On Friday 25 July 2008 07:28:19, M. David Peterson wrote: > I find the following verbiage at http://cherokee-project.com a little > strange: [...] > "After that there's really not much left to do to improve a High > Performance web server such as Cherokee." > > The above places a specific shelf-life on the Cherokee project as far as > innovation is concerned. While I recognize the reasoning, placing the > hard drive at the center of I/O operations seems to neglect the fact that > we're well into the transition between magnetic media and RAM-based > distributed storage (e.g. memcached), using the hard drive as nothing more > than a long term storage system. AKA, Magnetic Drives are the new Tape.
I wrote that some months ago, and I couldn't agree more with what you said. I didn't even consider it, but you are totally right. > Could I suggest a rewrite of the above to something similar to: > >> The speed at which any web server can serve requests for content is > >> both directly tied to and limited by the I/O speed of underlying > >> hardware and operating system. In this regard, a web servers > >> performance is measured by the latency incurred after an I/O request to > >> the underlying system has completed. Yes you could. I absolutely love it. In fact, I'm uploading the changes right now. I have to admit it really does make a difference being an English native speaker. Good work :) -- [EMAIL PROTECTED] http://unixwars.com _______________________________________________ Cherokee mailing list [email protected] http://lists.octality.com/listinfo/cherokee
