On 3/23/07, Fernando Perez <[EMAIL PROTECTED]> wrote: > On 3/23/07, William Stein <[EMAIL PROTECTED]> wrote: > > 3. And most importantly, 16 cores!!! sage.math is now has twice the > > processing power as it had before. > > Out of curiosity: > > 1. Where did you buy sage.math? We may be looking for a similar box > in the future. Has it worked out OK for you hardware-wise?
Western Scientific, which is a San Diego company. I purchased from them because at the time Sun didn't have 16-core opteron machines available. I've had problems with sage.math that might be hardware related, but could have just been os related. The main problems were in December, and reformatting the drives ext3 instead of reiserfs fixed the problem. We had one crash since then, and the console showed a kernel crash, so it might not be a hardware problem. The machine also broke during shipping to Seattle, but that wasn't Western Scientific's fault. In fact, they paid for shipping it back and forth to San Diego and did fix it promptly. Overall I would be more comfortable recommending Sun than Western Scientific, but often Western Scientific has more cutting edge hardware at an impressive price. > 2. Did Ubuntu simply not support the 16 cores? The official Ubuntu kernels support at most 8 cores. All my attempts to build a 16-core kernel (both an ubuntu one or a vanilla one) failed. My attempts to upgrade ubuntu resulted in an unbootable kernel. Ubuntu isn't aiming at enterprise server-class systems. In contrast, the Debian netboot amd64 CD supported 16 cores during the initial boot, and after the install everything seems to work fine. William --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/ -~----------~----~----~----~------~----~------~--~---
