Re: [PATCH] vhost: Make it more scalable by creating a vhost thread per device.

[Rick Jones <rick.jones2@hp.com>]

Sridhar Samudrala wrote:
> On Thu, 2010-04-08 at 17:14 -0700, Rick Jones wrote:
> 
>>>Here are the results with netperf TCP_STREAM 64K guest to host on a
>>>8-cpu Nehalem system.
>>
>>I presume you mean 8 core Nehalem-EP, or did you mean 8 processor Nehalem-EX?
> 
> 
> Yes. It is a 2 socket quad-core Nehalem. so i guess it is a 8 core
> Nehalem-EP.
> 
>>Don't get me wrong, I *like* the netperf 64K TCP_STREAM test, I lik it a lot!-) 
>>but I find it incomplete and also like to run things like single-instance TCP_RR 
>>and multiple-instance, multiple "transaction" (./configure --enable-burst) 
>>TCP_RR tests, particularly when concerned with "scaling" issues.
> 
> 
> Can we run multiple instance and multiple transaction tests with a
> single netperf commandline?

Do you count a shell for loop as a single command line?

> Is there any easy way to get consolidated throughput when a netserver on
> the host is servicing netperf clients from multiple guests?

I tend to use a script such as:

ftp://ftp.netperf.org/netperf/misc/runemomniagg2.sh

which presumes that netperf/netserver have been built with:

./configure --enable-omni --enable-burst ...

and uses the CSV output format of the omni tests.  When I want sums I then turn 
to a spreadsheet, or I suppose I could turn to awk etc.

The TCP_RR test can be flipped around request size for response size etc, so 
when I have a single sustem under test, I initiate the netperf commands on it, 
targetting netservers on the clients.  If I want inbound bulk throughput I use 
the TCP_MAERTS test rather than the TCP_STREAM test.

happy benchmarking,

rick jones