<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: mlerner</title><link>https://news.ycombinator.com/user?id=mlerner</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 06 Apr 2026 00:47:18 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=mlerner" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[Graham: Synchronizing Clocks by Leveraging Local Clock Properties (2022) [pdf]]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.usenix.org/system/files/nsdi22-paper-najafi_1.pdf">https://www.usenix.org/system/files/nsdi22-paper-najafi_1.pdf</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=44860832">https://news.ycombinator.com/item?id=44860832</a></p>
<p>Points: 59</p>
<p># Comments: 13</p>
]]></description><pubDate>Mon, 11 Aug 2025 04:54:20 +0000</pubDate><link>https://www.usenix.org/system/files/nsdi22-paper-najafi_1.pdf</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=44860832</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44860832</guid></item><item><title><![CDATA[Autothrottle: Resource Management for SLO-Targeted Microservices]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.usenix.org/conference/nsdi24/presentation/wang-zibo">https://www.usenix.org/conference/nsdi24/presentation/wang-zibo</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=42912911">https://news.ycombinator.com/item?id=42912911</a></p>
<p>Points: 22</p>
<p># Comments: 1</p>
]]></description><pubDate>Sun, 02 Feb 2025 23:17:38 +0000</pubDate><link>https://www.usenix.org/conference/nsdi24/presentation/wang-zibo</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=42912911</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42912911</guid></item><item><title><![CDATA[Resiliency at Scale: Managing Google's TPUv4 Machine Learning Supercomputer]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.micahlerner.com/2025/01/03/resiliency-at-scale-managing-googles-tpuv4-machine-learning-supercomputer.html">https://www.micahlerner.com/2025/01/03/resiliency-at-scale-managing-googles-tpuv4-machine-learning-supercomputer.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=42835880">https://news.ycombinator.com/item?id=42835880</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 27 Jan 2025 00:26:40 +0000</pubDate><link>https://www.micahlerner.com/2025/01/03/resiliency-at-scale-managing-googles-tpuv4-machine-learning-supercomputer.html</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=42835880</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42835880</guid></item><item><title><![CDATA[How the anti-cheat is anti-cheating so far]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.leagueoflegends.com/en-us/news/dev/dev-vanguard-x-lol-retrospective/">https://www.leagueoflegends.com/en-us/news/dev/dev-vanguard-x-lol-retrospective/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=41364059">https://news.ycombinator.com/item?id=41364059</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 27 Aug 2024 02:17:21 +0000</pubDate><link>https://www.leagueoflegends.com/en-us/news/dev/dev-vanguard-x-lol-retrospective/</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=41364059</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41364059</guid></item><item><title><![CDATA[ServiceRouter: Hyperscale and Minimal Cost Service Mesh at Meta]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.micahlerner.com/2024/03/28/servicerouter-hyperscale-and-minimal-cost-service-mesh-at-meta.html">https://www.micahlerner.com/2024/03/28/servicerouter-hyperscale-and-minimal-cost-service-mesh-at-meta.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39886114">https://news.ycombinator.com/item?id=39886114</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Sun, 31 Mar 2024 17:24:07 +0000</pubDate><link>https://www.micahlerner.com/2024/03/28/servicerouter-hyperscale-and-minimal-cost-service-mesh-at-meta.html</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39886114</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39886114</guid></item><item><title><![CDATA[A Cloud-Scale Characterization of Remote Procedure Calls]]></title><description><![CDATA[
<p>Article URL: <a href="https://newsletter.micahlerner.com/p/paper-review-a-cloud-scale-characterization">https://newsletter.micahlerner.com/p/paper-review-a-cloud-scale-characterization</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39598390">https://news.ycombinator.com/item?id=39598390</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 05 Mar 2024 01:34:44 +0000</pubDate><link>https://newsletter.micahlerner.com/p/paper-review-a-cloud-scale-characterization</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39598390</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39598390</guid></item><item><title><![CDATA[A Cloud-Scale Characterization of Google's Remote Procedure Calls]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.micahlerner.com/2024/03/03/a-cloud-scale-characterization-of-remote-procedure-calls.html">https://www.micahlerner.com/2024/03/03/a-cloud-scale-characterization-of-remote-procedure-calls.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39586099">https://news.ycombinator.com/item?id=39586099</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 04 Mar 2024 01:15:25 +0000</pubDate><link>https://www.micahlerner.com/2024/03/03/a-cloud-scale-characterization-of-remote-procedure-calls.html</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39586099</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39586099</guid></item><item><title><![CDATA[Gemini, Amazon's system for fast failure recovery in distributed model training]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.micahlerner.com/2024/01/30/gemini-fast-failure-recovery-in-distributed-training-with-in-memory-checkpoints.html">https://www.micahlerner.com/2024/01/30/gemini-fast-failure-recovery-in-distributed-training-with-in-memory-checkpoints.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39573995">https://news.ycombinator.com/item?id=39573995</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Sat, 02 Mar 2024 17:19:39 +0000</pubDate><link>https://www.micahlerner.com/2024/01/30/gemini-fast-failure-recovery-in-distributed-training-with-in-memory-checkpoints.html</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39573995</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39573995</guid></item><item><title><![CDATA[Defcon: Preventing overload with graceful feature degradation (2023)]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.micahlerner.com/2023/07/23/defcon-preventing-overload-with-graceful-feature-degradation.html">https://www.micahlerner.com/2023/07/23/defcon-preventing-overload-with-graceful-feature-degradation.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39554874">https://news.ycombinator.com/item?id=39554874</a></p>
<p>Points: 237</p>
<p># Comments: 95</p>
]]></description><pubDate>Thu, 29 Feb 2024 20:50:50 +0000</pubDate><link>https://www.micahlerner.com/2023/07/23/defcon-preventing-overload-with-graceful-feature-degradation.html</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39554874</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39554874</guid></item><item><title><![CDATA[MotherDuck: DuckDB in the Cloud and in the Client [pdf]]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.cidrdb.org/cidr2024/papers/p46-atwal.pdf">https://www.cidrdb.org/cidr2024/papers/p46-atwal.pdf</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39254198">https://news.ycombinator.com/item?id=39254198</a></p>
<p>Points: 5</p>
<p># Comments: 0</p>
]]></description><pubDate>Sun, 04 Feb 2024 20:39:32 +0000</pubDate><link>https://www.cidrdb.org/cidr2024/papers/p46-atwal.pdf</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39254198</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39254198</guid></item><item><title><![CDATA[Gemini: Fast Failure Recovery in Distributed Training with In-Memory Checkpoints]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.micahlerner.com/2024/01/30/gemini-fast-failure-recovery-in-distributed-training-with-in-memory-checkpoints.html">https://www.micahlerner.com/2024/01/30/gemini-fast-failure-recovery-in-distributed-training-with-in-memory-checkpoints.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39253777">https://news.ycombinator.com/item?id=39253777</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Sun, 04 Feb 2024 19:59:17 +0000</pubDate><link>https://www.micahlerner.com/2024/01/30/gemini-fast-failure-recovery-in-distributed-training-with-in-memory-checkpoints.html</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39253777</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39253777</guid></item><item><title><![CDATA[New comment by mlerner in "XFaaS: Hyperscale and Low Cost Serverless Functions at Meta"]]></title><description><![CDATA[
<p>Author of the paper summary here - my understanding is that XFaaS doesn't run functions that are run in response to user input (e.g. XFaaS does not execute code that fetches and returns data because a user clicked on a button).</p>
]]></description><pubDate>Wed, 31 Jan 2024 14:25:18 +0000</pubDate><link>https://news.ycombinator.com/item?id=39204143</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39204143</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39204143</guid></item><item><title><![CDATA[Gemini: Fast Failure Recovery in Distributed Training with In-Memory Checkpoints]]></title><description><![CDATA[
<p>Article URL: <a href="https://newsletter.micahlerner.com/p/gemini-fast-failure-recovery-in-distributed">https://newsletter.micahlerner.com/p/gemini-fast-failure-recovery-in-distributed</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39198732">https://news.ycombinator.com/item?id=39198732</a></p>
<p>Points: 4</p>
<p># Comments: 0</p>
]]></description><pubDate>Wed, 31 Jan 2024 01:52:28 +0000</pubDate><link>https://newsletter.micahlerner.com/p/gemini-fast-failure-recovery-in-distributed</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39198732</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39198732</guid></item><item><title><![CDATA[XFaaS: Hyperscale and Low Cost Serverless Functions at Meta]]></title><description><![CDATA[
<p>Article URL: <a href="https://newsletter.micahlerner.com/p/paper-review-xfaas-hyperscale-and">https://newsletter.micahlerner.com/p/paper-review-xfaas-hyperscale-and</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39158322">https://news.ycombinator.com/item?id=39158322</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Sat, 27 Jan 2024 18:31:12 +0000</pubDate><link>https://newsletter.micahlerner.com/p/paper-review-xfaas-hyperscale-and</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39158322</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39158322</guid></item><item><title><![CDATA[XFaaS: Hyperscale and Low Cost Serverless Functions at Meta]]></title><description><![CDATA[
<p>Article URL: <a href="https://newsletter.micahlerner.com/p/paper-review-xfaas-hyperscale-and">https://newsletter.micahlerner.com/p/paper-review-xfaas-hyperscale-and</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39112226">https://news.ycombinator.com/item?id=39112226</a></p>
<p>Points: 3</p>
<p># Comments: 1</p>
]]></description><pubDate>Wed, 24 Jan 2024 00:55:09 +0000</pubDate><link>https://newsletter.micahlerner.com/p/paper-review-xfaas-hyperscale-and</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39112226</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39112226</guid></item><item><title><![CDATA[Blueprint: A Toolchain for Highly-Reconfigurable Microservice Applications]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.micahlerner.com/2024/01/02/blueprint-a-toolchain-for-highly-reconfigurable-microservice-applications.html">https://www.micahlerner.com/2024/01/02/blueprint-a-toolchain-for-highly-reconfigurable-microservice-applications.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39080929">https://news.ycombinator.com/item?id=39080929</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Sun, 21 Jan 2024 18:09:08 +0000</pubDate><link>https://www.micahlerner.com/2024/01/02/blueprint-a-toolchain-for-highly-reconfigurable-microservice-applications.html</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39080929</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39080929</guid></item><item><title><![CDATA[Gemini: Fast Failure Recovery in Distributed Training with In-Memory Checkpoints [pdf]]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.cs.rice.edu/~eugeneng/papers/SOSP23.pdf">https://www.cs.rice.edu/~eugeneng/papers/SOSP23.pdf</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39074646">https://news.ycombinator.com/item?id=39074646</a></p>
<p>Points: 50</p>
<p># Comments: 13</p>
]]></description><pubDate>Sun, 21 Jan 2024 01:48:19 +0000</pubDate><link>https://www.cs.rice.edu/~eugeneng/papers/SOSP23.pdf</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39074646</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39074646</guid></item><item><title><![CDATA[XFaaS: Hyperscale and Low Cost Serverless Functions at Meta [pdf]]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.cis.upenn.edu/~haorz/papers/xfaas-sosp23.pdf">https://www.cis.upenn.edu/~haorz/papers/xfaas-sosp23.pdf</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39073736">https://news.ycombinator.com/item?id=39073736</a></p>
<p>Points: 4</p>
<p># Comments: 0</p>
]]></description><pubDate>Sat, 20 Jan 2024 23:41:23 +0000</pubDate><link>https://www.cis.upenn.edu/~haorz/papers/xfaas-sosp23.pdf</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39073736</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39073736</guid></item><item><title><![CDATA[Efficient Memory Management for Large Language Model Serving with PagedAttention]]></title><description><![CDATA[
<p>Article URL: <a href="https://newsletter.micahlerner.com/p/paper-review-efficient-memory-management">https://newsletter.micahlerner.com/p/paper-review-efficient-memory-management</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39072944">https://news.ycombinator.com/item?id=39072944</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Sat, 20 Jan 2024 22:09:43 +0000</pubDate><link>https://newsletter.micahlerner.com/p/paper-review-efficient-memory-management</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=39072944</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39072944</guid></item><item><title><![CDATA[Efficient Memory Management for Large Language Model Serving with PagedAttention]]></title><description><![CDATA[
<p>Article URL: <a href="https://newsletter.micahlerner.com/p/paper-review-efficient-memory-management">https://newsletter.micahlerner.com/p/paper-review-efficient-memory-management</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=38953971">https://news.ycombinator.com/item?id=38953971</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 11 Jan 2024 15:58:26 +0000</pubDate><link>https://newsletter.micahlerner.com/p/paper-review-efficient-memory-management</link><dc:creator>mlerner</dc:creator><comments>https://news.ycombinator.com/item?id=38953971</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=38953971</guid></item></channel></rss>