24.12.2013 Views

TOPS-10 Monitor Calls Manual Volume 2 AA-K039D ... - Trailing-Edge

TOPS-10 Monitor Calls Manual Volume 2 AA-K039D ... - Trailing-Edge

TOPS-10 Monitor Calls Manual Volume 2 AA-K039D ... - Trailing-Edge

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

GET TAB TABLES<br />

Before the values in these locations can be used, however, it is<br />

important to understand the way that response time is calculated. The<br />

intention of counting response time is to understand the amount of<br />

time it takes to reach one of the three measured events (input<br />

operation, output operation, or quantum expiration) . The user may,<br />

for example, type a command. The response time is the lag between the<br />

time the monitor accepts the command and the time it takes to reach<br />

one of the three events. Note that this does not include the time the<br />

user spends typing the command. The response time is counted in<br />

jiffies (ticks).<br />

The Responses Subtable collects response times for the first event for<br />

a job when it begins running a program. If the first event for a<br />

program is to output a * as a command prompt, the amount of time<br />

between the time the "RUN program" command is accepted by the monitor<br />

and the time the monitor sends the * to the user's terminal is<br />

measured as the response time for that program. If a job runs a<br />

program that immediately begins input from the terminal, the response<br />

time for that action is the amount of time between the time the "RUN<br />

program" command is accepted by the monitor and the time the terminal<br />

input operation is attempted. The response times are accumulated for<br />

all the jobs as they are scheduled to run.<br />

23-67

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!