Re: [boost] A fast-pathed rw-mutex algorithm for Boost...

From: Howard Hinnant (hinnant_at_[hidden])
Date: 2006-10-24 22:12:20

On Oct 24, 2006, at 9:29 PM, Peter Dimov wrote:

> Howard Hinnant wrote:
>
>> I don't know how to quantify it, but there is nothing quite like the
>> warm fuzzy of seeing this on your screen:
>>
>> writ thread 1 at 62%
>> writ thread 2 at 63%
>> writ thread 3 at 59%
>> writ thread 4 at 58%
>> read thread 5 at 62%
>> read thread 6 at 55%
>> read thread 7 at 64%
>> read thread 8 at 75%
>
> You are right in principle but the above isn't a good result for a
> multi-core. The read threads should have an advantage proportional  
> to the
> number of cores since they can run in parallel. I'll try to  
> embelish my test
> with something that measures starvation.

<nod> Indeed the above snapshot was taken from a single processor  
machine.  My memory tells me I've had similar experience with a dual  
core machine, maybe there was the advantage you speak of for readers,  
but I'm unable to verify that at the moment.

Much thanks for your efforts in this area.

-Howard