tamnd commited on
Commit
27f2a54
·
verified ·
1 Parent(s): 37ceb5b

Publish 2 shards CC-MAIN-2026-08/00176–01017

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 **16,232,818 documents across 839 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 839 files of CC-MAIN-2026-08 (16,232,818 pages total), projected to the full crawl of 100,000 WARC files.
168
 
169
- | Stage | 839 files (measured) | 100,000 files (projected) | Reduction |
170
  |---|---|---|---|
171
- | Raw WARC (.warc.gz, downloaded) | ~680.0 GB | ~83 TB | — |
172
  | HTML extracted (uncompressed) | 2.0 TB | ~295 TB | — |
173
- | Packed markdown WARC (.md.warc.gz) | ~34.4 GB | ~3.7 TB | **-98.3%** vs HTML |
174
  | Final Parquet (Zstd level 19) | 23.3 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 2.0 TB of uncompressed HTML down to 73.2 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: ~680.0 GB of raw gzipped WARCs becomes **23.3 GB of Parquet** — a **96.6% total reduction** — containing 16,232,818 clean markdown documents.
179
 
180
  ### Processing Times
181
 
182
- Pipeline timings across 839 shards of CC-MAIN-2026-08:
183
 
184
  ```
185
- Download (raw WARC) ████████████████████████ total 48h 54m 39s avg 3m 29s
186
  Convert (HTML → MD) █░░░░░░░░░░░░░░░░░░░░░░░ total 3h 56m 33s avg 16s
187
- Export (Parquet) █████░░░░░░░░░░░░░░░░░░░ total 10h 57m 16s avg 47s
188
  Publish (HuggingFace) ███░░░░░░░░░░░░░░░░░░░░░ total 6h 9m 33s avg 26s
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 **16,271,452 documents across 841 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 841 files of CC-MAIN-2026-08 (16,271,452 pages total), projected to the full crawl of 100,000 WARC files.
168
 
169
+ | Stage | 841 files (measured) | 100,000 files (projected) | Reduction |
170
  |---|---|---|---|
171
+ | Raw WARC (.warc.gz, downloaded) | ~681.7 GB | ~83 TB | — |
172
  | HTML extracted (uncompressed) | 2.0 TB | ~295 TB | — |
173
+ | Packed markdown WARC (.md.warc.gz) | ~34.5 GB | ~3.7 TB | **-98.3%** vs HTML |
174
  | Final Parquet (Zstd level 19) | 23.3 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 2.0 TB of uncompressed HTML down to 73.4 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: ~681.7 GB of raw gzipped WARCs becomes **23.3 GB of Parquet** — a **96.6% total reduction** — containing 16,271,452 clean markdown documents.
179
 
180
  ### Processing Times
181
 
182
+ Pipeline timings across 841 shards of CC-MAIN-2026-08:
183
 
184
  ```
185
+ Download (raw WARC) ████████████████████████ total 49h 9m 46s avg 3m 30s
186
  Convert (HTML → MD) █░░░░░░░░░░░░░░░░░░░░░░░ total 3h 56m 33s avg 16s
187
+ Export (Parquet) █████░░░░░░░░░░░░░░░░░░░ total 11h 0m 35s avg 47s
188
  Publish (HuggingFace) ███░░░░░░░░░░░░░░░░░░░░░ total 6h 9m 33s avg 26s
189
  ```
190
 
data/CC-MAIN-2026-08/00176.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5f160ddb340113f505cc2e53e1ccdbe30d9f97affb13170d5ad629044c6c772
3
+ size 30342108
data/CC-MAIN-2026-08/01017.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d751eff2192bfbdde93d842372a051c1e4d4271724dddc3f3776def55ef0825
3
+ size 28776192
stats.csv CHANGED
@@ -176,6 +176,7 @@ CC-MAIN-2026-08,172,19382,2557076183,96701504,30420156,2026-03-13T18:20:57Z,490,
176
  CC-MAIN-2026-08,173,19312,2574748357,94840263,30552947,2026-03-13T18:32:57Z,615,0,80,28
177
  CC-MAIN-2026-08,174,19430,2594383888,94151433,30292019,2026-03-13T18:41:24Z,426,0,97,25
178
  CC-MAIN-2026-08,175,19361,2535880308,90825657,29300682,2026-03-13T18:51:11Z,500,0,82,20
 
179
  CC-MAIN-2026-08,250,19200,2541559671,94959737,30214413,2026-03-13T17:42:47Z,511,0,85,19
180
  CC-MAIN-2026-08,251,19610,2594898345,97558492,31917087,2026-03-13T06:42:12Z,277,0,23,136
181
  CC-MAIN-2026-08,252,19240,2525928701,98163046,31104034,2026-03-13T06:47:10Z,238,0,49,10
@@ -504,6 +505,7 @@ CC-MAIN-2026-08,1013,19632,2589446285,96006808,30120430,2026-03-13T18:23:56Z,336
504
  CC-MAIN-2026-08,1014,19503,2566088846,94039374,29397377,2026-03-13T18:32:26Z,429,0,88,26
505
  CC-MAIN-2026-08,1015,19385,2577073795,90909836,28814674,2026-03-13T18:41:07Z,408,0,97,12
506
  CC-MAIN-2026-08,1016,19688,2601213579,96284971,29536751,2026-03-13T18:51:11Z,531,0,79,20
 
507
  CC-MAIN-2026-08,1101,19721,2641197765,95719568,30240239,2026-03-13T09:59:33Z,564,0,296,0
508
  CC-MAIN-2026-08,1102,19562,2608129691,99280759,31748974,2026-03-13T10:28:07Z,1518,0,191,0
509
  CC-MAIN-2026-08,1103,19270,2517580674,89253461,28361470,2026-03-13T10:55:39Z,1380,0,268,0
 
176
  CC-MAIN-2026-08,173,19312,2574748357,94840263,30552947,2026-03-13T18:32:57Z,615,0,80,28
177
  CC-MAIN-2026-08,174,19430,2594383888,94151433,30292019,2026-03-13T18:41:24Z,426,0,97,25
178
  CC-MAIN-2026-08,175,19361,2535880308,90825657,29300682,2026-03-13T18:51:11Z,500,0,82,20
179
+ CC-MAIN-2026-08,176,19391,2554258489,97225767,30342108,2026-03-13T19:00:26Z,460,0,98,0
180
  CC-MAIN-2026-08,250,19200,2541559671,94959737,30214413,2026-03-13T17:42:47Z,511,0,85,19
181
  CC-MAIN-2026-08,251,19610,2594898345,97558492,31917087,2026-03-13T06:42:12Z,277,0,23,136
182
  CC-MAIN-2026-08,252,19240,2525928701,98163046,31104034,2026-03-13T06:47:10Z,238,0,49,10
 
505
  CC-MAIN-2026-08,1014,19503,2566088846,94039374,29397377,2026-03-13T18:32:26Z,429,0,88,26
506
  CC-MAIN-2026-08,1015,19385,2577073795,90909836,28814674,2026-03-13T18:41:07Z,408,0,97,12
507
  CC-MAIN-2026-08,1016,19688,2601213579,96284971,29536751,2026-03-13T18:51:11Z,531,0,79,20
508
+ CC-MAIN-2026-08,1017,19243,2549255009,91047764,28776192,2026-03-13T19:00:26Z,447,0,101,0
509
  CC-MAIN-2026-08,1101,19721,2641197765,95719568,30240239,2026-03-13T09:59:33Z,564,0,296,0
510
  CC-MAIN-2026-08,1102,19562,2608129691,99280759,31748974,2026-03-13T10:28:07Z,1518,0,191,0
511
  CC-MAIN-2026-08,1103,19270,2517580674,89253461,28361470,2026-03-13T10:55:39Z,1380,0,268,0