Publish shard CC-MAIN-2026-08/01056
Browse files- README.md +11 -11
- data/CC-MAIN-2026-08/01056.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 **22,
|
| 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) | 2.7 TB | ~295 TB | — |
|
| 173 |
-
| Packed markdown WARC (.md.warc.gz) | ~47.
|
| 174 |
-
| Final Parquet (Zstd level 19) | 32.
|
| 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 2.7 TB of uncompressed HTML down to 101.
|
| 177 |
|
| 178 |
-
End to end: ~
|
| 179 |
|
| 180 |
### Processing Times
|
| 181 |
|
| 182 |
-
Pipeline timings across
|
| 183 |
|
| 184 |
```
|
| 185 |
-
Download (raw WARC) ████████████████████████ total 88h
|
| 186 |
Convert (HTML → MD) █░░░░░░░░░░░░░░░░░░░░░░░ total 5h 55m 4s avg 18s
|
| 187 |
-
Export (Parquet) █████░░░░░░░░░░░░░░░░░░░ total 21h
|
| 188 |
Publish (HuggingFace) ██░░░░░░░░░░░░░░░░░░░░░░ total 7h 31m 58s avg 23s
|
| 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 **22,524,060 documents across 1166 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 1166 files of CC-MAIN-2026-08 (22,524,060 pages total), projected to the full crawl of 100,000 WARC files.
|
| 168 |
|
| 169 |
+
| Stage | 1166 files (measured) | 100,000 files (projected) | Reduction |
|
| 170 |
|---|---|---|---|
|
| 171 |
+
| Raw WARC (.warc.gz, downloaded) | ~945.1 GB | ~83 TB | — |
|
| 172 |
| HTML extracted (uncompressed) | 2.7 TB | ~295 TB | — |
|
| 173 |
+
| Packed markdown WARC (.md.warc.gz) | ~47.8 GB | ~3.7 TB | **-98.3%** vs HTML |
|
| 174 |
+
| Final Parquet (Zstd level 19) | 32.3 GB | ~2.9 TB | **-32.5%** 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 2.7 TB of uncompressed HTML down to 101.6 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 68.3%.
|
| 177 |
|
| 178 |
+
End to end: ~945.1 GB of raw gzipped WARCs becomes **32.3 GB of Parquet** — a **96.6% total reduction** — containing 22,524,060 clean markdown documents.
|
| 179 |
|
| 180 |
### Processing Times
|
| 181 |
|
| 182 |
+
Pipeline timings across 1166 shards of CC-MAIN-2026-08:
|
| 183 |
|
| 184 |
```
|
| 185 |
+
Download (raw WARC) ████████████████████████ total 88h 21m 17s avg 4m 32s
|
| 186 |
Convert (HTML → MD) █░░░░░░░░░░░░░░░░░░░░░░░ total 5h 55m 4s avg 18s
|
| 187 |
+
Export (Parquet) █████░░░░░░░░░░░░░░░░░░░ total 21h 20m 3s avg 1m 5s
|
| 188 |
Publish (HuggingFace) ██░░░░░░░░░░░░░░░░░░░░░░ total 7h 31m 58s avg 23s
|
| 189 |
```
|
| 190 |
|
data/CC-MAIN-2026-08/01056.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dc0d98806f7d0c8cdeb71b88cba7c767159e9178babe1fae9748a37601d9c6f5
|
| 3 |
+
size 29492788
|
stats.csv
CHANGED
|
@@ -693,6 +693,7 @@ CC-MAIN-2026-08,1052,0,0,0,343,2026-03-14T01:11:50Z,451,0,0,5
|
|
| 693 |
CC-MAIN-2026-08,1053,19667,2614924242,94869149,29598323,2026-03-14T01:21:35Z,551,0,53,12
|
| 694 |
CC-MAIN-2026-08,1054,19494,2602930494,88502975,28683848,2026-03-14T01:29:20Z,357,0,62,15
|
| 695 |
CC-MAIN-2026-08,1055,19473,2595775562,93271749,29969403,2026-03-14T01:41:56Z,687,0,51,16
|
|
|
|
| 696 |
CC-MAIN-2026-08,1101,19721,2641197765,95719568,30240239,2026-03-13T09:59:33Z,564,0,296,0
|
| 697 |
CC-MAIN-2026-08,1102,19562,2608129691,99280759,31748974,2026-03-13T10:28:07Z,1518,0,191,0
|
| 698 |
CC-MAIN-2026-08,1103,19270,2517580674,89253461,28361470,2026-03-13T10:55:39Z,1380,0,268,0
|
|
|
|
| 693 |
CC-MAIN-2026-08,1053,19667,2614924242,94869149,29598323,2026-03-14T01:21:35Z,551,0,53,12
|
| 694 |
CC-MAIN-2026-08,1054,19494,2602930494,88502975,28683848,2026-03-14T01:29:20Z,357,0,62,15
|
| 695 |
CC-MAIN-2026-08,1055,19473,2595775562,93271749,29969403,2026-03-14T01:41:56Z,687,0,51,16
|
| 696 |
+
CC-MAIN-2026-08,1056,19160,2545973799,93816297,29492788,2026-03-14T01:51:50Z,513,0,86,0
|
| 697 |
CC-MAIN-2026-08,1101,19721,2641197765,95719568,30240239,2026-03-13T09:59:33Z,564,0,296,0
|
| 698 |
CC-MAIN-2026-08,1102,19562,2608129691,99280759,31748974,2026-03-13T10:28:07Z,1518,0,191,0
|
| 699 |
CC-MAIN-2026-08,1103,19270,2517580674,89253461,28361470,2026-03-13T10:55:39Z,1380,0,268,0
|