Trustin, Thank you for the reply. As 2.0 is not fully released yet I was not sure what the state was of this new filter and it sounds like it is still in the works.
For the time being we are looking into other options for traffic shaping outside our emulators that sit on top of MINA. FYI: I moved our stuff over to 2.0 a few weeks ago and the memory footprint went down dramatically as well as CPU load dropped some. We were already using heap buffers in 1.1 for both the socket and SSL but for 10k SSL client connections the overhead was still pretty high. The new version has been great. Cheers, -Scott -----Original Message----- From: Trustin Lee [mailto:[EMAIL PROTECTED] Sent: Monday, January 14, 2008 9:28 PM To: [email protected] Subject: Re: Traffic Shape Filter Hi Scott, TrafficShapingFilter is far from perfection. It will shape the traffic in the long term. It will not control the traffic inflows within a very short amount time. We need to figure out better way to shape traffic. Trustin On Jan 5, 2008 8:37 AM, Scott Peters <[EMAIL PROTECTED]> wrote: > Are there any examples of how to use the Traffic Shaping filter? I do > not see any unit test cases so it brings to question if this filter has > been tested or ready yet? > > With this code below I do not see any change in behavior that would show > the throughput is being slowed down. > > My goal is to allow us to simulate different network through puts [ > GPRS, EDGE, T1, etc ] that different parts of our product use. > > <code> > public static ScheduledExecutorService s_scheduler > Executors.newScheduledThreadPool(Runtime.getRuntime().availableProcessor > s()); > > Public sessionCreated(IoSession session) > { > ... > TrafficShapingFilter tsf = new TrafficShapingFilter(s_scheduler, > 100, 100); > session.getFilterChain().addFirst(TRAFFIC_SHAPING_FILTER, tsf); > ... > } > </code> > > > Thank you, > Scott Peters > -- what we call human nature is actually human habit -- http://gleamynode.net/ -- PGP Key ID: 0x0255ECA6
