From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 6730A46534; Tue, 8 Apr 2025 15:27:36 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 3154A40EDC; Tue, 8 Apr 2025 15:27:36 +0200 (CEST) Received: from mail-pf1-f170.google.com (mail-pf1-f170.google.com [209.85.210.170]) by mails.dpdk.org (Postfix) with ESMTP id 7615E40A89 for ; Tue, 8 Apr 2025 15:27:34 +0200 (CEST) Received: by mail-pf1-f170.google.com with SMTP id d2e1a72fcca58-72d3b48d2ffso5868460b3a.2 for ; Tue, 08 Apr 2025 06:27:34 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=networkplumber-org.20230601.gappssmtp.com; s=20230601; t=1744118853; x=1744723653; darn=dpdk.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:subject:cc:to:from:date:from:to:cc:subject:date :message-id:reply-to; bh=8GWDzoQ+uu2c+J6CcYrrmimZkb4X/6lL6q+QyTW/SDY=; b=QbOc3/XAEOGFyltLjjMGWPAyZswVGn/u2HivWXWN2iRobjo7J/emGsDHS/RDl5TIB6 jkNo0U2rxpQrRGgkMhnMnTH/0A5HjiKjos6glewwdsaMxofAc4bvfzxc/cL6xF3WgRX9 5WtrzGS0rrYvuCsHlQEH/9jEByLTINUKfDAvevqLRd60qcv5TI7QgEq3G/+kkD0o2q4k 2nI8VTnjPLdb0yiXHL8eof/k9D/YZ56SnBArLy6O3CeFmdkRyb1XzEsQaVdJsfDzHAfe 6fzYhjh3gku8seo6AzMh/kGMBfA6VLNuul3PYH6+HKwFF0AsPu+A/XYD5crA1GazLwCY 2nyA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1744118853; x=1744723653; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:subject:cc:to:from:date:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=8GWDzoQ+uu2c+J6CcYrrmimZkb4X/6lL6q+QyTW/SDY=; b=gudYoFC+bSMeNrzp3RWH3BkkloTlY4t4WVS8AjkKaueD5kOOT5DlTozgQ1IwgKdcYr alV0nDIDrlSJ3qpp3D61jWKQa05UV29Mf8dmP8nodmGAv+p18o+2/igYvK6YJnJXNjdu G9C1GQui+iGtYeRSsW+RZqvaNOaEi8LBfjO75WJsp+3/S34TyCTklkVf0bGDjzC/PSIb VFq9uI/yjmJPaco9wnCKXNULGMXD76Kiz/HRUzcqqVDAv+eu2fyavxz445kdieQ3C28x eB3HvlLTBV5Ldz4y2ifmXfO6/4oPM81UctyPW55Aj/gaFPwSNc4O6yZSEBvRHFydtY3v N2+A== X-Gm-Message-State: AOJu0YyusWXz2ljOuHJ6R5891Zd53Wp0Jii7tLmQemmVzzgMi0viy9V6 RVaeIg+v5rNae5LpQQuw3cLB4wdzOPQaftHduWjOeAWXtjMc/bWgwOXIkBSrqDM= X-Gm-Gg: ASbGnctclyXEVWZ9yS+ccUOVybqJ5swLXOdS4jtdflGkBaYUOPkfFLbN4qwY2zwoTdc zaYUuG+sdl73zvywl6Rd2x4H1gn0ieL2wnUvsvARJScFDCCB0T9UYv1wmRZq5KQkIwKnawXAl2P jetD5pTqptg3X/pOTij1J5wfQARF8vwUdinSz/C6MH2QWNj+m8BxoGinWJGuWCs6tnCjA/AUEdd 1H2uXBqaStj0FSepM5DcV0Gju8eI81rm2AH41BLAMlnkQcmPrL1MXrsOOt01Ns5PjPzcyjSrKEP 1Na9pH6kZXw6qPO7C7Aa+SzbILTt5/PKZB6Iu5473enCa6HfjkdqnnNZ1LqJmNnM8DBCrXxhvBY 0+pomk6oiJ+SiePb3/ZEQ X-Google-Smtp-Source: AGHT+IE4t6Ht4QNA3gM9O4QLu6xYW3h1n2sljVNvaNs5d1Hy3qM4PAyVpyffPvxYS9yHqaF0RJwyTw== X-Received: by 2002:a05:6a21:9982:b0:1f5:8d3b:e272 with SMTP id adf61e73a8af0-20104590ef9mr26057822637.1.1744118853579; Tue, 08 Apr 2025 06:27:33 -0700 (PDT) Received: from hermes.local (204-195-96-226.wavecable.com. [204.195.96.226]) by smtp.gmail.com with ESMTPSA id d2e1a72fcca58-739da0b3feesm10779109b3a.140.2025.04.08.06.27.33 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 08 Apr 2025 06:27:33 -0700 (PDT) Date: Tue, 8 Apr 2025 06:27:28 -0700 From: Stephen Hemminger To: =?UTF-8?B?6YOt5pil6Zye?= Cc: "dev@dpdk.org" , =?UTF-8?B?5byg5aOr5Lqa?= , =?UTF-8?B?6YeR5piO?= , SRC-B Security Subject: Re: Inquiry About DPDK's Roadmap for Dynamic CPU Scaling Support Message-ID: <20250408062728.013a5b61@hermes.local> In-Reply-To: <20250328033731epcms5p1393d177ee4de90debad11e9578e758f9@epcms5p1> References: <20250328033731epcms5p1393d177ee4de90debad11e9578e758f9@epcms5p1> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org On Fri, 28 Mar 2025 12:37:31 +0900 =E9=83=AD=E6=98=A5=E9=9C=9E wrote: > Dear=C2=A0DPDK Team, >=20 > I hope this email finds you well.=C2=A0 >=20 > I'm writing to inquire about DPDK's future plans regarding dynamic CPU in= itialization support, particularly in Kubernetes environments where exclusi= ve CPU Pod scaling is being enhanced. >=20 > Backgroud: > With Kubernetes upcoming improvements to InPlacePodVerticalScaling for ex= clusive CPU Pods (Guaranteed QoS Pods), we have encountered a limitation wh= en using DPDK: > =E2=80=A2 A Pod initialized with=C2=A05 exclusive CPUs=C2=A0can have DPDK= =E2=80=99s EAL successfully bind and utilize these cores. > =E2=80=A2 However, when the Pod scales up (e.g., adding=C2=A02 new CPUs),= DPDK currently lacks native mechanisms to detect and initialize the newly = allocated cores. This forces us to either restart the Pod (causing service = disruption) or leave additional CPUs underutilized. >=20 > Key Questions: > =E2=80=A2 Is there an active DPDK roadmap item to support binded and=C2= =A0initialized=C2=A0CPUs scaling up and down? > =E2=80=A2 If not, would the DPDK community consider contributions in this= area, especially given Kubernetes=E2=80=99 evolving CPU management capabil= ities? >=20 > We believe such functionality would significantly benefit cloud-native NF= V workloads. Any guidance or references to ongoing work would be greatly ap= preciated. >=20 > Reference Link: >=20 > =C2=A0 =C2=A0 =C2=A0 =C2=A0 InPlacePodVerticalScaling=C2=A0Enhancements f= or=C2=A0guaranteed pods:=C2=A0https://github.com/kubernetes/enhancements/bl= ob/master/keps/sig-node/1287-in-place-update-pod-resources/README.md#future= -enhancements >=20 > =C2=A0 =C2=A0 =C2=A0 =C2=A0 A PR to implement=C2=A0guaranteed pods resize= : https://github.com/kubernetes/kubernetes/pull/129719 >=20 > Thank you for your time and expertise. Looking forward to your insights. I have seen no plans for doing this. It would mostly impact the application. One way of doing it now would be launch DPDK application with max possible = CPU's and instead of using remote_launch across all workers, be more selective. It doesn't make sense for DPDK to grow direct access to Kubernetes API's. But maybe the kernel API for hotplug has something.