[infinispan-dev] ClockService

[Bela Ban]

On 1/30/13 1:45 AM, Manik Surtani wrote:
>
> On 29 Jan 2013, at 18:27, Mircea Markus <mmarkus at redhat.com 
> <mailto:mmarkus at redhat.com>> wrote:
>
>> I guess Bela's question was about why System. currentTimeMillis() 
>> would trigger a context switch, I'm also curious about that one :-)
>
> On its own it doesn't. But if we have a cached clock with a scheduled 
> thread updating that cached value periodically, that happens in an 
> async thread.

Just make sure you have 7 long variables between the cached var and the 
next var, so they're on different cache lines... :-) This is what the 
Disruptor does. What if a cache line is > 64 bytes ?

The Oracle article wasn't good IMO, so what's the problem again ? The 
slowness of System.currentTimsMillis()/nanoTime() ? Does this have an 
impact on overall Infinispan performance at all ?

-- 
Bela Ban, JGroups lead (http://www.jgroups.org)