Zappos - the leading eCommerce site for shoes and apparel - recently talked about their best practices of delivering WOW Performance to their customers. Zappos re-architected their web-site and went from Perl to Enterprise Java as the need to scale and perform was driven by explosive business growth and performance problems in their old architecture. Performance is the key to business success for every eCommerce site. Zappos picked an Application Performance Management Solution that enabled them to deliver their #1 Core Value to their Clients: "Deliver WOW through Service" Read more about why Zappos decided to re-architect their site, how they transformed their load-testing process and what now enables them to deliver WOW Performance to their end-users. Edit: first link fixed.
- Posted by: Andreas Grabner
- Posted on: March 02 2010 01:29 EST
- From perl to java ? by shawn spencer on March 02 2010 01:59 EST
- Re: Best Practices from Zappos to deliver WOW Performance by amar n on March 02 2010 04:37 EST
- Re: Best Practices from Zappos to deliver better performance by Cameron Purdy on March 02 2010 15:29 EST
- Re: Best Practices from Zappos to deliver better performance by William Louth on March 03 2010 16:31 EST
Hey i saw another article on TSS last week which said how Perl, php etc is making it big time into enterprise space.... hmm. is this ironic that a major wb site just moved from perl to java ????? So with all due respect to non - OO programmers PHP and perl and all that scriping junk is good for small , silly wbe sites.
Not able to access best practices link...
here is the link to actual webinar - Zappos Webinar
Judging purely by the links, I'm going to guess that this has something to do with Dynatrace? If so, perhaps it would be a good idea to state that up front? Peace, Cameron Purdy | Oracle Coherence
As an expert in this field I hold the view that call path tracing (strangely referred to as transaction tracing) is not scalable in production at runtime or analysis time so I would be interested in hearing from the Zappos team their thoughts on this as in my view you have to severely limit your instrumentation plan (limited coverage) with such an approach unless of course the performance of your application is terrible effectively hiding the excessive overhead of call path tracing. Note I have built contextual trace technology before moving onto much more scalable approaches and I have requested on many occasions shootouts with trace based vendors - everyone declined. http://williamlouth.wordpress.com/2010/03/02/does-transaction-tracing-scale-analysis/ http://williamlouth.wordpress.com/2010/02/17/dont-trace-calls-record-metering-changes/