class3 - group jobs into classes that have"similar...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
Sheet1 Page 1 Preformance measures +system Oriented +-Throughput +-Utilitation +User oriented +-Response time +-Turnaround time MEASUREMENT mesure of time (continuous) mesure of events (discrete) we need to quantify "jobs" each job is a workload defined in terms of service demands and intensity a service demand is the work required by the system and intensity is how many or how often lets assume now that its not just a disk it also a CPU define as a tupple ID(cpu,disk)
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: group jobs into classes that have "similar properties" ie close to each other euclidian distance = sqrt((x1-x2)^2+(y1-y2)2) manhattan distance = abs(x2-x1) + abs(y2-y1) <-- easyer to use he says example woarkload=(CPUtime,datatransfer) T1 90 13000 | 13 T2 9213004 |13 T3 9112999 |13 T4 500 9000 | 9 T5 510 9005 | 9 T6 499 8998 | 9 T7 700 25500 |25 T8 70125501 |25 T9 70225499 |25 solution get the "Z score"...
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online