[Live-devel] optimising openRTSP

Anoop P.A. Anoop_P.A at pmc-sierra.com
Wed Oct 29 08:39:19 PDT 2008

if it is 16 * 20 * 32 KB means 10 MBps. I am having gigE network. 
My problem is openRTSP consuming lots of cpu so I am unable to run other s/w.
What is this "ReorderingPacketBuffer." function doing?. is this issue ( what i a seeing) have any relation with free mem ?

From: live-devel-bounces at ns.live555.com on behalf of Ross Finlayson
Sent: Wed 10/29/2008 6:26 PM
To: LIVE555 Streaming Media - development & use
Subject: Re: [Live-devel] optimising openRTSP

	I have profiled my system and found that around 8.7 % time is being spend in Reordering Packet Buffer <<OpenRTSP_profiling.log>>

	VMA      samples   %time   App-name      Symbol-name

	00465778 35862     8.7658  openRTSP ReorderingPacketBuffer::storePacket(BufferedPacket*)

	And around 3.4 % time is being spend in

	00463ac8 13521     3.3049  openRTSP                 seqNumLT(unsigned short, unsigned short)

	I am attaching entire oprofilelog for openRTSP log kindly help me with some suggestion to dig out the problem

I don't see a problem here.

It sounds like you just have high-bandwidth streams:

	PS. These logs captured while recording 16 mjpeg streams (each with 20 FPS and 32KB framesize)

That's almost 82 Mbps.  Are you sure you have sufficient network capacity for this (i.e., you don't just have 100 Mbps Ethernet, which would not be sufficient)?

If your problem is packet loss (but you have sufficient network capacity), then you need to increase your OS's buffer size for incoming network packets.  See <http://www.live555.com/liveMedia/faq.html#packet-loss>

Ross Finlayson
Live Networks, Inc.

More information about the live-devel mailing list