Scaling Server Performance

An anonymous reader writes "When Ace's Hardware's article Hitchhiker's Guide to the Mainframe was posted on Slashdot, they got 590,000 hits and over 250,000 page requests during one day. This kind of traffic caused only a 21% average CPU load to their Java-based web server, which is powered by a single 550MHz UltraSparc-II CPU. In their newest article, Scaling Server Performance, Ace's Hardware explains how this was possible."

Comments

Maybe

[Jack Wagner]

This looks to be very Sparc specific to me. It's a well known fact that Intel proc's use a low grade silicon wafer for their on-board caches and Sparcs use the same commercial grade chip, which is why you'll always see better performance on a UltraSparc chip than an X86 chip. Plus Sparcs can handle bandwidth caching better because they follow the IEET network layer specs and X86 boxii don't.

Of course you would need to have two X86 boxii in order to handle a load like that anyways simply because of the horrendous implementation of Java on X86, but that's another story for another day.

Garth Brooks covers this in his famous book "The Mythical Man Month" where he proves in a controled lab environment that Java under X86 runs on the order of Olog(n) slower than it does on a RISC chip like an UltraSparc.

Warmest regards,
--Jack

Re:Ones that crashs

[/dev/trash]

50% mySQL is the cause.
50% 56k dialup as web hosting deal.