Concurrency models (was: Timer)

[Alan Kennedy]

[Alan Kennedy]
>> While reading about the C10K problem, I came across a great academic
>> paper about architecting high-performance network servers using a
>> combination of event-driven and threaded architectures, with a
>> specially defined IPC mechanism to tie it all together. It involved
>> the nice concept of "back pressure" (my words), whereby when a
>> component became saturated it could notify the event-scheduler, and
>> cut-off the sources of requests.
>>
>> I'd love to find that paper again.

[Michael Sparks]
> Was it any of the papers listed here ?
>    * http://citeseer.nj.nec.com/context/17325/298025
> 
> http://citeseer.nj.nec.com/532228.html perhaps? (Though the overcommital
> handling sounds more like a side effect rather than explicit notifcation
> to the scheduler)

Not that particular paper, but a shorter one by the same author, Matt
Welsh. Here it is

A Design Framework for Highly Concurrent Systems
http://www.eecs.harvard.edu/~mdw/papers/events.pdf

I see that the paper you selected is Matt Welsh's PhD thesis, and goes
into the subject in a lot more detail than the paper I had originally
read.

An Architecture for Highly Concurrent, Well-Conditioned Internet
Services
http://www.eecs.harvard.edu/~mdw/papers/mdw-phdthesis.pdf

I look forward to reading Dr. Welsh's thesis, and his other
publications.

http://www.eecs.harvard.edu/~mdw/pubs.html

I won't comment further on "over committal" or "back pressure" until
I've gone through the papers again in some detail.

kind regards,

-- 
alan kennedy
-----------------------------------------------------
check http headers here: http://xhaus.com/headers
email alan:              http://xhaus.com/mailto/alan