Overview

The Graphalytics datasets are compressed using zstd. The total size of the compressed archives is approx. 350 GB. When decompressed, the datasets require approximately 1.5 TB of disk space.

For detailed information on the datasets, see the table with their statistics.

Download scripts

March 2023. Some Graphalytics datasets were had incorrect data or expected results, and were fixed. If you downloaded the datasets prior to March 2023, some datasets had missing/incorrect reference outputs for certain algorithms. Therefore, we recommend to download the datasets again.

Feb 2026. Vertex and edge files are now available in Parquet format.

dataset nodes edges scale size package nodes edges
cit-Patents 3M 16M XS 119.1 MB tar.zst v.parq e.parq
com-friendster 65M 1B XL 6.7 GB tar.zst v.parq e.parq
datagen-7_5-fb 633k 34M S 162.3 MB tar.zst v.parq e.parq
datagen-7_6-fb 754k 42M S 200.0 MB tar.zst v.parq e.parq
datagen-7_7-zf 13M 32M S 434.5 MB tar.zst v.parq e.parq
datagen-7_8-zf 16M 41M S 544.3 MB tar.zst v.parq e.parq
datagen-7_9-fb 1M 85M S 401.2 MB tar.zst v.parq e.parq
datagen-8_0-fb 1M 107M M 502.5 MB tar.zst v.parq e.parq
datagen-8_1-fb 2M 134M M 625.4 MB tar.zst v.parq e.parq
datagen-8_2-zf 43M 106M M 1.4 GB tar.zst v.parq e.parq
datagen-8_3-zf 53M 130M M 1.7 GB tar.zst v.parq e.parq
datagen-8_4-fb 3M 269M M 1.2 GB tar.zst v.parq e.parq
datagen-8_5-fb 4M 332M L 1.5 GB tar.zst v.parq e.parq
datagen-8_6-fb 5M 421M L 1.9 GB tar.zst v.parq e.parq
datagen-8_7-zf 145M 340M L 4.6 GB tar.zst v.parq e.parq
datagen-8_8-zf 168M 413M L 5.3 GB tar.zst v.parq e.parq
datagen-8_9-fb 10M 848M L 3.7 GB tar.zst v.parq e.parq
datagen-9_0-fb 12M 1B XL 4.6 GB tar.zst v.parq e.parq
datagen-9_1-fb 16M 1B XL 5.8 GB tar.zst v.parq e.parq
datagen-9_2-zf 434M 1B XL 13.7 GB tar.zst v.parq e.parq
datagen-9_3-zf 555M 1B XL 17.4 GB tar.zst v.parq e.parq
datagen-9_4-fb 29M 2B XL 14.0 GB tar.zst v.parq e.parq
datagen-sf3k-fb 33M 2B XL 12.7 GB tar.zst v.parq e.parq
datagen-sf10k-fb 100M 9B 2XL 40.5 GB tar.zst v.parq e.parq
dota-league 61k 50M S 114.3 MB tar.zst v.parq e.parq
graph500-22 2M 64M S 202.4 MB tar.zst v.parq e.parq
graph500-23 4M 129M M 410.6 MB tar.zst v.parq e.parq
graph500-24 8M 260M M 847.7 MB tar.zst v.parq e.parq
graph500-25 17M 523M L 1.7 GB tar.zst v.parq e.parq
graph500-26 32M 1B XL 3.4 GB tar.zst v.parq e.parq
graph500-27 63M 2B XL 7.1 GB tar.zst v.parq e.parq
graph500-28 121M 4B 2XL 14.4 GB tar.zst v.parq e.parq
graph500-29 232M 8B 2XL 29.6 GB tar.zst v.parq e.parq
graph500-30 447M 17B 3XL 60.8 GB tar.zst v.parq e.parq
kgs 832k 17M XS 65.7 MB tar.zst v.parq e.parq
twitter_mpi 52M 1B XL 5.7 GB tar.zst v.parq e.parq
wiki-Talk 2M 5M 2XS 34.9 MB tar.zst v.parq e.parq
example-directed 10 17 - 1.0 KB tar.zst v.parq e.parq
example-undirected 9 12 - 1.0 KB tar.zst v.parq e.parq
test-bfs-directed <100 <100 - <2.0 KB tar.zst v.parq e.parq
test-bfs-undirected <100 <100 - <2.0 KB tar.zst v.parq e.parq
test-cdlp-directed <100 <100 - <2.0 KB tar.zst v.parq e.parq
test-cdlp-undirected <100 <100 - <2.0 KB tar.zst v.parq e.parq
test-pr-directed <100 <100 - <2.0 KB tar.zst v.parq e.parq
test-pr-undirected <100 <100 - <2.0 KB tar.zst v.parq e.parq
test-lcc-directed <100 <100 - <2.0 KB tar.zst v.parq e.parq
test-lcc-undirected <100 <100 - <2.0 KB tar.zst v.parq e.parq
test-wcc-directed <100 <100 - <2.0 KB tar.zst v.parq e.parq
test-wcc-undirected <100 <100 - <2.0 KB tar.zst v.parq e.parq
test-sssp-directed <100 <100 - <2.0 KB tar.zst v.parq e.parq
test-sssp-undirected <100 <100 - <2.0 KB tar.zst v.parq e.parq