We have an application where we have limited room (19" rack with room for 5 1U servers). The only constraints are that: 1) the operating system must be a flavor of Linux 2) video data is pouring in over 6 IEEE 1394 connections (3 pairs)
We must operate on this data as fast as possible. Each 1394 pair of connections must be available together. We are currently running a test with one 1394 pair on a dual Xenon P4 at 2.4 GHz primarily because the math utilities we are using are optimized for the P4. When we upgrade, should we just increase the clock speed to 3.2 GHz (what motherboard?) or does AMD have something better? Money is not too much of an impediment.
Thanks!
You are about to answer a thread that has been inactive for more than 6 months. If you still wish to proceed, please ensure that your posting is original and does not duplicate or overlap any prior responses to this thread.