September 02, 2010
If you ever get the urge to build your own supercomputer, just take a look at what Chris Fenton has done. According to a blog post on his website, Fenton has managed to construct a working 1/10 scale version of the original Cray-1 machine. He's calling it the Cray-1A.
Built and sold by Cray Research in the 1970s, the Cray-1, with its iconic look and, at the time, unparalleled performance, jump-started the supercomputing industry. Today you can see the machines displayed in a handful of museums, but if you want to encounter a working version, you're going to have to drop by Fenton's house.
An electrical engineer by trade, Fenton is one of those guys who just gets a kick out of building stuff. His website chronicles his various homebrew electronic projects: GPS altimeters, IED detectors, gas guns -- you know, everyday gadgets you'd use around the house. Thus the need for the Cray-1 to round out the collection.
Thanks to a Cray-1 hardware reference manual located online, Fenton was able to reverse engineer the design onto an FPGA device using Verilog as the hardware description language. The final design was implemented on a Xilinx Spartan-3E 1600 development board. "This is basically the biggest FPGA you can buy that doesn’t cost thousands of dollars for a devkit," writes Fenton. "The Cray occupies about 75% of the logic resources, and all of the block RAM."
Not only did he reproduce a binary-compatible, cycle-accurate supercomputer, he also packaged it up to look like a miniature version of the original Cray-1, complete with doll-sized wraparound benching. The only thing missing is the software (oops). But not for lack of trying:
After searching the internet exhaustively, I contacted the Computer History Museum and they didn’t have any either. They also informed me that apparently SGI destroyed Cray’s old software archives before spinning them off again in the late 90’s. I filed a couple of FOIA requests with scary government agencies that also came up dry. I wound up e-mailing back and forth with a bunch of former Cray employees and also came up *mostly* dry. My current best hope is a guy I was able to track down that happened to own an 80 MB ‘disk pack’ from a Cray-1 Maintenance Control Unit (the Cray-1 was so complicated, it required a dedicated mini-computer just to boot it!), although it still remains to be seen if I’ll actually get a chance to try to recover it.
Fenton admits that without the software stack, the Cray-1A is not all that useful (unlike that gas gun!). Meanwhile he's rewriting the CAL assembler for the architecture. Once that's done, he could theoretically compose any software he wanted, although I imagine recoding the Fortran compiler and OS would chew up most of Fenton's remaining free time. Not that that would stop him.
Posted by Michael Feldman - September 02, 2010 @ 3:48 PM, Pacific Daylight Time
Michael Feldman is the editor of HPCwire.
No Recent Blog Comments
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?