tamnd commited on
Commit
4557f2a
·
verified ·
1 Parent(s): ce8989e

Publish 2 shards CC-MAIN-2026-08/00192–00353

Browse files
README.md CHANGED
@@ -32,7 +32,7 @@ configs:
32
 
33
  Open Index is a large-scale web text dataset built from [Common Crawl](https://commoncrawl.org). Every page goes through a pipeline that extracts the main content from raw HTML, converts it to clean Markdown using [trafilatura](https://github.com/adbar/trafilatura), and packages the result into Parquet files with full WARC metadata preserved.
34
 
35
- The dataset currently includes crawl **CC-MAIN-2026-08** with **18,061,800 documents across 933 shards**. We plan to add more snapshots over time.
36
 
37
  Open Index is released under the **Open Data Commons Attribution License (ODC-By) v1.0**, the same license used by Common Crawl.
38
 
@@ -164,27 +164,27 @@ Empty conversions (pages where trafilatura could not extract meaningful content)
164
 
165
  ### Compression Ratios
166
 
167
- Numbers below are actual measurements summed across all 933 files of CC-MAIN-2026-08 (18,061,800 pages total), projected to the full crawl of 100,000 WARC files.
168
 
169
- | Stage | 933 files (measured) | 100,000 files (projected) | Reduction |
170
  |---|---|---|---|
171
- | Raw WARC (.warc.gz, downloaded) | ~756.2 GB | ~83 TB | — |
172
  | HTML extracted (uncompressed) | 2.2 TB | ~295 TB | — |
173
- | Packed markdown WARC (.md.warc.gz) | ~38.3 GB | ~3.7 TB | **-98.3%** vs HTML |
174
  | Final Parquet (Zstd level 19) | 25.9 GB | ~2.9 TB | **-32.4%** vs packed WARC |
175
 
176
- The big win is the HTML → Markdown step: trafilatura strips all tags, scripts, styles, navigation, and ads, keeping only the main content. This cuts 2.2 TB of uncompressed HTML down to 81.5 GB of markdown — a **98.3% reduction** — before any file-level compression is applied. Parquet with Zstd level 19 then compresses the markdown a further 68.2%.
177
 
178
- End to end: ~756.2 GB of raw gzipped WARCs becomes **25.9 GB of Parquet** — a **96.6% total reduction** — containing 18,061,800 clean markdown documents.
179
 
180
  ### Processing Times
181
 
182
- Pipeline timings across 933 shards of CC-MAIN-2026-08:
183
 
184
  ```
185
- Download (raw WARC) ████████████████████████ total 60h 49m 49s avg 3m 54s
186
- Convert (HTML → MD) █░░░░░░░░░░░░░░░░░░░░░░░ total 4h 47m 46s avg 18s
187
- Export (Parquet) █████░░░░░░░░░░░░░░░░░░░ total 13h 57m 10s avg 53s
188
  Publish (HuggingFace) ██░░░░░░░░░░░░░░░░░░░░░░ total 6h 42m 30s avg 25s
189
  ```
190
 
 
32
 
33
  Open Index is a large-scale web text dataset built from [Common Crawl](https://commoncrawl.org). Every page goes through a pipeline that extracts the main content from raw HTML, converts it to clean Markdown using [trafilatura](https://github.com/adbar/trafilatura), and packages the result into Parquet files with full WARC metadata preserved.
34
 
35
+ The dataset currently includes crawl **CC-MAIN-2026-08** with **18,100,775 documents across 935 shards**. We plan to add more snapshots over time.
36
 
37
  Open Index is released under the **Open Data Commons Attribution License (ODC-By) v1.0**, the same license used by Common Crawl.
38
 
 
164
 
165
  ### Compression Ratios
166
 
167
+ Numbers below are actual measurements summed across all 935 files of CC-MAIN-2026-08 (18,100,775 pages total), projected to the full crawl of 100,000 WARC files.
168
 
169
+ | Stage | 935 files (measured) | 100,000 files (projected) | Reduction |
170
  |---|---|---|---|
171
+ | Raw WARC (.warc.gz, downloaded) | ~757.9 GB | ~83 TB | — |
172
  | HTML extracted (uncompressed) | 2.2 TB | ~295 TB | — |
173
+ | Packed markdown WARC (.md.warc.gz) | ~38.4 GB | ~3.7 TB | **-98.3%** vs HTML |
174
  | Final Parquet (Zstd level 19) | 25.9 GB | ~2.9 TB | **-32.4%** vs packed WARC |
175
 
176
+ The big win is the HTML → Markdown step: trafilatura strips all tags, scripts, styles, navigation, and ads, keeping only the main content. This cuts 2.2 TB of uncompressed HTML down to 81.6 GB of markdown — a **98.3% reduction** — before any file-level compression is applied. Parquet with Zstd level 19 then compresses the markdown a further 68.2%.
177
 
178
+ End to end: ~757.9 GB of raw gzipped WARCs becomes **25.9 GB of Parquet** — a **96.6% total reduction** — containing 18,100,775 clean markdown documents.
179
 
180
  ### Processing Times
181
 
182
+ Pipeline timings across 935 shards of CC-MAIN-2026-08:
183
 
184
  ```
185
+ Download (raw WARC) ████████████████████████ total 60h 58m 16s avg 3m 54s
186
+ Convert (HTML → MD) █░░░░░░░░░░░░░░░░░░░░░░░ total 4h 56m 25s avg 19s
187
+ Export (Parquet) █████░░░░░░░░░░░░░░░░░░░ total 14h 0m 40s avg 53s
188
  Publish (HuggingFace) ██░░░░░░░░░░░░░░░░░░░░░░ total 6h 42m 30s avg 25s
189
  ```
190
 
data/CC-MAIN-2026-08/00192.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62024666836703e73f8f44af261e5e409bb5939f14b070f4a8ce6f49009f6180
3
+ size 29228248
data/CC-MAIN-2026-08/00353.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1cdac189c54cd8a257b9621bca282eae324963c75c29e1a77b2433c05d03ee5
3
+ size 30335173
stats.csv CHANGED
@@ -192,6 +192,7 @@ CC-MAIN-2026-08,188,19353,2586975300,95129494,29885882,2026-03-13T20:56:38Z,596,
192
  CC-MAIN-2026-08,189,19332,2543793720,95695763,30417911,2026-03-13T21:06:56Z,568,0,84,16
193
  CC-MAIN-2026-08,190,19346,2567101082,94155671,29866255,2026-03-13T21:16:12Z,437,0,99,21
194
  CC-MAIN-2026-08,191,19387,2570610546,100901027,30983932,2026-03-13T21:29:14Z,631,0,143,37
 
195
  CC-MAIN-2026-08,250,19200,2541559671,94959737,30214413,2026-03-13T17:42:47Z,511,0,85,19
196
  CC-MAIN-2026-08,251,19610,2594898345,97558492,31917087,2026-03-13T06:42:12Z,277,0,23,136
197
  CC-MAIN-2026-08,252,19240,2525928701,98163046,31104034,2026-03-13T06:47:10Z,238,0,49,10
@@ -295,6 +296,7 @@ CC-MAIN-2026-08,349,19533,2591710395,93588653,29936978,2026-03-13T20:55:15Z,0,57
295
  CC-MAIN-2026-08,350,19432,2562615888,91646102,29091302,2026-03-13T21:06:26Z,479,0,177,14
296
  CC-MAIN-2026-08,351,19601,2571162986,96301161,32049090,2026-03-13T21:15:41Z,0,435,120,24
297
  CC-MAIN-2026-08,352,19231,2557508135,96230865,30275599,2026-03-13T21:28:41Z,628,0,160,29
 
298
  CC-MAIN-2026-08,401,19441,2586676962,94334130,29618260,2026-03-13T11:41:00Z,0,0,33,0
299
  CC-MAIN-2026-08,402,19446,2552785517,96654791,30560442,2026-03-13T11:49:21Z,0,443,57,0
300
  CC-MAIN-2026-08,403,19320,2579048144,91406292,29341198,2026-03-13T11:56:46Z,0,414,29,0
 
192
  CC-MAIN-2026-08,189,19332,2543793720,95695763,30417911,2026-03-13T21:06:56Z,568,0,84,16
193
  CC-MAIN-2026-08,190,19346,2567101082,94155671,29866255,2026-03-13T21:16:12Z,437,0,99,21
194
  CC-MAIN-2026-08,191,19387,2570610546,100901027,30983932,2026-03-13T21:29:14Z,631,0,143,37
195
+ CC-MAIN-2026-08,192,19505,2557456629,92443840,29228248,2026-03-13T21:39:11Z,507,0,105,0
196
  CC-MAIN-2026-08,250,19200,2541559671,94959737,30214413,2026-03-13T17:42:47Z,511,0,85,19
197
  CC-MAIN-2026-08,251,19610,2594898345,97558492,31917087,2026-03-13T06:42:12Z,277,0,23,136
198
  CC-MAIN-2026-08,252,19240,2525928701,98163046,31104034,2026-03-13T06:47:10Z,238,0,49,10
 
296
  CC-MAIN-2026-08,350,19432,2562615888,91646102,29091302,2026-03-13T21:06:26Z,479,0,177,14
297
  CC-MAIN-2026-08,351,19601,2571162986,96301161,32049090,2026-03-13T21:15:41Z,0,435,120,24
298
  CC-MAIN-2026-08,352,19231,2557508135,96230865,30275599,2026-03-13T21:28:41Z,628,0,160,29
299
+ CC-MAIN-2026-08,353,19470,2597274403,94820809,30335173,2026-03-13T21:39:11Z,0,519,105,0
300
  CC-MAIN-2026-08,401,19441,2586676962,94334130,29618260,2026-03-13T11:41:00Z,0,0,33,0
301
  CC-MAIN-2026-08,402,19446,2552785517,96654791,30560442,2026-03-13T11:49:21Z,0,443,57,0
302
  CC-MAIN-2026-08,403,19320,2579048144,91406292,29341198,2026-03-13T11:56:46Z,0,414,29,0