[Cats] Question on token length in draft-zhang-cats-token-aware-ts-00
Yina Dai <daiyn5@mail2.sysu.edu.cn> Tue, 28 July 2026 11:11 UTC
Return-Path: <daiyn5@mail2.sysu.edu.cn>
X-Original-To: cats@mail2.ietf.org
Delivered-To: cats@mail2.ietf.org
Received: from localhost (localhost [127.0.0.1]) by mail2.ietf.org (Postfix) with ESMTP id 6314E11FC45CA for <cats@mail2.ietf.org>; Tue, 28 Jul 2026 04:11:28 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=ietf.org; s=ietf1; t=1785237088; bh=1r1JMgX658EBve3Wmmj69Kggd3erstKrW7tT+VP605Q=; h=To:Cc:Subject:From:Date; b=OJowql4bUN7jWrkTwYdRwRwo4BtdoY1SbzJwRYc52f1VmhdfPrCDUB2RJtWygVBo/ d9kKueI+w3/7okMafT19n9jRJRqtAfBWwcAboQs2Nzqk6PmMIWi0epFIleY8XQQvYI 9FTNOBvWjW9qOdCVZRnxKpveL72n/9jXuF73kfZw=
X-Virus-Scanned: amavisd-new at ietf.org
X-Spam-Flag: NO
X-Spam-Score: -1.996
X-Spam-Level:
X-Spam-Status: No, score=-1.996 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, MSGID_FROM_MTA_HEADER=0.001, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_VALIDITY_CERTIFIED_BLOCKED=0.001, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: mail2.ietf.org (amavisd-new); dkim=pass (1024-bit key) header.d=mail2.sysu.edu.cn
Received: from mail2.ietf.org ([166.84.6.31]) by localhost (mail2.ietf.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Ph1C5A1fi4QX for <cats@mail2.ietf.org>; Tue, 28 Jul 2026 04:11:18 -0700 (PDT)
Received: from mail-m10146.netease.com (mail-m10146.netease.com [154.81.10.146]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-256) server-digest SHA256) (No client certificate requested) by mail2.ietf.org (Postfix) with ESMTPS id C688911FC4431 for <cats@ietf.org>; Tue, 28 Jul 2026 04:10:58 -0700 (PDT)
Content-Type: multipart/alternative; BOUNDARY="=_Part_270818_1024108940.1785237048558"
Message-ID: <AOsA-gDlKpC98Pf8bqlYwKo0.1.1785237048558.Hmail.daiyn5@mail2.sysu.edu.cn>
To: zhangnh12 <zhangnh12@chinaunicom.cn>, hanmy12 <hanmy12@chinaunicom.cn>, yixx3 <yixx3@chinaunicom.cn>
X-Priority: 3
X-Mailer: HMail Webmail Server V2.0 Copyright (c) 2016-163.com Sirius_WEB_MAC_1.65.2
X-Originating-IP: 112.95.163.118
MIME-Version: 1.0
Received: from daiyn5@mail2.sysu.edu.cn( [112.95.163.118] ) by ajax-webmail ( [127.0.0.1] ) ; Tue, 28 Jul 2026 19:10:48 +0800 (GMT+08:00)
From: Yina Dai <daiyn5@mail2.sysu.edu.cn>
Date: Tue, 28 Jul 2026 19:10:48 +0800
X-HM-Tid: 0a9fa83314130302kunmaa68acd4120b4
X-HM-MType: 1
X-HM-NTES-SC: AL0_4z5B86Wr4Tz9jdMF+bhXMczeGSIZXwBQgqb8hRnZcQpDYHtzFgoOJ1LWcv dh5j9ol6nrRMeg7Qi1pNE1YXhTuN9h5NaODfWyq7t0SsREMh/aCNdmYu+cPrkBfFRVdl5EIFKPVE UlrC7OxNBb3owUgpHsKNBfLcorxCBbEA87U6I=
X-HM-Spam-Status: e1kfGhgUHx5ZQUpXWQgPGg8OCBgUHx5ZQUlOS1dZFg8aDwILHllBWSg2Ly tZV1koWUFITzdXWRgWCB1ZQUpXWS1ZQUlXWQ8JGhUIEh9ZQVkaHxpKVkIZGhkeHU9PHR0ZTFYVFA kWGhdVEwETFhoSFyQUDg9ZV1kYEgtZQVlKSklVQk5VSk1IVUpKQ1lXWRYaDxIVHRRZQVlPS0hVSk tISk5OSFVKS0tVSkJLS1kG
DKIM-Signature: a=rsa-sha256; b=MPBbFCApigYKqujhdioPQzmL0Mo0kL6IaocdhAZK31Xurfd7+j4Ts+wddmzDlo5RavKopgfeQbwmMB8ts6exyCWTaSx3xEw93bzu6XtQ0iltvqwN/kPv6joIht3myZUsI6tSGawdBwFY8TBz4Skjn1nPE4ZsNNQPV8gcxdUPXiA=; s=default; c=relaxed/relaxed; d=mail2.sysu.edu.cn; v=1; bh=1r1JMgX658EBve3Wmmj69Kggd3erstKrW7tT+VP605Q=; h=date:mime-version:subject:message-id:from;
Message-ID-Hash: LVEQEAQMN7E4LO2YXEUU6JULTK7X6OD5
X-Message-ID-Hash: LVEQEAQMN7E4LO2YXEUU6JULTK7X6OD5
X-MailFrom: daiyn5@mail2.sysu.edu.cn
X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; emergency; loop; banned-address; member-moderation; nonmember-moderation; administrivia; implicit-dest; max-recipients; max-size; news-moderation; no-subject; digests; suspicious-header
CC: cats <cats@ietf.org>
X-Mailman-Version: 3.3.9rc6
Precedence: list
Subject: [Cats] Question on token length in draft-zhang-cats-token-aware-ts-00
List-Id: "Computing-Aware Traffic Steering (CATS)" <cats.ietf.org>
Archived-At: <https://mailarchive.ietf.org/arch/msg/cats/JE9PRTHXZ7mLzLGO16z0PnqKVRM>
List-Archive: <https://mailarchive.ietf.org/arch/browse/cats>
List-Help: <mailto:cats-request@ietf.org?subject=help>
List-Owner: <mailto:cats-owner@ietf.org>
List-Post: <mailto:cats@ietf.org>
List-Subscribe: <mailto:cats-join@ietf.org>
List-Unsubscribe: <mailto:cats-leave@ietf.org>
Hi authors, Thank you for sharing draft-zhang-cats-token-aware-ts-00. I have one question regarding the use of estimated token length for traffic steering. The draft appears to use token length as an indicator of the computing demand of an AI request. However, a short input may still lead to long reasoning or output generation, while a long input may require only a relatively simple response. Could the authors clarify whether token length is intended to be a primary steering factor, or only one auxiliary input together with other request and service-instance information? It may also be useful to clarify whether the draft refers to input tokens, expected output tokens, or both. Best regards, Yina Dai