[Flent-users] Query regarding aggregate netperf data points

[Flent-users] Query regarding aggregate netperf data points

[Sourabh jain]

[-- Attachment #1.1: Type: text/plain, Size: 779 bytes --]

Hello All,

I have a query regaring data aggregation functionality of Flent. In the
tests like tcp_8up and tcp_8down the interim value generated by different
instances of netperf are varying too much (between 0.2 to 0.58) though
interval is same for all i.e. 0.2 as compare to tcp_2up or tcp_2down
examples.

The point is system lag due to less core and process scheduling, the data
point collected during test may have different timestamps then how
aggregation is performed.

My question is how Flent handle this discrete time value of netperf to plot
total bandwidth graph?

Is Flent has some threshold that says that result for all process will be
consider for x point if they come within assumed threshold or there is some
other mechanism.


Thanks and Regards,
Sourabh Jain

[-- Attachment #1.2: Type: text/html, Size: 954 bytes --]

[-- Attachment #2: Type: text/plain, Size: 151 bytes --]

_______________________________________________
Flent-users mailing list
Flent-users@flent.org
http://flent.org/mailman/listinfo/flent-users_flent.org