tcmay@got.net wrote:
Not in cracking "truly large" problems by brute force. Even if each of the million processors is capable of 100 MIPS (which is unlikely, given the PIM approach and the fine-granularity, few-bit-or-less word size, etc.), this is only 10^8 MIPS. For problems that (for instance) 10^75 machines would have to spend 10^10 years on, not even a drop in an ocean.
The problem is there are still people and organizations that use 512-bit RSA keys. The DOE recentedly awarded Intel a contract to build a computer with 9072 Pentium Pro processors. I doubt that it will be used for factoring keys, but if it were, it will be able to factor a 512-bit number in a matter of months. The boundary delimiting "truly large" problems and merely extremely expensive ones inches up all the time. Less than a decade ago people thought factoring RSA-129 was a "truly large" problem. Wei Dai