In a recent note, Jim Harrison said: > Date: Fri, 6 Apr 2007 17:28:54 -0400 > > Not really. Since the *ix boxes max out at 30% utilization, you need > This statement has been repeated so often among mainframe partisans that it has come to be accepted without question. Can someone provide a source, attribution, citation, whatever?
What does it mean to "max out" Is it that regardless of how much work is available to be dispatched, the utilization never rises above 30%? If so, what other bottleneck is constraining throughput? Is it that at some very low utilization, delta, the throughput is epsilon, and if the relation were linear, at 100% utilization the throughput would be expected to be epsilon/delta, but as the workload increases, the utilization rises to 100%, but the throughput never rises above 0.3*epsilon/delta? If so, what form of overhead is consuming 70% of the "utilization"? For what sort of workload mix does the statement apply? I find it hard to believe that at least for certain sorts of problems such as cryptanalysis, meteorological simulations, and molecular orbital computation the CPU, at least, would not be fully utilized for productive computation with little overhead. Inquiring minds want to know. -- gil -- StorageTek INFORMATION made POWERFUL ---------------------------------------------------------------------- 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
