Files
publish-assistant/cloud-edge/digests/atc-2025/index.html
2026-08-18 13:39:21 +00:00

24 lines
33 KiB
HTML
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
<!doctype html><html lang=en dir=auto data-theme=auto><head><meta charset=utf-8><meta http-equiv=X-UA-Compatible content="IE=edge"><meta name=viewport content="width=device-width,initial-scale=1,shrink-to-fit=no"><meta name=robots content="index, follow"><title>ATC 2025 Digest | Publish Assistant</title><meta name=keywords content="operating-systems,cloud,storage,networking"><meta name=description content="13 papers selected.
ASTERINAS: A Linux ABI-Compatible, Rust-Based Framekernel OS with a Small and Sound TCB
Yuke Peng, Hongliang Tian, Junyang Zhang, Ruihan Li et al.
TL;DR — A production-grade OS kernel written in Rust that exposes a full Linux ABI while confining unsafe code to a small, formally-audited framekernel core.
Why notable — ASTERINAS demonstrates that Linux compatibility and memory-safety guarantees are not mutually exclusive — unsafe Rust is isolated to under 5 kloc of framework code, giving systems operators a credible path toward a safer Linux-compatible kernel without sacrificing application portability."><meta name=author content="Publish Assistant"><link rel=canonical href=https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/atc-2025/><link crossorigin=anonymous href=/vincent/publish-assistant/assets/css/stylesheet.d72f07832e13c592b3edba91680bfe70f01daac396179bcace0ac36e8e0494c6.css integrity="sha256-1y8Hgy4TxZKz7bqRaAv+cPAdqsOWF5vKzgrDbo4ElMY=" rel="preload stylesheet" as=style><link rel=icon href=https://pub.sqrt.fr/vincent/publish-assistant/favicon.ico><link rel=icon type=image/png sizes=16x16 href=https://pub.sqrt.fr/vincent/publish-assistant/favicon-16x16.png><link rel=icon type=image/png sizes=32x32 href=https://pub.sqrt.fr/vincent/publish-assistant/favicon-32x32.png><link rel=apple-touch-icon href=https://pub.sqrt.fr/vincent/publish-assistant/apple-touch-icon.png><link rel=mask-icon href=https://pub.sqrt.fr/vincent/publish-assistant/safari-pinned-tab.svg><meta name=theme-color content="#2e2e33"><meta name=msapplication-TileColor content="#2e2e33"><link rel=alternate hreflang=en href=https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/atc-2025/><noscript><style>#theme-toggle,.top-link{display:none}</style><style>@media(prefers-color-scheme:dark){:root{--theme:rgb(29, 30, 32);--entry:rgb(46, 46, 51);--primary:rgb(218, 218, 219);--secondary:rgb(155, 156, 157);--tertiary:rgb(65, 66, 68);--content:rgb(196, 196, 197);--code-block-bg:rgb(46, 46, 51);--code-bg:rgb(55, 56, 62);--border:rgb(51, 51, 51);color-scheme:dark}.list{background:var(--theme)}.toc{background:var(--entry)}}</style></noscript><script>localStorage.getItem("pref-theme")==="dark"?document.querySelector("html").dataset.theme="dark":localStorage.getItem("pref-theme")==="light"?document.querySelector("html").dataset.theme="light":window.matchMedia("(prefers-color-scheme: dark)").matches?document.querySelector("html").dataset.theme="dark":document.querySelector("html").dataset.theme="light"</script><meta property="og:url" content="https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/atc-2025/"><meta property="og:site_name" content="Publish Assistant"><meta property="og:title" content="ATC 2025 Digest"><meta property="og:description" content="13 papers selected.
ASTERINAS: A Linux ABI-Compatible, Rust-Based Framekernel OS with a Small and Sound TCB Yuke Peng, Hongliang Tian, Junyang Zhang, Ruihan Li et al.
TL;DR — A production-grade OS kernel written in Rust that exposes a full Linux ABI while confining unsafe code to a small, formally-audited framekernel core.
Why notable — ASTERINAS demonstrates that Linux compatibility and memory-safety guarantees are not mutually exclusive — unsafe Rust is isolated to under 5 kloc of framework code, giving systems operators a credible path toward a safer Linux-compatible kernel without sacrificing application portability."><meta property="og:locale" content="en_us"><meta property="og:type" content="article"><meta property="article:section" content="cloud-edge"><meta property="article:published_time" content="2025-07-09T00:00:00+00:00"><meta property="article:modified_time" content="2025-07-09T00:00:00+00:00"><meta property="article:tag" content="Operating-Systems"><meta property="article:tag" content="Cloud"><meta property="article:tag" content="Storage"><meta property="article:tag" content="Networking"><meta name=twitter:card content="summary"><meta name=twitter:title content="ATC 2025 Digest"><meta name=twitter:description content="13 papers selected.
ASTERINAS: A Linux ABI-Compatible, Rust-Based Framekernel OS with a Small and Sound TCB Yuke Peng, Hongliang Tian, Junyang Zhang, Ruihan Li et al.
TL;DR — A production-grade OS kernel written in Rust that exposes a full Linux ABI while confining unsafe code to a small, formally-audited framekernel core.
Why notable — ASTERINAS demonstrates that Linux compatibility and memory-safety guarantees are not mutually exclusive — unsafe Rust is isolated to under 5 kloc of framework code, giving systems operators a credible path toward a safer Linux-compatible kernel without sacrificing application portability."><script type=application/ld+json>{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Edge and Cloud Systems","item":"https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/"},{"@type":"ListItem","position":2,"name":"Digests","item":"https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/"},{"@type":"ListItem","position":3,"name":"ATC 2025 Digest","item":"https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/atc-2025/"}]}</script><script type=application/ld+json>{"@context":"https://schema.org","@type":"BlogPosting","headline":"ATC 2025 Digest","name":"ATC 2025 Digest","description":"13 papers selected.\nASTERINAS: A Linux ABI-Compatible, Rust-Based Framekernel OS with a Small and Sound TCB Yuke Peng, Hongliang Tian, Junyang Zhang, Ruihan Li et al.\nTL;DR — A production-grade OS kernel written in Rust that exposes a full Linux ABI while confining unsafe code to a small, formally-audited framekernel core.\nWhy notable — ASTERINAS demonstrates that Linux compatibility and memory-safety guarantees are not mutually exclusive — unsafe Rust is isolated to under 5 kloc of framework code, giving systems operators a credible path toward a safer Linux-compatible kernel without sacrificing application portability.\n","keywords":["operating-systems","cloud","storage","networking"],"articleBody":"13 papers selected.\nASTERINAS: A Linux ABI-Compatible, Rust-Based Framekernel OS with a Small and Sound TCB Yuke Peng, Hongliang Tian, Junyang Zhang, Ruihan Li et al.\nTL;DR — A production-grade OS kernel written in Rust that exposes a full Linux ABI while confining unsafe code to a small, formally-audited framekernel core.\nWhy notable — ASTERINAS demonstrates that Linux compatibility and memory-safety guarantees are not mutually exclusive — unsafe Rust is isolated to under 5 kloc of framework code, giving systems operators a credible path toward a safer Linux-compatible kernel without sacrificing application portability.\n→ Read paper Rex: Closing the language-verifier gap with safe and usable kernel extensions Jinghao Jia, Ruowen Qin, Milo Craun, Egor Lukiyanov et al.\nTL;DR — Rex introduces a new kernel-extension framework that replaces eBPFs in-kernel verifier with a Rust-typed, LLVM-based toolchain to safely express programs that eBPF currently rejects.\nWhy notable — The eBPF verifiers conservatism silently limits what practitioners can implement; Rex shows how a language-level safety guarantee can replace ad-hoc bytecode verification without changing the kernel ABI, opening the door to far richer kernel extensions.\n→ Read paper PageFlex: Flexible and Efficient User-space Delegation of Linux Paging Policies with eBPF Anil Yelam, Kan Wu, Zhiyuan Guo, Suli Yang et al.\nTL;DR — PageFlex lets applications plug in custom page-replacement and allocation policies via eBPF hooks without modifying the kernel, achieving performance competitive with kernel-native policies.\nWhy notable — Memory management policy has historically been locked inside the kernel; PageFlexs eBPF delegation mechanism gives cloud operators a principled way to tailor paging behavior per workload, directly addressing the one-size-fits-all limitation of the Linux page allocator.\n→ Read paper μEFI: A Microkernel-Style UEFI with Isolation and Transparency Le Chen, Yiyang Wu, Jinyu Gu 0001, Yubin Xia et al.\nTL;DR — μEFI restructures UEFI firmware around microkernel principles so that individual UEFI drivers are isolated from each other and from the boot-time trusted computing base.\nWhy notable — Firmware vulnerabilities are notoriously hard to patch and can persist through OS reinstalls; μEFIs approach substantially reduces the blast radius of a compromised UEFI driver while remaining compatible with existing UEFI software, making it directly relevant to secure-boot infrastructure.\n→ Read paper Z-LFS: A Zoned Namespace-tailored Log-structured File System for Commodity Small-zone ZNS SSDs Inhwi Hwang, Sangjin Lee 0003, Sunggon Kim, Hyeonsang Eom et al.\nTL;DR — Z-LFS is an LFS designed around the tight zone-size constraints of commodity ZNS SSDs, using fine-grained segment management and zone-aware garbage collection to avoid the capacity and write-amplification pitfalls of existing approaches.\nWhy notable — ZNS SSDs offer significant cost and endurance advantages but mainstream file systems waste capacity on small-zone devices; Z-LFS shows that rethinking LFS segment layout specifically for small zones yields competitive throughput with substantially lower write amplification.\n→ Read paper Crash Consistency in Block-Level Caching Systems: An Open CAS Case Study Shaohua Duan, Youmin Chen\nTL;DR — A systematic study of crash-consistency bugs in the widely deployed Open CAS block-layer cache, revealing a class of ordering violations that can silently corrupt data on unexpected power loss.\nWhy notable — Block-level caches are invisible to file systems and often assumed to be transparent, making these bugs particularly insidious; the papers taxonomy and detection methodology are directly actionable for operators running NVMe caching in production storage stacks.\n→ Read paper HotRAP: Hot Record Retention and Promotion for LSM-trees with Tiered Storage Jiansheng Qiu, Fangzhou Yuan, Mingyu Gao 0001, Huanchen Zhang\nTL;DR — HotRAP adds a retention-and-promotion layer to LSM-tree compaction that keeps frequently accessed records in faster storage tiers by tracking access heat across compaction boundaries.\nWhy notable — Tiered storage deployments with LSM engines (RocksDB, LevelDB) routinely see hot data demoted to slow tiers during compaction; HotRAPs lightweight heat tracking improves read latency by up to 5x on skewed workloads without changing the external LSM API.\n→ Read paper TGW: Operating an Efficient and Resilient Cloud Gateway at Scale Yifan Yang 0009, Lin He 0004, Jiasheng Zhou, Xiaoyi Shi et al.\nTL;DR — TGW describes the architecture, traffic engineering, and operational lessons of a production cloud gateway handling hundreds of Tbps at a major cloud provider.\nWhy notable — Production-scale gateway papers with real traffic data are rare; TGWs account of how to sustain sub-millisecond failover and linear-scale throughput under adversarial traffic patterns provides a concrete reference design for anyone building or operating large-scale edge infrastructure.\n→ Read paper SwCC: Software-Programmable and Per-Packet Congestion Control in RDMA Engine Hongjing Huang, Jie Zhang 0081, Xuzheng Chen, Ziyu Song et al.\nTL;DR — SwCC embeds a programmable per-packet congestion-control engine directly inside an RDMA NIC, allowing operators to deploy and hot-swap CC algorithms without CPU involvement or ASIC redesign.\nWhy notable — RDMA congestion control has historically been frozen in NIC firmware, forcing cluster-wide firmware upgrades to try new algorithms; SwCCs programmable datapath brings the velocity of software-defined networking to the RDMA layer at near-line-rate performance.\n→ Read paper Opening Up Kernel-Bypass TCP Stacks Shinichi Awamoto, Michio Honda\nTL;DR — A framework that exposes kernel-bypass TCP (DPDK-based) stacks to unmodified POSIX applications by transparently interposing at the syscall level, without requiring application changes or root privileges.\nWhy notable — Kernel-bypass networking has been limited to purpose-built applications; this works zero-modification deployment model makes microsecond-latency TCP accessible to the broad ecosystem of existing networked software, which has significant practical implications for latency-sensitive cloud services.\n→ Read paper Accelerating Nested Virtualization with HyperTurtle Ori Ben Zur, Jakob Krebs, Shai Aviram Bergman, Mark Silberstein\nTL;DR — HyperTurtle reduces the performance penalty of nested virtualization by selectively forwarding L2 guest hypercalls directly to the L0 hypervisor, bypassing the L1 intermediary for common fast paths.\nWhy notable — Nested virtualization is increasingly important for confidential computing and cloud-in-cloud deployments, but the overhead is often prohibitive; HyperTurtles selective bypass approach cuts nested VM overhead by up to 60% on I/O-intensive workloads with no guest modifications.\n→ Read paper KVCache Cache in the Wild: Characterizing and Optimizing KVCache Cache at a Large Cloud Provider Jiahao Wang, Jinbo Han, Xingda Wei, Sijie Shen et al.\nTL;DR — A production measurement study of LLM KV-cache behavior across a large cloud fleet, revealing access patterns and reuse characteristics that inform a redesigned caching policy reducing GPU memory pressure significantly.\nWhy notable — As LLM serving becomes a dominant cloud workload, KV-cache management is a critical bottleneck; this paper provides the first at-scale empirical characterization of KV-cache reuse in a real deployment, and its findings directly shaped policy changes that improved cache hit rates by over 30%.\n→ Read paper GREYHOUND: Hunting Fail-Slows in Hybrid-Parallel Training at Scale Tianyuan Wu, Wei Wang 0030, Yinghao Yu, Siran Yang et al.\nTL;DR — GREYHOUND is a runtime monitoring system that automatically detects and isolates fail-slow stragglers in large-scale hybrid-parallel LLM training jobs before they degrade the entire training run.\nWhy notable — Fail-slow faults are notoriously harder to detect than fail-stop failures and can silently extend training jobs by hours; GREYHOUNDs production deployment demonstrates that fine-grained per-layer timing signals can catch slow nodes within seconds, making it a practical reliability tool for anyone running large training clusters.\n→ Read paper ","wordCount":"1186","inLanguage":"en","datePublished":"2025-07-09T00:00:00Z","dateModified":"2025-07-09T00:00:00Z","author":{"@type":"Person","name":"Publish Assistant"},"mainEntityOfPage":{"@type":"WebPage","@id":"https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/atc-2025/"},"publisher":{"@type":"Organization","name":"Publish Assistant","logo":{"@type":"ImageObject","url":"https://pub.sqrt.fr/vincent/publish-assistant/favicon.ico"}}}</script></head><body id=top><header class=header><nav class=header-nav><div class=logo><a href=https://pub.sqrt.fr/vincent/publish-assistant/ accesskey=h title="Publish Assistant (Alt + H)">Publish Assistant</a>
<span class=logo-sep>/</span>
<a class=logo-topic href=/vincent/publish-assistant/cloud-edge/ title="Edge and Cloud Systems">Edge and Cloud Systems</a><div class=logo-switches><button id=theme-toggle class=theme-toggle accesskey=t title="(Alt + T)" aria-label="Toggle theme">
<svg class="moon" width="18" height="18" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round"><path d="M21 12.79A9 9 0 1111.21 3 7 7 0 0021 12.79z"/></svg>
<svg class="sun" width="18" height="18" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round"><circle cx="12" cy="12" r="5"/><line x1="12" y1="1" x2="12" y2="3"/><line x1="12" y1="21" x2="12" y2="23"/><line x1="4.22" y1="4.22" x2="5.64" y2="5.64"/><line x1="18.36" y1="18.36" x2="19.78" y2="19.78"/><line x1="1" y1="12" x2="3" y2="12"/><line x1="21" y1="12" x2="23" y2="12"/><line x1="4.22" y1="19.78" x2="5.64" y2="18.36"/><line x1="18.36" y1="5.64" x2="19.78" y2="4.22"/></svg></button></div></div><ul id=menu class=menu><li><a href=/vincent/publish-assistant/cloud-edge/venues/ title=Venues><span>Venues</span></a></li><li><a href=/vincent/publish-assistant/cloud-edge/calendar/ title=Calendar><span>Calendar</span></a></li><li><a href=/vincent/publish-assistant/cloud-edge/digests/ title=Digests><span class=active>Digests</span></a></li></ul></nav></header><main class=main><article class=post-single><header class=post-header><nav class=breadcrumbs role=navigation aria-label=Breadcrumb><a href=/vincent/publish-assistant/cloud-edge/digests/>Digests</a>
<svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="feather feather-chevron-right"><polyline points="9 18 15 12 9 6"/></svg></nav><h1 class="post-title entry-hint-parent">ATC 2025 Digest</h1><div class=post-meta><span title='2025-07-09 00:00:00 +0000 UTC'>July 9, 2025</span>&nbsp;·&nbsp;<span>Publish Assistant</span></div></header><div class="post-content md-content"><p>13 papers selected.</p><hr><h3 id=asterinas-a-linux-abi-compatible-rust-based-framekernel-os-with-a-small-and-sound-tcb>ASTERINAS: A Linux ABI-Compatible, Rust-Based Framekernel OS with a Small and Sound TCB<a hidden class=anchor aria-hidden=true href=#asterinas-a-linux-abi-compatible-rust-based-framekernel-os-with-a-small-and-sound-tcb>#</a></h3><p><em>Yuke Peng, Hongliang Tian, Junyang Zhang, Ruihan Li <em>et al.</em></em></p><p><strong>TL;DR</strong> — A production-grade OS kernel written in Rust that exposes a full Linux ABI while confining unsafe code to a small, formally-audited framekernel core.</p><p><strong>Why notable</strong> — ASTERINAS demonstrates that Linux compatibility and memory-safety guarantees are not mutually exclusive — unsafe Rust is isolated to under 5 kloc of framework code, giving systems operators a credible path toward a safer Linux-compatible kernel without sacrificing application portability.</p><p><a href=https://www.usenix.org/conference/atc25/presentation/peng-yuke>→ Read paper</a></p><hr><h3 id=rex-closing-the-language-verifier-gap-with-safe-and-usable-kernel-extensions>Rex: Closing the language-verifier gap with safe and usable kernel extensions<a hidden class=anchor aria-hidden=true href=#rex-closing-the-language-verifier-gap-with-safe-and-usable-kernel-extensions>#</a></h3><p><em>Jinghao Jia, Ruowen Qin, Milo Craun, Egor Lukiyanov <em>et al.</em></em></p><p><strong>TL;DR</strong> — Rex introduces a new kernel-extension framework that replaces eBPF&rsquo;s in-kernel verifier with a Rust-typed, LLVM-based toolchain to safely express programs that eBPF currently rejects.</p><p><strong>Why notable</strong> — The eBPF verifier&rsquo;s conservatism silently limits what practitioners can implement; Rex shows how a language-level safety guarantee can replace ad-hoc bytecode verification without changing the kernel ABI, opening the door to far richer kernel extensions.</p><p><a href=https://www.usenix.org/conference/atc25/presentation/jia>→ Read paper</a></p><hr><h3 id=pageflex-flexible-and-efficient-user-space-delegation-of-linux-paging-policies-with-ebpf>PageFlex: Flexible and Efficient User-space Delegation of Linux Paging Policies with eBPF<a hidden class=anchor aria-hidden=true href=#pageflex-flexible-and-efficient-user-space-delegation-of-linux-paging-policies-with-ebpf>#</a></h3><p><em>Anil Yelam, Kan Wu, Zhiyuan Guo, Suli Yang <em>et al.</em></em></p><p><strong>TL;DR</strong> — PageFlex lets applications plug in custom page-replacement and allocation policies via eBPF hooks without modifying the kernel, achieving performance competitive with kernel-native policies.</p><p><strong>Why notable</strong> — Memory management policy has historically been locked inside the kernel; PageFlex&rsquo;s eBPF delegation mechanism gives cloud operators a principled way to tailor paging behavior per workload, directly addressing the one-size-fits-all limitation of the Linux page allocator.</p><p><a href=https://www.usenix.org/conference/atc25/presentation/yelam>→ Read paper</a></p><hr><h3 id=μefi-a-microkernel-style-uefi-with-isolation-and-transparency>μEFI: A Microkernel-Style UEFI with Isolation and Transparency<a hidden class=anchor aria-hidden=true href=#μefi-a-microkernel-style-uefi-with-isolation-and-transparency>#</a></h3><p><em>Le Chen, Yiyang Wu, Jinyu Gu 0001, Yubin Xia <em>et al.</em></em></p><p><strong>TL;DR</strong> — μEFI restructures UEFI firmware around microkernel principles so that individual UEFI drivers are isolated from each other and from the boot-time trusted computing base.</p><p><strong>Why notable</strong> — Firmware vulnerabilities are notoriously hard to patch and can persist through OS reinstalls; μEFI&rsquo;s approach substantially reduces the blast radius of a compromised UEFI driver while remaining compatible with existing UEFI software, making it directly relevant to secure-boot infrastructure.</p><p><a href=https://www.usenix.org/conference/atc25/presentation/chen-le>→ Read paper</a></p><hr><h3 id=z-lfs-a-zoned-namespace-tailored-log-structured-file-system-for-commodity-small-zone-zns-ssds>Z-LFS: A Zoned Namespace-tailored Log-structured File System for Commodity Small-zone ZNS SSDs<a hidden class=anchor aria-hidden=true href=#z-lfs-a-zoned-namespace-tailored-log-structured-file-system-for-commodity-small-zone-zns-ssds>#</a></h3><p><em>Inhwi Hwang, Sangjin Lee 0003, Sunggon Kim, Hyeonsang Eom <em>et al.</em></em></p><p><strong>TL;DR</strong> — Z-LFS is an LFS designed around the tight zone-size constraints of commodity ZNS SSDs, using fine-grained segment management and zone-aware garbage collection to avoid the capacity and write-amplification pitfalls of existing approaches.</p><p><strong>Why notable</strong> — ZNS SSDs offer significant cost and endurance advantages but mainstream file systems waste capacity on small-zone devices; Z-LFS shows that rethinking LFS segment layout specifically for small zones yields competitive throughput with substantially lower write amplification.</p><p><a href=https://www.usenix.org/conference/atc25/presentation/hwang>→ Read paper</a></p><hr><h3 id=crash-consistency-in-block-level-caching-systems-an-open-cas-case-study>Crash Consistency in Block-Level Caching Systems: An Open CAS Case Study<a hidden class=anchor aria-hidden=true href=#crash-consistency-in-block-level-caching-systems-an-open-cas-case-study>#</a></h3><p><em>Shaohua Duan, Youmin Chen</em></p><p><strong>TL;DR</strong> — A systematic study of crash-consistency bugs in the widely deployed Open CAS block-layer cache, revealing a class of ordering violations that can silently corrupt data on unexpected power loss.</p><p><strong>Why notable</strong> — Block-level caches are invisible to file systems and often assumed to be transparent, making these bugs particularly insidious; the paper&rsquo;s taxonomy and detection methodology are directly actionable for operators running NVMe caching in production storage stacks.</p><p><a href=https://www.usenix.org/conference/atc25/presentation/duan-shaohua>→ Read paper</a></p><hr><h3 id=hotrap-hot-record-retention-and-promotion-for-lsm-trees-with-tiered-storage>HotRAP: Hot Record Retention and Promotion for LSM-trees with Tiered Storage<a hidden class=anchor aria-hidden=true href=#hotrap-hot-record-retention-and-promotion-for-lsm-trees-with-tiered-storage>#</a></h3><p><em>Jiansheng Qiu, Fangzhou Yuan, Mingyu Gao 0001, Huanchen Zhang</em></p><p><strong>TL;DR</strong> — HotRAP adds a retention-and-promotion layer to LSM-tree compaction that keeps frequently accessed records in faster storage tiers by tracking access heat across compaction boundaries.</p><p><strong>Why notable</strong> — Tiered storage deployments with LSM engines (RocksDB, LevelDB) routinely see hot data demoted to slow tiers during compaction; HotRAP&rsquo;s lightweight heat tracking improves read latency by up to 5x on skewed workloads without changing the external LSM API.</p><p><a href=https://www.usenix.org/conference/atc25/presentation/qiu>→ Read paper</a></p><hr><h3 id=tgw-operating-an-efficient-and-resilient-cloud-gateway-at-scale>TGW: Operating an Efficient and Resilient Cloud Gateway at Scale<a hidden class=anchor aria-hidden=true href=#tgw-operating-an-efficient-and-resilient-cloud-gateway-at-scale>#</a></h3><p><em>Yifan Yang 0009, Lin He 0004, Jiasheng Zhou, Xiaoyi Shi <em>et al.</em></em></p><p><strong>TL;DR</strong> — TGW describes the architecture, traffic engineering, and operational lessons of a production cloud gateway handling hundreds of Tbps at a major cloud provider.</p><p><strong>Why notable</strong> — Production-scale gateway papers with real traffic data are rare; TGW&rsquo;s account of how to sustain sub-millisecond failover and linear-scale throughput under adversarial traffic patterns provides a concrete reference design for anyone building or operating large-scale edge infrastructure.</p><p><a href=https://www.usenix.org/conference/atc25/presentation/yang-yifan>→ Read paper</a></p><hr><h3 id=swcc-software-programmable-and-per-packet-congestion-control-in-rdma-engine>SwCC: Software-Programmable and Per-Packet Congestion Control in RDMA Engine<a hidden class=anchor aria-hidden=true href=#swcc-software-programmable-and-per-packet-congestion-control-in-rdma-engine>#</a></h3><p><em>Hongjing Huang, Jie Zhang 0081, Xuzheng Chen, Ziyu Song <em>et al.</em></em></p><p><strong>TL;DR</strong> — SwCC embeds a programmable per-packet congestion-control engine directly inside an RDMA NIC, allowing operators to deploy and hot-swap CC algorithms without CPU involvement or ASIC redesign.</p><p><strong>Why notable</strong> — RDMA congestion control has historically been frozen in NIC firmware, forcing cluster-wide firmware upgrades to try new algorithms; SwCC&rsquo;s programmable datapath brings the velocity of software-defined networking to the RDMA layer at near-line-rate performance.</p><p><a href=https://www.usenix.org/conference/atc25/presentation/huang-hongjing>→ Read paper</a></p><hr><h3 id=opening-up-kernel-bypass-tcp-stacks>Opening Up Kernel-Bypass TCP Stacks<a hidden class=anchor aria-hidden=true href=#opening-up-kernel-bypass-tcp-stacks>#</a></h3><p><em>Shinichi Awamoto, Michio Honda</em></p><p><strong>TL;DR</strong> — A framework that exposes kernel-bypass TCP (DPDK-based) stacks to unmodified POSIX applications by transparently interposing at the syscall level, without requiring application changes or root privileges.</p><p><strong>Why notable</strong> — Kernel-bypass networking has been limited to purpose-built applications; this work&rsquo;s zero-modification deployment model makes microsecond-latency TCP accessible to the broad ecosystem of existing networked software, which has significant practical implications for latency-sensitive cloud services.</p><p><a href=https://www.usenix.org/conference/atc25/presentation/awamoto>→ Read paper</a></p><hr><h3 id=accelerating-nested-virtualization-with-hyperturtle>Accelerating Nested Virtualization with HyperTurtle<a hidden class=anchor aria-hidden=true href=#accelerating-nested-virtualization-with-hyperturtle>#</a></h3><p><em>Ori Ben Zur, Jakob Krebs, Shai Aviram Bergman, Mark Silberstein</em></p><p><strong>TL;DR</strong> — HyperTurtle reduces the performance penalty of nested virtualization by selectively forwarding L2 guest hypercalls directly to the L0 hypervisor, bypassing the L1 intermediary for common fast paths.</p><p><strong>Why notable</strong> — Nested virtualization is increasingly important for confidential computing and cloud-in-cloud deployments, but the overhead is often prohibitive; HyperTurtle&rsquo;s selective bypass approach cuts nested VM overhead by up to 60% on I/O-intensive workloads with no guest modifications.</p><p><a href=https://www.usenix.org/conference/atc25/presentation/zur>→ Read paper</a></p><hr><h3 id=kvcache-cache-in-the-wild-characterizing-and-optimizing-kvcache-cache-at-a-large-cloud-provider>KVCache Cache in the Wild: Characterizing and Optimizing KVCache Cache at a Large Cloud Provider<a hidden class=anchor aria-hidden=true href=#kvcache-cache-in-the-wild-characterizing-and-optimizing-kvcache-cache-at-a-large-cloud-provider>#</a></h3><p><em>Jiahao Wang, Jinbo Han, Xingda Wei, Sijie Shen <em>et al.</em></em></p><p><strong>TL;DR</strong> — A production measurement study of LLM KV-cache behavior across a large cloud fleet, revealing access patterns and reuse characteristics that inform a redesigned caching policy reducing GPU memory pressure significantly.</p><p><strong>Why notable</strong> — As LLM serving becomes a dominant cloud workload, KV-cache management is a critical bottleneck; this paper provides the first at-scale empirical characterization of KV-cache reuse in a real deployment, and its findings directly shaped policy changes that improved cache hit rates by over 30%.</p><p><a href=https://www.usenix.org/conference/atc25/presentation/wang-jiahao>→ Read paper</a></p><hr><h3 id=greyhound-hunting-fail-slows-in-hybrid-parallel-training-at-scale>GREYHOUND: Hunting Fail-Slows in Hybrid-Parallel Training at Scale<a hidden class=anchor aria-hidden=true href=#greyhound-hunting-fail-slows-in-hybrid-parallel-training-at-scale>#</a></h3><p><em>Tianyuan Wu, Wei Wang 0030, Yinghao Yu, Siran Yang <em>et al.</em></em></p><p><strong>TL;DR</strong> — GREYHOUND is a runtime monitoring system that automatically detects and isolates fail-slow stragglers in large-scale hybrid-parallel LLM training jobs before they degrade the entire training run.</p><p><strong>Why notable</strong> — Fail-slow faults are notoriously harder to detect than fail-stop failures and can silently extend training jobs by hours; GREYHOUND&rsquo;s production deployment demonstrates that fine-grained per-layer timing signals can catch slow nodes within seconds, making it a practical reliability tool for anyone running large training clusters.</p><p><a href=https://www.usenix.org/conference/atc25/presentation/wu-tianyuan>→ Read paper</a></p></div><footer class=post-footer><ul class=post-tags><li><a href=https://pub.sqrt.fr/vincent/publish-assistant/tags/operating-systems/>Operating-Systems</a></li><li><a href=https://pub.sqrt.fr/vincent/publish-assistant/tags/cloud/>Cloud</a></li><li><a href=https://pub.sqrt.fr/vincent/publish-assistant/tags/storage/>Storage</a></li><li><a href=https://pub.sqrt.fr/vincent/publish-assistant/tags/networking/>Networking</a></li></ul><nav class=paginav><a class=prev href=https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/hpdc-2025/><span class=title>« Prev</span>
<span>HPDC 2025 Digest</span>
</a><a class=next href=https://pub.sqrt.fr/vincent/publish-assistant/cloud-edge/digests/mobisys-2025/><span class=title>Next »</span>
<span>MobiSys 2025 Digest</span></a></nav></footer></article></main><footer class=footer><span>&copy; 2026 <a href=https://pub.sqrt.fr/vincent/publish-assistant/>Publish Assistant</a></span> ·
<span>Powered by
<a href="https://gohugo.io/?utm_source=papermod" rel=noopener target=_blank>Hugo</a> &
<a href=https://github.com/adityatelange/hugo-PaperMod/ rel=noopener target=_blank>PaperMod</a></span></footer><a href=#top id=top-link class="top-link hidden" aria-label="go to top" title="Go to Top (Alt + G)" accesskey=g><svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="feather feather-chevrons-up"><polyline points="17 11 12 6 7 11"/><polyline points="17 18 12 13 7 18"/></svg>
</a><script>let menu=document.getElementById("menu");if(menu){const e=localStorage.getItem("menu-scroll-position");e&&(menu.scrollLeft=parseInt(e,10)),menu.onscroll=function(){localStorage.setItem("menu-scroll-position",menu.scrollLeft)}}document.querySelectorAll('a[href^="#"]').forEach(e=>{e.addEventListener("click",function(e){e.preventDefault();var t=this.getAttribute("href").substr(1);window.matchMedia("(prefers-reduced-motion: reduce)").matches?document.querySelector(`[id='${decodeURIComponent(t)}']`).scrollIntoView():document.querySelector(`[id='${decodeURIComponent(t)}']`).scrollIntoView({behavior:"smooth"}),t==="top"?history.replaceState(null,null," "):history.pushState(null,null,`#${t}`)})})</script><script>var toplink=document.getElementById("top-link");window.onscroll=function(){const e=window.innerHeight;document.body.scrollTop>e||document.documentElement.scrollTop>e?toplink.classList.remove("hidden"):toplink.classList.add("hidden")}</script><script>document.getElementById("theme-toggle").addEventListener("click",()=>{const e=document.querySelector("html");e.dataset.theme==="dark"?(e.dataset.theme="light",localStorage.setItem("pref-theme","light")):(e.dataset.theme="dark",localStorage.setItem("pref-theme","dark"))})</script></body></html>