From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id E1F5EC4167D for ; Thu, 2 Nov 2023 06:43:16 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233245AbjKBGnR (ORCPT ); Thu, 2 Nov 2023 02:43:17 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:52532 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229445AbjKBGnP (ORCPT ); Thu, 2 Nov 2023 02:43:15 -0400 Received: from mgamail.intel.com (mgamail.intel.com [192.55.52.43]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 165DF125; Wed, 1 Nov 2023 23:43:10 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1698907390; x=1730443390; h=from:to:cc:subject:in-reply-to:references:date: message-id:mime-version; bh=eFzutUwI91AlprFudf7uXPVWde/i5DfJ1a5Vzf3IEHM=; b=DBBfIHJrEAcsME4CLAzNvgjMAbnWGkNG3lVW6oVO9u52dEwoXYo+OM7E bqEl1Cn57FmuQaUUKERqXpL9orCgCNbnzdFNU8gmn6f8lGoq0nyXQh1YC b8EpHspZqQW4MdNvHOSXebAJed+O5DZYPQlKll/R/lG/DU7FyK9MxRB0W oJN4s6nY0SJtQyVoEBOYCD/3dBPx6S/QeCkXMkXTeboaIa2WodnSjyeNV ed2TOsY3SaJKDfk0x0/SZs0MyPd8oRElGtsuGOKjn8m65mhEoSyY2+8vr w24B36Vl1mQ/LYLyxCOqmmYfX58Z3okAlBRfN1DuzdGPoUKj7ots07wek g==; X-IronPort-AV: E=McAfee;i="6600,9927,10881"; a="474887758" X-IronPort-AV: E=Sophos;i="6.03,270,1694761200"; d="scan'208";a="474887758" Received: from orsmga003.jf.intel.com ([10.7.209.27]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 01 Nov 2023 23:43:09 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10881"; a="711040332" X-IronPort-AV: E=Sophos;i="6.03,270,1694761200"; d="scan'208";a="711040332" Received: from yhuang6-desk2.sh.intel.com (HELO yhuang6-desk2.ccr.corp.intel.com) ([10.238.208.55]) by orsmga003-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 01 Nov 2023 23:43:05 -0700 From: "Huang, Ying" To: Ravi Jonnalagadda Cc: , , , , , , , , , , , , , , , Subject: Re: [RFC PATCH v3 0/4] Node Weights and Weighted Interleave In-Reply-To: <20231101092923.283-1-ravis.opensrc@micron.com> (Ravi Jonnalagadda's message of "Wed, 1 Nov 2023 14:59:23 +0530") References: <87il6m6w2j.fsf@yhuang6-desk2.ccr.corp.intel.com> <20231101092923.283-1-ravis.opensrc@micron.com> Date: Thu, 02 Nov 2023 14:41:03 +0800 Message-ID: <87a5rw1wu8.fsf@yhuang6-desk2.ccr.corp.intel.com> User-Agent: Gnus/5.13 (Gnus v5.13) MIME-Version: 1.0 Content-Type: text/plain; charset=ascii Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Ravi Jonnalagadda writes: >>> On Tue, Oct 31, 2023 at 04:56:27PM +0100, Michal Hocko wrote: >>>> On Tue 31-10-23 11:21:42, Johannes Weiner wrote: >>>> > On Tue, Oct 31, 2023 at 10:53:41AM +0100, Michal Hocko wrote: >>>> > > On Mon 30-10-23 20:38:06, Gregory Price wrote: >> >>[snip] >> >>>> >>>> > This hopefully also explains why it's a global setting. The usecase is >>>> > different from conventional NUMA interleaving, which is used as a >>>> > locality measure: spread shared data evenly between compute >>>> > nodes. This one isn't about locality - the CXL tier doesn't have local >>>> > compute. Instead, the optimal spread is based on hardware parameters, >>>> > which is a global property rather than a per-workload one. >>>> >>>> Well, I am not convinced about that TBH. Sure it is probably a good fit >>>> for this specific CXL usecase but it just doesn't fit into many others I >>>> can think of - e.g. proportional use of those tiers based on the >>>> workload - you get what you pay for. >>>> >>>> Is there any specific reason for not having a new interleave interface >>>> which defines weights for the nodemask? Is this because the policy >>>> itself is very dynamic or is this more driven by simplicity of use? >>> >>> A downside of *requiring* weights to be paired with the mempolicy is >>> that it's then the application that would have to figure out the >>> weights dynamically, instead of having a static host configuration. A >>> policy of "I want to be spread for optimal bus bandwidth" translates >>> between different hardware configurations, but optimal weights will >>> vary depending on the type of machine a job runs on. >>> >>> That doesn't mean there couldn't be usecases for having weights as >>> policy as well in other scenarios, like you allude to above. It's just >>> so far such usecases haven't really materialized or spelled out >>> concretely. Maybe we just want both - a global default, and the >>> ability to override it locally. >> >>I think that this is a good idea. The system-wise configuration with >>reasonable default makes applications life much easier. If more control >>is needed, some kind of workload specific configuration can be added. > > Glad that we are in agreement here. For bandwidth expansion use cases > that this interleave patchset is trying to cater to, most applications > would have to follow the "reasanable defaults" for weights. > The necessity for applications to choose different weights while > interleaving would probably be to do capacity expansion which the > default memory tiering implementation would anyway support and provide > better latency. > >>And, instead of adding another memory policy, a cgroup-wise >>configuration may be easier to be used. The per-workload weight may >>need to be adjusted when we deploying different combination of workloads >>in the system. >> >>Another question is that should the weight be per-memory-tier or >>per-node? In this patchset, the weight is per-source-target-node >>combination. That is, the weight becomes a matrix instead of a vector. >>IIUC, this is used to control cross-socket memory access in addition to >>per-memory-type memory access. Do you think the added complexity is >>necessary? > > Pros and Cons of Node based interleave: > Pros: > 1. Weights can be defined for devices with different bandwidth and latency > characteristics individually irrespective of which tier they fall into. > 2. Defining the weight per-source-target-node would be necessary for multi > socket systems where few devices may be closer to one socket rather than other. > Cons: > 1. Weights need to be programmed for all the nodes which can be tedious for > systems with lot of NUMA nodes. 2. More complex, so need justification, for example, practical use case. > Pros and Cons of Memory Tier based interleave: > Pros: > 1. Programming weight per initiator would apply for all the nodes in the tier. > 2. Weights can be calculated considering the cumulative bandwidth of all > the nodes in the tier and need to be programmed once for all the nodes in a > given tier. > 3. It may be useful in cases where numa nodes with similar latency and bandwidth > characteristics increase, possibly with pooling use cases. 4. simpler. > Cons: > 1. If nodes with different bandwidth and latency characteristics are placed > in same tier as seen in the current mainline kernel, it will be difficult to > apply a correct interleave weight policy. > 2. There will be a need for functionality to move nodes between different tiers > or create new tiers to place such nodes for programming correct interleave weights. > We are working on a patch to support it currently. Thanks! If we have such system, we will need this. > 3. For systems where each numa node is having different characteristics, > a single node might end up existing in different memory tier, which would be > equivalent to node based interleaving. No. A node can only exist in one memory tier. > On newer systems where all CXL memory from different devices under a > port are combined to form single numa node, this scenario might be > applicable. You mean the different memory ranges of a NUMA node may have different performance? I don't think that we can deal with this. > 4. Users may need to keep track of different memory tiers and what nodes are present > in each tier for invoking interleave policy. I don't think this is a con. With node based solution, you need to know your system too. >> >>> Could you elaborate on the 'get what you pay for' usecase you >>> mentioned? >> -- Best Regards, Huang, Ying