Caldarale, Charles R wrote:
From: Gregor Schneider [mailto:rc4...@googlemail.com]
Subject: Re: Performance of APR
However, I haven't seen any benchmarks comparing NIO with APR,
especially when it comes to SSL
SSL handling is the same for the NIO and the traditional Java connector, so APR
wins there. Filip has reported that the NIO connector is slightly slower than
the traditional one in terms of overall throughput (likely due to extra thread
context switching), but has much better scalability.
NIO is a bit smarter when it comes to SSL, because NIO can keep
connections alive, during higher concurrency, even NIO will outperform
the regular connector by magnitude. When the number of connections is
less than maxThreads, then the performance is the same.
NIO also does a non blocking hand shake, so it can also handshake (which
on SSL is very expensive) on a higher number of connections than both
APR and the regular connector.
With that said, I can create a test case where each connector
outperforms the others. There isn't one rule that one is better than the
other. When it comes to maturity/stability, the older blocking connector
has a few more years than both APR and NIO. The longer the code has been
around, the more use cases are there.
As Mladen pointed out, APR's use of OpenSSL is superior to that of JSSE.
The hybrid model of APR (blocking read/write with polling capabilities)
is a very good fit for the servlet spec which defines IO as blocking anyway.
The NIO connector is slowly maturing, still a few more bugs that will be
fixed in 6.0.19, makes it a very good alternative too.
Filip
- Chuck
THIS COMMUNICATION MAY CONTAIN CONFIDENTIAL AND/OR OTHERWISE PROPRIETARY
MATERIAL and is thus for use only by the intended recipient. If you received
this in error, please contact the sender and delete the e-mail and its
attachments from all computers.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org
