[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4990838b-660d-46a2-b21c-67adcba61ff9@lucifer.local>
Date: Fri, 27 Jun 2025 16:30:12 +0100
From: Lorenzo Stoakes <lorenzo.stoakes@...cle.com>
To: siddhartha@...ip.in
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org, mgorman@...e.de,
Vlastimil Babka <vbabka@...e.cz>
Subject: Re: [PATCH] mm: limit THP alignment – performance gain observed in AI inference workloads
+cc Vlata
On Fri, Jun 27, 2025 at 04:09:16PM +0530, siddhartha@...ip.in wrote:
> Hi all,
>
> I wanted to share validation data from a Hugging Face-based AI inferencing
> workload,
> which was significantly impacted by the THP alignment logic introduced in
> commit efa7df3e3bb5.
>
> Using transformer models with dynamic input lengths on Intel Xeon (Cooper
> Lake),
> we observed up to a 3200% throughput improvement after applying the patch
> from Oct 2024:
>
> mm: limit THP alignment of anonymous mappings to PMD-aligned sizes
All congratulations are owed to Vlastimil Babka for doing this, cc'd :)
I gather he enjoys novelty beer mugs as tokens of thanks ;)
>
> Metrics:
> - Model: BERT-base
> - Inference engine: Transformers + ONNX Runtime
> - Kernel: 6.6 vs patched 6.6.8
> - Batch size: 8-32, input length: 64-512 tokens
> - Metric: inference throughput (samples/sec)
>
> Thanks for the fix -- this change had real impact on a production-relevant
> workload.
>
> Best Regards,
> Siddhartha Sharma
> ISV @ Kenip
> Solution Link: https://www.intel.com/content/www/us/en/partner/showcase/offering/a5bHo00000045YUIAY/deadlock-clearance.html
>
Powered by blists - more mailing lists