Publish shard CC-MAIN-2026-08/01268
Browse files- README.md +10 -10
- data/CC-MAIN-2026-08/01268.parquet +3 -0
- stats.csv +1 -0
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 **17,
|
| 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
|
| 168 |
|
| 169 |
-
| Stage |
|
| 170 |
|---|---|---|---|
|
| 171 |
-
| Raw WARC (.warc.gz, downloaded) | ~
|
| 172 |
| HTML extracted (uncompressed) | 2.1 TB | ~295 TB | — |
|
| 173 |
-
| Packed markdown WARC (.md.warc.gz) | ~36.
|
| 174 |
| Final Parquet (Zstd level 19) | 24.6 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 2.1 TB of uncompressed HTML down to 77.
|
| 177 |
|
| 178 |
-
End to end: ~
|
| 179 |
|
| 180 |
### Processing Times
|
| 181 |
|
| 182 |
-
Pipeline timings across
|
| 183 |
|
| 184 |
```
|
| 185 |
-
Download (raw WARC) ████████████████████████ total 54h
|
| 186 |
Convert (HTML → MD) █░░░░░░░░░░░░░░░░░░░░░░░ total 4h 30m 54s avg 18s
|
| 187 |
-
Export (Parquet) █████░░░░░░░░░░░░░░░░░░░ total 12h
|
| 188 |
Publish (HuggingFace) ██░░░░░░░░░░░░░░░░░░░░░░ total 6h 26m 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 **17,205,106 documents across 889 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 889 files of CC-MAIN-2026-08 (17,205,106 pages total), projected to the full crawl of 100,000 WARC files.
|
| 168 |
|
| 169 |
+
| Stage | 889 files (measured) | 100,000 files (projected) | Reduction |
|
| 170 |
|---|---|---|---|
|
| 171 |
+
| Raw WARC (.warc.gz, downloaded) | ~720.6 GB | ~83 TB | — |
|
| 172 |
| HTML extracted (uncompressed) | 2.1 TB | ~295 TB | — |
|
| 173 |
+
| Packed markdown WARC (.md.warc.gz) | ~36.5 GB | ~3.7 TB | **-98.3%** vs HTML |
|
| 174 |
| Final Parquet (Zstd level 19) | 24.6 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 2.1 TB of uncompressed HTML down to 77.6 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: ~720.6 GB of raw gzipped WARCs becomes **24.6 GB of Parquet** — a **96.6% total reduction** — containing 17,205,106 clean markdown documents.
|
| 179 |
|
| 180 |
### Processing Times
|
| 181 |
|
| 182 |
+
Pipeline timings across 889 shards of CC-MAIN-2026-08:
|
| 183 |
|
| 184 |
```
|
| 185 |
+
Download (raw WARC) ████████████████████████ total 54h 29m 27s avg 3m 40s
|
| 186 |
Convert (HTML → MD) █░░░░░░░░░░░░░░░░░░░░░░░ total 4h 30m 54s avg 18s
|
| 187 |
+
Export (Parquet) █████░░░░░░░░░░░░░░░░░░░ total 12h 23m 18s avg 50s
|
| 188 |
Publish (HuggingFace) ██░░░░░░░░░░░░░░░░░░░░░░ total 6h 26m 33s avg 26s
|
| 189 |
```
|
| 190 |
|
data/CC-MAIN-2026-08/01268.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c75a0bd2069f5f0552e8d1b466d9b9d876cf67871c0ce94b210ed6ced65f0a80
|
| 3 |
+
size 30296209
|
stats.csv
CHANGED
|
@@ -621,6 +621,7 @@ CC-MAIN-2026-08,1264,19357,2587024454,96413559,29815462,2026-03-13T19:36:26Z,318
|
|
| 621 |
CC-MAIN-2026-08,1265,19480,2554786069,97256048,29914477,2026-03-13T19:47:56Z,454,0,239,23
|
| 622 |
CC-MAIN-2026-08,1266,19338,2557272375,95022217,29673971,2026-03-13T19:55:56Z,323,0,155,13
|
| 623 |
CC-MAIN-2026-08,1267,19508,2601615925,94415426,29354363,2026-03-13T20:04:27Z,348,0,139,18
|
|
|
|
| 624 |
CC-MAIN-2026-08,1301,19417,2601765043,93758124,29790521,2026-03-13T11:40:27Z,0,0,0,0
|
| 625 |
CC-MAIN-2026-08,1302,19471,2592470194,95488689,29747494,2026-03-13T11:40:27Z,0,0,0,0
|
| 626 |
CC-MAIN-2026-08,1303,19681,2609544680,95027524,30250671,2026-03-13T11:40:27Z,0,0,0,0
|
|
|
|
| 621 |
CC-MAIN-2026-08,1265,19480,2554786069,97256048,29914477,2026-03-13T19:47:56Z,454,0,239,23
|
| 622 |
CC-MAIN-2026-08,1266,19338,2557272375,95022217,29673971,2026-03-13T19:55:56Z,323,0,155,13
|
| 623 |
CC-MAIN-2026-08,1267,19508,2601615925,94415426,29354363,2026-03-13T20:04:27Z,348,0,139,18
|
| 624 |
+
CC-MAIN-2026-08,1268,19593,2603183686,94744617,30296209,2026-03-13T20:13:26Z,487,0,71,0
|
| 625 |
CC-MAIN-2026-08,1301,19417,2601765043,93758124,29790521,2026-03-13T11:40:27Z,0,0,0,0
|
| 626 |
CC-MAIN-2026-08,1302,19471,2592470194,95488689,29747494,2026-03-13T11:40:27Z,0,0,0,0
|
| 627 |
CC-MAIN-2026-08,1303,19681,2609544680,95027524,30250671,2026-03-13T11:40:27Z,0,0,0,0
|