<div>Ian</div>
<div> </div>
<div>Thanks for the lesson, I probably should read up on the architecture rather than assuming it's a Cisco in a different coloured box, maybe some formal training rather than the pdf's I found and our SE lab.</div>
<div> </div>
<div>Thanks</div>
<div>Rick<br><br></div>
<div class="gmail_quote">On 1 March 2011 20:11, Ian Calderbank <span dir="ltr"><<a href="mailto:ian@calderbankconsulting.co.uk">ian@calderbankconsulting.co.uk</a>></span> wrote:<br>
<blockquote style="BORDER-LEFT: #ccc 1px solid; MARGIN: 0px 0px 0px 0.8ex; PADDING-LEFT: 1ex" class="gmail_quote">Redback does forwarding using the PPA, end of story. (apart from some very<br>special exceptions).<br><br>
Process-switching concept doesn't exist outside of cisco. Seems your<br>friendly local redback SE isn't giving you the basic architecture lesson?<br><br><history><br>Ye olde Cisco way of forwarding packets = process switching, fast-switching<br>
yada yada<br>Cisco invented Cef then better DCEF = Distributed forwarding using line card<br>processors (VIP's on 7500's in the mid 90s) to perform all forwarding<br>actions (except those that CEF didn't support.. which was the issue).<br>
<br>Juniper did it properly with a box with a "strict separation of control and<br>forwarding planes" architecture (enter stage left to thunderous applause the<br>M series). impossible for a transit packet to touch the control processor.<br>
The forwarding processor (either a shared forwarding blade, or a chip on a<br>line card) does all the forwarding work. Control cpu cruises at 1% if the<br>network is converged/stable, regardless of the forwarding load.<br>
<br>Redback followed suit with the SE. Did a bit more work on modularising the<br>control plane than J, different backplane design, different asics of course,<br>but still the fundamental separation of forwarding and control processors<br>
and data paths, transit packet unable to touch the RP.<br></history><br><br>As david mentions there are corner case applications where you might<br>deliberately redirect the packet to the RP such as http url redirect .<br>
Session Border controller does the same for SIP I would expect.<br><br>For a regular forwarding policy, it just adds another link in the PPA's<br>chain of actions (called a PED graph) that are to be applied to that<br>
circuit. Each PED costs you a bit of resource (mem , cpu) on the PPA. But so<br>long as your PPA's are not overcooked - don't worry about it.<br><br>cheers<br>Ian<br><br>-----Original Message-----<br>From: <a href="mailto:redback-nsp-bounces@puck.nether.net">redback-nsp-bounces@puck.nether.net</a><br>
[mailto:<a href="mailto:redback-nsp-bounces@puck.nether.net">redback-nsp-bounces@puck.nether.net</a>] On Behalf Of<br><a href="mailto:redback-nsp-request@puck.nether.net">redback-nsp-request@puck.nether.net</a><br>Sent: 28 February 2011 17:00<br>
To: <a href="mailto:redback-nsp@puck.nether.net">redback-nsp@puck.nether.net</a><br>Subject: redback-nsp Digest, Vol 38, Issue 13<br><br>Send redback-nsp mailing list submissions to<br> <a href="mailto:redback-nsp@puck.nether.net">redback-nsp@puck.nether.net</a><br>
<br>To subscribe or unsubscribe via the World Wide Web, visit<br> <a href="https://puck.nether.net/mailman/listinfo/redback-nsp" target="_blank">https://puck.nether.net/mailman/listinfo/redback-nsp</a><br>or, via email, send a message with subject or body 'help' to<br>
<a href="mailto:redback-nsp-request@puck.nether.net">redback-nsp-request@puck.nether.net</a><br><br>You can reach the person managing the list at<br> <a href="mailto:redback-nsp-owner@puck.nether.net">redback-nsp-owner@puck.nether.net</a><br>
<br>When replying, please edit your Subject line so it is more specific than<br>"Re: Contents of redback-nsp digest..."<br><br><br>Today's Topics:<br><br> 1. forwarding policy (Richard Clayton)<br> 2. Re: forwarding policy (<a href="mailto:david.freedman@uk.clara.net">david.freedman@uk.clara.net</a>)<br>
<br><br>----------------------------------------------------------------------<br><br>Message: 1<br>Date: Sun, 27 Feb 2011 22:05:43 +0000<br>From: Richard Clayton <<a href="mailto:sledge121@gmail.com">sledge121@gmail.com</a>><br>
To: <a href="mailto:redback-nsp@puck.nether.net">redback-nsp@puck.nether.net</a><br>Subject: [rbak-nsp] forwarding policy<br>Message-ID:<br> <AANLkTinwxkUwkiStm6PuGt=xdPaf1TjPNeeyid7f=<a href="mailto:K-c@mail.gmail.com">K-c@mail.gmail.com</a>><br>
Content-Type: text/plain; charset="iso-8859-1"<br>
<div class="im"><br>Am I correct in thinking that a Redback forwarding policy is process<br></div>
<div class="im">switched, the same as Cisco policy-based routing.<br><br>Thanks<br>Rick<br></div>-------------- next part --------------<br>An HTML attachment was scrubbed...<br>URL:<br><<a href="https://puck.nether.net/pipermail/redback-nsp/attachments/20110227/a50d24e4/attachment-0001.html" target="_blank">https://puck.nether.net/pipermail/redback-nsp/attachments/20110227/a50d24e4<br>
/attachment-0001.html</a>><br><br>------------------------------<br><br>Message: 2<br>Date: Sun, 27 Feb 2011 22:10:28 +0000 (GMT)<br>From: <a href="mailto:david.freedman@uk.clara.net">david.freedman@uk.clara.net</a><br>
To: <a href="mailto:sledge121@gmail.com">sledge121@gmail.com</a><br>Cc: <a href="mailto:redback-nsp@puck.nether.net">redback-nsp@puck.nether.net</a><br>Subject: Re: [rbak-nsp] forwarding policy<br>Message-ID:<br> <<a href="mailto:Pine.BSF.4.58.1102272206340.31883@singularity.convergence.cx">Pine.BSF.4.58.1102272206340.31883@singularity.convergence.cx</a>><br>
Content-Type: TEXT/PLAIN; charset=US-ASCII<br>
<div>
<div></div>
<div class="h5"><br>>Am I correct in thinking that a Redback forwarding policy is process<br>>switched<br><br>Not quite sure what you mean here, my understanding is that unless you are<br>using XCRP services (such as the http server), then the forwarding actions<br>
occur in hardware, on the PPA,<br><br><br>[thingy]redback# show forward policy foo<br><br>Policy-Name Type Grid Qs Slots Ports Bound DnLd<br>Status<br>foo forward 14 0 1 4 in<br>
<br>Slot#: 1 2 3 4 5 6 7 8 9 10 11 12 13 14<br>iPPA dnld:<br>ePPA dnld:<br>iPPA ports: 4<br>ePPA ports:<br><br>Class-Name Action Mode IP-Addr/Option Bound Int,msec<br>Output-Name<br>
redirect-class redir local<br>drop-class drop<br><br>Total policy map: 1<br><br>I believe this is possible because these are FPGA , though I'm sure there<br>are some internal limits to how nested operations can be.<br>
<br>Dave.<br><br><br><br></div></div>------------------------------<br><br>_______________________________________________<br>redback-nsp mailing list<br><a href="mailto:redback-nsp@puck.nether.net">redback-nsp@puck.nether.net</a><br>
<a href="https://puck.nether.net/mailman/listinfo/redback-nsp" target="_blank">https://puck.nether.net/mailman/listinfo/redback-nsp</a><br><br><br>End of redback-nsp Digest, Vol 38, Issue 13<br>*******************************************<br>
<br>_______________________________________________<br>redback-nsp mailing list<br><a href="mailto:redback-nsp@puck.nether.net">redback-nsp@puck.nether.net</a><br><a href="https://puck.nether.net/mailman/listinfo/redback-nsp" target="_blank">https://puck.nether.net/mailman/listinfo/redback-nsp</a><br>
</blockquote></div><br>