While the new Gojira supercomputer is not more powerful than the University of Tokyo's Oakleaf-FX at 1 petaflop of performance if you look at it from a price to performance ratio the $5,500 Gojira is more than impressive. It has a peak theoretical performance of 729 teraflops by using over 71,000 Ivy Bridge cores across several Amazon Web Service regions and providing the equivalent of 70.75 years of compute time. The cluster was built in an incredibly short time, going from zero to 50,000 cores in 23 minutes and hitting the peak after 60 minutes. You won't be playing AC Unity on it any time soon but if you want to rapidly test virtual prototypes these guys can do it for an insanely low price. Catch more at The Register and ZDNet, the Cycle Computing page seems to be down for the moment.
"Cycle Computing has helped hard drive giant Western Digital shove a month's worth of simulations into eight hours on Amazon cores."
Here is some more Tech News from around the web:
- Mac OS X Yosemite disables third-party SSD driver support @ The Inquirer
- Digitimes Research: Lenovo, Asustek to launch US$149 Chromebook
- DAY ZERO, and COUNTING: EVIL 'UNICORN' all-Windows vuln – are YOU patched? @ The Register
- FLASH better than DISK for archiving, say academics. Are they stark, raving mad? @ The Register
Gojira is not a
Gojira is not a supercomputer, and $5500, is for the rental on those cloud cores. The Gojira is the size of the run, and if you need the cores they are there for a rental price, there is even a service in the EU that uses Power8’s and bet that Amazon will be buying some of Tyan’s Licensed from OpenPower and fabbed by globalfoundries Power8’s. or Amazon will be doing the same thing as Google, and having some custom fabbed licensed power8 based server chips, and motherboards made for those big cloud jobs that are being spun up on their supercomputers as a service. These types of services will demand the lowest possible cost, and having server chips made at the cost of a license, fabrication, and a much smaller amount of engineering, relative to the cost of developing a server chip from scratch, will be a great savings to the providers of the service.
Nvidia is going to make billions providing its GPU accelerators to the nascent licensed power8 based server market, starting with Tyan’s, and others’ server SKUs in 2015. Those Bernoulli equations eat up lots of core time, to simulate a new read/write head’s design, as it rides on that layer of fluid(gas in this case). Even the newly announced PowerVR 7 GPU designs are offering a 64 bit FP option, for some ARM, or MIPS, GPU acceleration, on low power sipping computing clusters that are not x86 based.
Chipzilla, is going to have some Gojira sized competition in the server/HPC/supercomputing as a service market, come 2015, and beyond.