Skip to content

turbo-persistence: drop key compression dictionary from SST files#90608

Merged
sokra merged 3 commits intocanaryfrom
sokra/streaming-compaction
Feb 26, 2026
Merged

turbo-persistence: drop key compression dictionary from SST files#90608
sokra merged 3 commits intocanaryfrom
sokra/streaming-compaction

Conversation

@lukesandberg
Copy link
Contributor

@lukesandberg lukesandberg commented Feb 26, 2026

Summary

  • Remove the zstd key compression dictionary from SST files, simplifying the on-disk format
  • Key blocks now use plain LZ4 compression (same as value blocks), eliminating the zstd dependency
  • Update the turbo-persistence README to reflect the format changes

Why

The key compression dictionary added complexity (zstd dependency, dictionary computation, dual decompression paths) with diminishing returns. Plain LZ4 is already used for value blocks and provides sufficient compression for key blocks. This simplifies both the SST file format and the compaction/inspection tooling.

This main goal is to unblock streaming SST writes.

What Changed

On-disk format (breaking):

  • SST files no longer start with a serialized key compression dictionary
  • Meta file entries no longer include the key_compression_dictionary_length field

Code:

  • Removed zstd dependency from turbo-persistence
  • Removed dictionary computation in StaticSortedFileBuilder
  • Simplified sst_inspect tool (no more dictionary-aware decompression)
  • Removed key_compression_dictionary_length from StaticSortedFileMetaData, MetaEntry, and MetaFileBuilder

Docs:

  • Updated README meta file format (removed dictionary length field)
  • Updated README SST file format (removed dictionary region)
  • Updated compression description to reference LZ4 directly

Test Plan

  • Existing turbo-persistence tests and benchmarks pass with the format changes
  • sst_inspect tool updated to work without dictionary logic

sokra and others added 2 commits February 26, 2026 18:43
Key blocks are now compressed with plain LZ4 (no dictionary), same as
value blocks. This removes the zstd dependency (used only for dictionary
building) and simplifies the SST file format by eliminating the
dictionary region at the start of each file.

Breaking change to the meta file and SST on-disk format.
…n dictionary

Remove references to the key compression dictionary from the meta file
and SST file format descriptions, and update the small value block
compression note to no longer mention dictionaries.
@nextjs-bot nextjs-bot added created-by: Turbopack team PRs by the Turbopack team. Turbopack Related to Turbopack with Next.js. labels Feb 26, 2026
@lukesandberg lukesandberg requested a review from sokra February 26, 2026 17:52
@nextjs-bot
Copy link
Collaborator

nextjs-bot commented Feb 26, 2026

Tests Passed

@nextjs-bot
Copy link
Collaborator

Stats from current PR

✅ No significant changes detected

📊 All Metrics
📖 Metrics Glossary

Dev Server Metrics:

  • Listen = TCP port starts accepting connections
  • First Request = HTTP server returns successful response
  • Cold = Fresh build (no cache)
  • Warm = With cached build artifacts

Build Metrics:

  • Fresh = Clean build (no .next directory)
  • Cached = With existing .next directory

Change Thresholds:

  • Time: Changes < 50ms AND < 10%, OR < 2% are insignificant
  • Size: Changes < 1KB AND < 1% are insignificant
  • All other changes are flagged to catch regressions

⚡ Dev Server

Metric Canary PR Change Trend
Cold (Listen) 455ms 456ms ▁▁█▁▂
Cold (Ready in log) 438ms 439ms ▁▁█▁▂
Cold (First Request) 1.228s 1.268s ▃▁█▁▃
Warm (Listen) 456ms 457ms ▁▁█▁▂
Warm (Ready in log) 443ms 444ms ▁▁█▁▂
Warm (First Request) 343ms 341ms ▁▁█▁▃
📦 Dev Server (Webpack) (Legacy)

📦 Dev Server (Webpack)

Metric Canary PR Change Trend
Cold (Listen) 456ms 456ms ▁▁█▁▁
Cold (Ready in log) 458ms 457ms ▄▆█▆▁
Cold (First Request) 1.957s 1.959s ▃▄█▅▁
Warm (Listen) 457ms 456ms ▁▁█▁▁
Warm (Ready in log) 460ms 456ms ▆▅█▄▁
Warm (First Request) 1.972s 1.982s ▄▄█▅▁

⚡ Production Builds

Metric Canary PR Change Trend
Fresh Build 3.881s 3.880s ▁▁█▁▃
Cached Build 3.883s 3.917s ▁▁█▁▃
📦 Production Builds (Webpack) (Legacy)

📦 Production Builds (Webpack)

Metric Canary PR Change Trend
Fresh Build 14.567s 14.563s ▁▁█▂▁
Cached Build 14.687s 14.641s ▁▁█▂▁
node_modules Size 475 MB 475 MB ▁▁▁▁▁
📦 Bundle Sizes

Bundle Sizes

⚡ Turbopack

Client

Main Bundles: **400 kB** → **400 kB** ✅ -13 B

80 files with content-based hashes (individual files not comparable between builds)

Server

Middleware
Canary PR Change
middleware-b..fest.js gzip 764 B 763 B
Total 764 B 763 B ✅ -1 B
Build Details
Build Manifests
Canary PR Change
_buildManifest.js gzip 451 B 452 B
Total 451 B 452 B ⚠️ +1 B

📦 Webpack

Client

Main Bundles
Canary PR Change
5528-HASH.js gzip 5.54 kB N/A -
6280-HASH.js gzip 58.3 kB N/A -
6335.HASH.js gzip 169 B N/A -
912-HASH.js gzip 4.59 kB N/A -
e8aec2e4-HASH.js gzip 62.6 kB N/A -
framework-HASH.js gzip 59.7 kB 59.7 kB
main-app-HASH.js gzip 255 B 251 B 🟢 4 B (-2%)
main-HASH.js gzip 39.1 kB 39.1 kB
webpack-HASH.js gzip 1.68 kB 1.68 kB
262-HASH.js gzip N/A 4.59 kB -
2889.HASH.js gzip N/A 169 B -
5602-HASH.js gzip N/A 5.55 kB -
6948ada0-HASH.js gzip N/A 62.6 kB -
9544-HASH.js gzip N/A 59 kB -
Total 232 kB 233 kB ⚠️ +717 B
Polyfills
Canary PR Change
polyfills-HASH.js gzip 39.4 kB 39.4 kB
Total 39.4 kB 39.4 kB
Pages
Canary PR Change
_app-HASH.js gzip 194 B 194 B
_error-HASH.js gzip 183 B 180 B 🟢 3 B (-2%)
css-HASH.js gzip 331 B 330 B
dynamic-HASH.js gzip 1.81 kB 1.81 kB
edge-ssr-HASH.js gzip 256 B 256 B
head-HASH.js gzip 351 B 352 B
hooks-HASH.js gzip 384 B 383 B
image-HASH.js gzip 580 B 581 B
index-HASH.js gzip 260 B 260 B
link-HASH.js gzip 2.5 kB 2.5 kB
routerDirect..HASH.js gzip 320 B 319 B
script-HASH.js gzip 386 B 386 B
withRouter-HASH.js gzip 315 B 315 B
1afbb74e6ecf..834.css gzip 106 B 106 B
Total 7.97 kB 7.97 kB ✅ -2 B

Server

Edge SSR
Canary PR Change
edge-ssr.js gzip 125 kB 125 kB
page.js gzip 254 kB 254 kB
Total 379 kB 379 kB ⚠️ +327 B
Middleware
Canary PR Change
middleware-b..fest.js gzip 616 B 614 B
middleware-r..fest.js gzip 156 B 155 B
middleware.js gzip 43.8 kB 43.9 kB
edge-runtime..pack.js gzip 842 B 842 B
Total 45.4 kB 45.5 kB ⚠️ +99 B
Build Details
Build Manifests
Canary PR Change
_buildManifest.js gzip 715 B 718 B
Total 715 B 718 B ⚠️ +3 B
Build Cache
Canary PR Change
0.pack gzip 4.01 MB 4.03 MB 🔴 +14.1 kB (+0%)
index.pack gzip 103 kB 102 kB
index.pack.old gzip 104 kB 102 kB 🟢 1.35 kB (-1%)
Total 4.22 MB 4.23 MB ⚠️ +12 kB

🔄 Shared (bundler-independent)

Runtimes
Canary PR Change
app-page-exp...dev.js gzip 320 kB 320 kB
app-page-exp..prod.js gzip 170 kB 170 kB
app-page-tur...dev.js gzip 319 kB 319 kB
app-page-tur..prod.js gzip 169 kB 169 kB
app-page-tur...dev.js gzip 316 kB 316 kB
app-page-tur..prod.js gzip 168 kB 168 kB
app-page.run...dev.js gzip 316 kB 316 kB
app-page.run..prod.js gzip 168 kB 168 kB
app-route-ex...dev.js gzip 70.8 kB 70.8 kB
app-route-ex..prod.js gzip 49.2 kB 49.2 kB
app-route-tu...dev.js gzip 70.8 kB 70.8 kB
app-route-tu..prod.js gzip 49.2 kB 49.2 kB
app-route-tu...dev.js gzip 70.4 kB 70.4 kB
app-route-tu..prod.js gzip 49 kB 49 kB
app-route.ru...dev.js gzip 70.4 kB 70.4 kB
app-route.ru..prod.js gzip 49 kB 49 kB
dist_client_...dev.js gzip 324 B 324 B
dist_client_...dev.js gzip 326 B 326 B
dist_client_...dev.js gzip 318 B 318 B
dist_client_...dev.js gzip 317 B 317 B
pages-api-tu...dev.js gzip 43.2 kB 43.2 kB
pages-api-tu..prod.js gzip 32.9 kB 32.9 kB
pages-api.ru...dev.js gzip 43.2 kB 43.2 kB
pages-api.ru..prod.js gzip 32.8 kB 32.8 kB
pages-turbo....dev.js gzip 52.5 kB 52.5 kB
pages-turbo...prod.js gzip 38.5 kB 38.5 kB
pages.runtim...dev.js gzip 52.5 kB 52.5 kB
pages.runtim..prod.js gzip 38.4 kB 38.4 kB
server.runti..prod.js gzip 62 kB 62 kB
Total 2.82 MB 2.82 MB ⚠️ +2 B
📎 Tarball URL
next@https://vercel-packages.vercel.app/next/prs/90608/next

@codspeed-hq
Copy link

codspeed-hq bot commented Feb 26, 2026

Merging this PR will not alter performance

✅ 17 untouched benchmarks
⏩ 3 skipped benchmarks1


Comparing sokra/streaming-compaction (9ad1e83) with canary (1cf02b2)

Open in CodSpeed

Footnotes

  1. 3 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports.

Remove leftover code from the key compression dictionary removal:

- Remove `blocks_start()` (always returned 0) and inline callers
- Remove dictionary parameters from `read_block()`, `decompress_into_arc()`,
  and `compress_into_buffer()`
- Remove `dict` and `long_term` fields from `CompressionConfig` enum
- Drop unused `_total_key_size` parameter from `write_static_stored_file()`
  and its call sites
- Simplify `compress_into_buffer()` to use `lz4::compress_to_vec()` directly
  instead of constructing a `Compressor` object
@lukesandberg lukesandberg marked this pull request as ready for review February 26, 2026 19:12
@sokra sokra merged commit d52877e into canary Feb 26, 2026
160 of 161 checks passed
@sokra sokra deleted the sokra/streaming-compaction branch February 26, 2026 20:24
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

created-by: Turbopack team PRs by the Turbopack team. Turbopack Related to Turbopack with Next.js.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants