knowledge-flow
← Categories

Performance

160 件の記事

2026

11

  1. It's all about the allocator
  2. Writing a DSL for high performance parallel programming in OxCaml
  3. Postgres not using your index? It's counting pages, not rows
  4. Probabilistic data structures in Go
  5. Python 3.15の新プロファイラーTachyonの紹介
  6. ISUCON2026 で優勝しました（準備編）
  7. 並列度を下げたら速くなった — go build の重複コンパイルを実測で見る
  8. 【React】再レンダリングが起きるタイミングと memo / useCallback で防ぐ方法
  9. JIT bitmap text rendering for microcontrollers
  10. Just how far can the FOR UPDATE SKIP LOCKED Postgres queue go?
  11. SSA ベースの .NET IL 最適化コンパイラーを作ってみた
  12. Every Windows 11 app should kill Electron, and Discord just showed why
  13. AWS CLI v2と出力が同じで最大550倍速いawsrをRustで作った
  14. X、Webアプリ基盤をゼロから再構築 ―3つのサーバーを統合し低速回線での表示を大幅にスピードアップ
  15. Blender's UI and Runtime: Much Faster Than GPUI
  16. #27 緊急！ISUCON2026優勝者に聞く、完全自律AIエージェントの作り方と今後のOSS | kawasy | 八百万のOSS
  17. A new Go API for JSON
  18. Shlemiel the Painter Is Back: How Zig’s libc Rewrite Made SQLite Slower
  19. How to speed up the Rust compiler in November 2026
  20. PG Phriday: Burnt TOAST
  21. How we halved DNS build delay and cut memory use by 85%
  22. ISUCON2026 で優勝しました（当日編）
  23. The fastest double-to-string algorithm you’ve never heard of
  24. ISUCON 2026 参加記
  25. The AI cost of JavaScript frameworks
  26. Switch Lowering in GCC

10Rust ビルドの高速化と、Python 3.15 の速度

  1. Go 1.28's HTTP/2 Parking: Measuring Memory Savings
  2. Billion-query-per-second lookup tables in C++
  3. Python's experimental JIT to be removed
  4. Chromium(V8)のArray.prototype.lastIndexOfを最大約26000倍高速化した
  5. C++やRustよりも高速に動作しメモリ使用量も少ないプログラミング言語「Goose」
  6. SIMD and Go
  7. Why WooCommerce Is Slow: 924 ms to Build a Page, 27 ms to Serve It
  8. How close can pure functional programming get to C?
  9. Origin Range Requests: Serving large files without range-request tuning
  10. TTFBは速いのに表示が遅いVercelのページ、関数リージョンが既定のままだった
  11. Faster than earcut: triangulating 3.7M map polygons in 489 ms
  12. Cutting our inference bill by 68% by being obsessive about optimization
  13. 2026 is an exciting year for the Rust compiler (and me)
  14. Making all divisions slower to fix a bug in 0.000000000000000000542% of cases
  15. リトルの法則で考える障害時のキャパシティプランニング
  16. One Request, Many Callers: Coalescing Duplicate fetch() Calls in JavaScript
  17. Pele – Mermaid diagrams as themeable SVG
  18. Vitest v5.0 で新しくなったBenchmarkをCIで実行して比較する - プププなテクブ
  19. Windows 11、8GBメモリのPCも軽快に 基本メモリ使用量を削減
  20. Lambda Web FunctionsでCold Startがどの程度の間隔で起きるのか計測してみた
  21. アプリとDBの間の往復遅延が応答時間に与える影響を TiDB Cloud Starterで計測した
  22. JPEG XL vs AVIF for web use-cases
  23. Better HTTP/1 connection reuse in Go 1.27
  24. Next.jsの公開ページで、1回遊ぶ間に読むchunkを48本から16本にした
  25. Rewinding the exponential: How we keep hardware usage down
  26. LINEスキマニのトップページにおけるパフォーマンス改善（インターンレポート）
  27. Parsing JSON at tens of gigabytes per second… in Python
  28. Making np.searchsorted up to 25× Faster in NumPy 2.5
  29. Making React Context Cheap with React Compiler
  30. Why DuckDB 2.0 is faster
  31. Satori GCをMSBuild SDKとして公開しました！
  32. Introducing on-demand CPU and memory profiling with flamegraphs for Workers and Durable Objects
  33. Lifeguard: A static analyzer for Python lazy imports compatibility
  34. How fast is Python 3.15?
  35. Release mold 3.0.0
  36. How MySQL Implicit Type Conversion Turned an Indexed Query Into a 7M-Row Scan
  37. Emitting metadata early makes building/checking Rust up to twice as fast
  38. Hacking the Go compiler to efficiently map IPv4 to IPv6
  39. Two-Stack Sliding-Window Aggregation
  40. Introducing Workers KV Instant — powered by Quicksilver
  41. vCPUを減らせばVMが速くなる！？ Linux 7.4でマージが期待される“Steal Governor”
  42. RIP, vector database
  43. 5x faster Edge Functions: How we replaced v8 isolates with Firecracker MicroVMs
  44. free-threaded Pythonでasyncioイベントループを複数動かす ―速くなる処理、速くならない処理
  45. How to speed up the Rust compiler in September 2026

09モデルの高速化と、割り当てを減らす最適化

  1. Zero-Growth Stack, Real Gains: How Stack Allocation Can Save 10% CPU in Go
  2. How fast is the web? Explore billions of real-user measurements with BEACON
  3. Ember-1
  4. Adding logit penalty for "wait", "maybe" and "perhaps" to Qwen models improves their accuracy
  5. 【福田昭のセミコン業界最前線】メモリコスト6分の1でAI推論、Samsungの「zNAND-O」
  6. Eliminate First: How We Architected a 60% + BigQuery Cost Reduction
  7. Packing Binary Is Fun, Actually
  8. Improving site performance by shipping more CSS
  9. The Reality of AWS Lambda in High-Load Production
  10. The state of SIMD in Rust in 2026
  11. Optimizing a Lock-Free Ring Buffer
  12. Platform-independent SIMD in Go
  13. EXPLAIN で詰まったときの三つ目の手札、Optimizer Trace の話
  14. 「Claudeを2週間で3倍速にした方法」をAnthropicが解説
  15. ISR pages aren't stale because ISR is broken. They're stale because nobody visited them.
  16. Jev + Generative UI で Web 制作の LLM トークンを 72% 削減できた話（3ページの企業サイトで実測）
  17. Build-Time Syntax Highlighting With Shiki and Rehype
  18. Why didn't anybody tell me about Redis hash slots?
  19. Adaptive Lossless Floating-Point Encoding in Apache Parquet
  20. We ported the original Doom to SQL
  21. OpenAIが10億人以上・40ペタバイト超えのユーザーデータを扱う自社製ストレージシステム「Habitat」について解説
  22. We just shipped support for the ugliest part of HTTP: Vary
  23. AI coding has made CI a bottleneck, so we reworked ours to keep up
  24. Textbook review: Is Parallel Programming Hard, And, If So, What Can You Do About It?
  25. I Benchmarked Five Ways to Speed Up a Slow React Table. Two of Them Did Nothing.
  26. Jev
  27. 【TypeSafe】Jev で書類項目の突合はどこまでできるか？精度・速度・コストを測ってみた
  28. Adversarial examples for fast hash functions
  29. Writing Your Own Allocator in Rust (and Why You Should)
  30. Reverse Engineering ChatGPT Web: How OpenAI Built for a Billion Users
  31. The last mile of a long road: faster NumPy in the browser
  32. You can run git on object storage if you re-make packfiles
  33. I built a free carbon & page-weight auditor for the web — feedback welcome
  34. Qwen3.8-27Bを9分の1に圧縮も性能98.2%維持の「Bonsai 2 27B」
  35. Saving another 100TB of RAM with math (and Rust)
  36. BigQueryのスロット、返し忘れていませんか? fluid scalingによるスロット費用最適化
  37. Jemalloc 5.4.0
  38. Benchmarking Wild vs Mold
  39. Introducing TIN: full-text search for Postgres
  40. モバイルアプリのネイティブとクロスプラットフォームのパフォーマンスを比較する / ネイティブ v.s. Flutter v.s. React Native (Expo)
  41. Training a 4B model to produce 81% faster query plans than Postgres
  42. Why building a Rust LSP is hard
  43. PeakHour 6
  44. Astera Labs Releases Leo 2 CXL Memory Controllers and Leo X Controller for Rackscale Fabric-Attached Memory
  45. Performance Improvements in .NET 11
  46. React 互換の軽量ランタイム TanStack Redact とは
  47. Amazon S3で「ストレージコスト80％削減」――たった6カ月で成功した“再現性のある方法”とは？
  48. Principles for Fast Tokio Applications
  49. The case against JPEG XL
  50. 40ms Go GC stop-the-world pauses caused by swap
  51. SpotifyのエンジニアがClaude Codeのトークン消費を約90％削減した方法とは?
  52. DeepSeek v4.1 Flashを動かしたくてFP4に対応していないA100を、33tok/sから673 tok/sまで持っていって気がつくと公式APIより速くなっていた話
  53. I made a build visualizer to understand Bun’s compile times
  54. terraform plan -minimal-refreshでplanを爆速にする
  55. 118M Queries per Second on Neki
  56. 単体テストの実行時間を6割程度削減してみた
  57. JEP 544: Ahead-of-Time Code Compilation
  58. React Native ExecuTorch is now up to 92x faster
  59. Automatic Key Exchange: faster, post-quantum secure origin handshakes for 45 billion daily connections (and counting)
  60. Swap, memory pressure and the Go GC
  61. This Month in Ladybird - August 2026
  62. From a Chocolate Wrapper to Concurrent InnoDB Page Splits
  63. Railsフレームワークを本格的にRactorに対応させるShopifyの取り組み（翻訳）
  64. JavaScriptを1文字ずつ削る。手ごとの削減幅を実測した10個と、壊れる境界
  65. Debian Code Search: Fast TurboPFor with Go SIMD
  66. Language Models Can Control Their Own Attention
  67. C++26: std::hive
  68. Ok, but does it scale?
  69. The NX bit is not just about security
  70. TanStack Virtual を使ってみた話
  71. How Swiss Tables Work in Go's Built-in Map
  72. BraveがChrome・Edge・Firefoxを速度と性能の両方で上回っているというベンチマーク結果を公開
  73. EAS Observe
  74. Static Allocation, Constant Work
  75. How we could save petabytes of cache storage with Zstandard and Pingora
  76. Wasmi 2.0 - Engineering of the Fastest Wasm Interpreters
  77. io_uring without readahead
  78. The Browser's Main Thread Is Expensive

08「ソフトはもう遅くない」論争と、メモリ削減

  1. Samsung's Processing-in-Memory (PIM)
  2. Zod v4.5 adds schema compilation (3-9x faster validation)
  3. How we saved 100 terabytes of memory by optimizing 1.1.1.1's DNS cache
  4. Understanding useMemo and useCallback
  5. 「Google Chrome 152」安定版リリース、CPU性能に応じてウェブサイトを最適化できる「CPU Performance API」が追加
  6. Replacing a Rust Enum with a 64-bit Word Made My Interpreter 17% Faster
  7. How I Made a Canvas JSON Viewer Fast with Viewport Virtualization
  8. There's no reason for software to be slow anymore
  9. There continue to be reasons for software to be slow
  10. Rust Glancer: Rust LSP using 100x less RAM
  11. Compile-Time Improvements in LLVM 23