>But even so: the ROT is(was) if an instruction takes a second, an I/O takes 2 >weeks. >Could 1 ms I/O responsetime even come near the break-even piont?
Considering that there is more than one instruction in the I/O path. Considering that VIO's CPU requirement is larger. Considering buffering. The answer is a definite MAYBE. Lately, with today's I/O response, I have found no benefit with VIO. - Too busy driving to stop for gas! ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
