KDC performance test - lookaside cache impact, testing framework

[Greg Hudson]

On 06/18/2012 10:47 AM, Nico Williams wrote:
> What's the point of the lookaside cache?  A: To avoid re-computing
> replies to retransmited requests, which might occur in the event that
> the KDC is much too busy.

Retransmitted requests can also happen if a KDC needs a few seconds to 
process a request, e.g. to contact a slow OTP server.  Since we're 
trying to support that case, I'm (so far) more interested in making the 
lookaside cache efficient than dropping it.

And, of course, they can happen when a packet is lost.  Usually that's 
rare, so maybe that's not important.

> Also, how would the lookaside cache work in the multi-processor option
> case?  For a multi-threaded KDC you'd want to make sure to lock only
> lookaside table buckets so as to decrease contention.

If we do a multi-threaded KDC, it might make sense to move to the 
single-element bucket design I mentioned in my last message.  The 
lookaside cache is a rather small piece of code; it isn't really 
necessary to make it accomodate future design possibilities.