DatagramChannel performance issue

[Alan Bateman]

On 03/09/2018 16:26, Martin Thompson wrote:
> Is that a direct ByteBuffer?
A direct buffer or a region of memory allocated with Unsafe will do too. 
The former might be easier as it will be deallocated via the cleaner 
mechanism.


> :
> The posted results are for Linux (Ubuntu 18.04 LTS). I've also tested
> on Windows 10 and the ratios are similar.
>
> I tested with -Djava.net.preferIPv4Stack=true. The single source case
> is similar. The dual source case does see a reduction in latency of a
> few hundred nanoseconds.
There are a couple of additional field accesses in the JNI code for IPv6 
addresses so this probably explains that.

-Alan