Source code for our paper: GPU-Accelerated Batch-Dynamic Subgraph Matching

• Ubuntu 20.04
• CUDA 11.1
• NVIDIA GTX 3090

1. Build the project with make.
2. Put the dataset in the datasets folder, which should include the data_graph and query_graph folders.
3. Run the project with ./gamma -d ../datasets/data_graph/<data_graph> -i ../datasets/data_graph/<edge_stream> -q ../datasets/query_graph/<query_graph>.

All the datasets used can be download from here

v id label
...
e src dst
...

where lines start with v denotes vertices and those start with e denotes edges, id is the vertex id, label is the vertex label, src and dst are the source and destination vertex id of an edge, respectively.

e src dst
...