Re: [mpls] Concerns about ISD

Tony Li <tony.li@tony.li> Thu, 14 April 2022 03:28 UTC

Return-Path: <tony1athome@gmail.com>
X-Original-To: mpls@ietfa.amsl.com
Delivered-To: mpls@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6D6A93A0B6C for <mpls@ietfa.amsl.com>; Wed, 13 Apr 2022 20:28:57 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.51
X-Spam-Level:
X-Spam-Status: No, score=-1.51 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FORGED_FROMDOMAIN=0.248, FREEMAIL_FROM=0.001, HEADER_FROM_DIFFERENT_DOMAINS=0.248, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01, URIBL_BLOCKED=0.001] autolearn=no autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id dTgvRDHOs_xS for <mpls@ietfa.amsl.com>; Wed, 13 Apr 2022 20:28:53 -0700 (PDT)
Received: from mail-pj1-x1032.google.com (mail-pj1-x1032.google.com [IPv6:2607:f8b0:4864:20::1032]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1F4903A0B68 for <mpls@ietf.org>; Wed, 13 Apr 2022 20:28:53 -0700 (PDT)
Received: by mail-pj1-x1032.google.com with SMTP id md4so3916120pjb.4 for <mpls@ietf.org>; Wed, 13 Apr 2022 20:28:53 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=sender:from:message-id:mime-version:subject:date:in-reply-to:cc:to :references; bh=thWtwSFdrinrU0HPV/WrVBFpH+VSeZUR3ueOCJeAGfs=; b=BLdKB7YdT9LA0bJBshxKk6cxMrNTNIp6EKw8QxIYvvCfztPXMI3gWKny/oVO7eNxqx IgJq4maO5kc3AdqLERRh/k/lMOryZ2VCXBjpYu2CGK5JgVhhblAi2muksLVAi8vefwSv rgfd2KXXPz0kxHiaq1VjwN5hiqpwqWhts+eCS8/wxi+qGcocx8U5exZL1+BRlNREQlJz XggHBrDRbsGWtVHg1Uz1BlkNd7TGkDLzx6kg8FYI2PNvBP9eEIvfdzx3gInF9rwRR+HJ Q2y3B57nMiIGGRKVy8xEiIa/2n6IdlYDOyhuWKrV/Lq814Or5olmaPaLQ3mPdvMTYqhv QNpA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:sender:from:message-id:mime-version:subject:date :in-reply-to:cc:to:references; bh=thWtwSFdrinrU0HPV/WrVBFpH+VSeZUR3ueOCJeAGfs=; b=zk0MZG/zc2gSSEqRabZhXzvWOQOZCG/33Jak1cgn/DecO3aL3AM+4oD7QzDG5fxOCi 1DtuP7n8LZX4mqn+nJ6Oiqfh+YMBlue/bdqcgT+GJLZsNLhKuN9vRERQxQazEl7h3Zyd t7qH3pP3m5hVPK0Rl0sL9wWQ8zHu8Gt/kebLEdYXVbrkjWF5FSBSfXolULSlkNTzX9kE In9BDd3W+t4yXH6byZ+EDGCaLadFyf2tX58dSckd/23OrX3ljpYQCndFVwdwcoDohWZR eSa4uUUwWS2ll/bvYlb6wivBN3s6PCcvkXM2ZtW+8QGC5kP2Z0HYSp7wWIxLzdX+HYP7 J5oQ==
X-Gm-Message-State: AOAM5308vuiJlkQhB0yZBfj0uzIWT4mC/Jhp08G/Kb1JUQVpt4WGzcfv QaSmcfe2CNNFB/ZUH2HcyetM4arx42Q=
X-Google-Smtp-Source: ABdhPJx+lFCwWgjdZ4J+riBa1GTFF4nRMRZ0W7QzvkzofvxoyRGz4OFFJv6TktS6pm4uXLJFBL9xrw==
X-Received: by 2002:a17:902:c405:b0:158:63be:1776 with SMTP id k5-20020a170902c40500b0015863be1776mr17965486plk.107.1649906931883; Wed, 13 Apr 2022 20:28:51 -0700 (PDT)
Received: from smtpclient.apple (c-67-169-103-239.hsd1.ca.comcast.net. [67.169.103.239]) by smtp.gmail.com with ESMTPSA id m17-20020a17090a859100b001bc20ddcc67sm4223039pjn.34.2022.04.13.20.28.50 (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Wed, 13 Apr 2022 20:28:51 -0700 (PDT)
Sender: Tony Li <tony1athome@gmail.com>
From: Tony Li <tony.li@tony.li>
Message-Id: <EA787FB0-1969-41B1-BE37-6956470E18F1@tony.li>
Content-Type: multipart/alternative; boundary="Apple-Mail=_99BCCAAE-885A-4F28-9137-DB1AEAE1A306"
Mime-Version: 1.0 (Mac OS X Mail 15.0 \(3693.60.0.1.1\))
Date: Wed, 13 Apr 2022 20:28:50 -0700
In-Reply-To: <PH0PR13MB4795DCEA91F17EF5672767629AEC9@PH0PR13MB4795.namprd13.prod.outlook.com>
Cc: John E Drake <jdrake@juniper.net>, Kireeti Kompella <kireeti.kompella@gmail.com>, "mpls@ietf.org" <mpls@ietf.org>
To: Haoyu Song <haoyu.song@futurewei.com>
References: <BY3PR13MB4787BFC0BE610AEDEC0925919AED9@BY3PR13MB4787.namprd13.prod.outlook.com> <60FA12CB-9955-4A19-97AC-917FD9AC1D64@gmail.com> <BY3PR13MB47874837B2BC69992E5103839AEC9@BY3PR13MB4787.namprd13.prod.outlook.com> <BY3PR05MB8081FA851029B917B5626EBDC7EC9@BY3PR05MB8081.namprd05.prod.outlook.com> <BY3PR13MB47878C6DDF33073B5B783B119AEC9@BY3PR13MB4787.namprd13.prod.outlook.com> <BY3PR05MB80818C8F19F77E449EE9E357C7EC9@BY3PR05MB8081.namprd05.prod.outlook.com> <PH0PR13MB4795DCEA91F17EF5672767629AEC9@PH0PR13MB4795.namprd13.prod.outlook.com>
X-Mailer: Apple Mail (2.3693.60.0.1.1)
Archived-At: <https://mailarchive.ietf.org/arch/msg/mpls/RArM117b7dbvqZRmFPATdSe8cjU>
Subject: Re: [mpls] Concerns about ISD
X-BeenThere: mpls@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: Multi-Protocol Label Switching WG <mpls.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mpls>, <mailto:mpls-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/mpls/>
List-Post: <mailto:mpls@ietf.org>
List-Help: <mailto:mpls-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mpls>, <mailto:mpls-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 14 Apr 2022 03:28:58 -0000

Hi Haoyu,

> On Apr 13, 2022, at 4:37 PM, Haoyu Song <haoyu.song@futurewei.com> wrote:
> 
> I’m a hardware guy and I’d like to discuss with your hardware experts on the design tradeoffs. You keep saying that you don’t believe my analysis so I really want to learn what’s wrong with it or which part you don’t agree with.
> You can show them my slides to begin with. Thanks!


Well, I’m no hardware guy, but here’s the issues that I have with your presentation:

- You start off the an example of the Trident architecture. All well and good, but that’s one and only one data point. It’s always dangerous to extrapolate a solution from a single data point.  

- Other architectures simply provide the start of the packet to the network processor and map in some number of bytes. If you’re very lucky, some are mapped into registers already for you. The rest you have to issue reads for.  In the cases that I’ve dealt with, the cost of the computation is trivial, the time is spent doing reads. The cost of parsing is irrelevant: you just need to minimize the reads. You actually agree with this point on slide 12. ISD allows us to minimize the number of reads.

- You mention header chains. AFAIK, no one is proposing header chains.

- Similarly, you object to TLVs. No one is proposing TLVs. On this, we agree, this would be a poor design choice.

- Then you make this point:

	• If no PSD, and ISD is at ToS, and no need to scan other labels and payload, 
	  then there’s some gain on parsing speed, but negligible compared to the 
	  cost for actual header/data processing cost

I disagree with this completely.  If there is no PSD and only ISD, then we will only need to read to the end of ISD.  If the data were in PSD instead of being in ISD, then not only do we have to deal with the MNA label, but we have to parse down to the bottom of the label stack.  That’s a linear search through the stack, reading  every label.  This is where PSD is very, very, very painful.  BTW, this is true if ISD isn’t at the TOS either.  Any place other than the bottom of stack and ISD is better than PSD.

Regards,
Tony