Re: [SMARTY] yet another benchmark
| From: | Steve Manes | Date: | Fri, 14 Jun 2002 17:21:50 +0000 |
| Subject: | Re: [SMARTY] yet another benchmark | ||
| References: | 1 | Groups: | php.smarty.general |
| Request: | Send a blank email to php-smarty-general+get-4864@lists.php.net to get a copy of this message | ||
At 09:30 AM 6/14/02 -0500, Monte Ohrt wrote:
Well that's not really the point. First of all, you are correct in saying that 2.7 seconds vs 3 seconds doesn't make much difference when you look at one page load. But, when you look at 1 million page loads, it _does_ make a difference in system load and performance over time.Absolutely true. Our PHP sites at CCI (www.blackplanet.com, et al) deliver 800 million page views monthly with estimates of 1.4 billion/month in 2003. That doesn't tell the whole story because our traffic tends to peak on blocks of time during the day (lunchtime, early evening). Rendering speed isn't only important to us but also resource consumption. We're running our 142 web servers near the WFO so a 10% decrease in speed or a 10% increase in overhead means, simplistically speaking, that 14+ web servers need to be added to the clusters to compensate for it... or about fifty grand, not including installation, rack fees and the add'l hardware to support them. -----------------------[ http://www.magpie.com ]-------=o&>o-------
Steve Manes "When in doubt, get a bigger hammer"Brooklyn, USA