail.com> Cc: "freebsd-fs@freebsd.org" <freebsd-fs@freebsd.org> Subject: Re: zfs l2arc warmup Message-ID: <alpine.GSO.2.01.1403280927500.1735@freddy.simplesystems.org> In-Reply-To: <CAFfb-hr=wR6nxqL%2B4tn-y2eQEw4n_g7rZoK9rRLnm_Ldcm1TZQ@mail.gmail.com> References: <CAFfb-hpi20062%2BHCrSVhey1hVk9TAcOZAWgHSAP93RSov3sx4A@mail.gmail.com> <CALfReydi_29L5tVe1P-aiFnm_0T4JJt72Z1zKouuj8cjHLKhnw@mail.gmail.com> <CAFfb-hpZos5-d3xo8snU1aVER5u=dSFRx-B-oqjFRTkT83w0Kg@mail.gmail.com> <20140328005911.GA30665@neutralgood.org> <CAFfb-hr=wR6nxqL%2B4tn-y2eQEw4n_g7rZoK9rRLnm_Ldcm1TZQ@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 28 Mar 2014, Joar Jegleim wrote: > The server is actually far from any limit, in fact it has so 'little' > to do I've been a bit put off to figure out why our frontpage won't be > more snappy. The lack of "snappy" is likely to be an application problem rather than a server problem. Take care not to blame the server for an application design problem. You may be over-building your server when all that is actually needed is some simplification of the web content. The design of the application is important. The design of the content provided to the web client is important. Something I learned about recently which could be really helpful to you is there is a Firefox tool called "Web Developer Toolbar" which has a "Network" option. This option will show all files loaded for a given web page, including the time when the request was initiated, and when it completed. You may find that the apparent latency problem is not your server at all. You may find that there are many requests to servers not under your control. The performance problem is likely be due to the design of the content passed to the browser. For example, I just requested to initially load an application-generated page and I see that the base page loaded in 722ms and then there were two more subsequent loads in parallel requiring 335ms and 445ms, and then one more load subsequent to that requiring 262ms. The entire page load time was 1.7 seconds. The load time was dominated by the chain of dependencies. If I reload the page (request is now 'hot' on the server) then I see several of the response times substantially diminish, but some others remain virtually the same, resuling in a page load time of 1.13 seconds. >From what I have been seeing, web page load times often don't have much at all to do with the performance of the server. Bob -- Bob Friesenhahn bfriesen@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?alpine.GSO.2.01.1403280927500.1735>