ADGNN: Towards Scalable GNN Training with Aggregation-Difference Aware Sampling

The argument list:

--role_id=server,0
--worker_server_num=2,2
--ifctx=true
--data_path=/mnt/data/cora
--iter_lr_pttMethod_printInterval=200,0.01,hash,1,cuda,8
--hidden=16
--vtx_edge_feat_class_train_val_test=2708,5278,1433,7,140,500,1000
--iter_lr_pttMethod_printInterval=1000,0.2,hash,50
--sampleInfo=1:1,1000000,random
--adConfig=1:1,2,1000:1000,false,5,false,1
--servers=127.0.0.1:2001,127.0.0.1:2002,127.0.0.1:2003,127.0.0.1:2004,127.0.0.1:2005,127.0.0.1:2006
--workers=127.0.0.1:3001,127.0.0.1:3002,127.0.0.1:3003,127.0.0.1:3004,127.0.0.1:3005,127.0.0.1:3006

Note that, /mnt/data/cora is the common-shared nfs directory. Also, you can create the directory in the local to simulate the distributed environment.

##How to Install ADGNN

If you just want to using ADGNN to build your own GNN models, you need to install the python dependencies: python3.6 and requirements in "python/requirements.txt" on Ubuntu16.04 (other versions of Ubuntu can also work). Then use "ldd cmake/build/example2.cpython-36m-x86_64-linux-gnu.so" to detect if all dependencies are satisfied.

otherwise:

If you intend to modify the core codes of ADGNN in c++, beyond the python dependencies, you need to install cmake, grpc, protobuf, pybind11

mkdir cmake/build && cd cmake/build
cmake ../..
make

If build successfully, it will generate new 4 grpc and protobuf files (".grpc.pb.cc and .h","pb.cc and .h" ) and dynamic link libraries lib/pb11_ec.cpython-36m-x86_64-linux-gnu.so and lib/datatrans.cpython-36m-x86_64-linux-gnu.so. Then, you can run EC-Graph following the instructions in "How to Run an Example"

You can use docker to run ADGNN on the servers, please see details in Dockerfile.

##How to Run an Example

1: Install a distributed file system, e.g., NFS, HDFS. Set the shared-directory as "/mnt/data". If you just want to run ADGNN on a single-machine, you can just "mkdir /mnt/data" without installing NFS.

2: Processing the data format to the ADGNN format by using programs in "python/data_processing".

Two files will be created (all separators are "\t"). 

featsClass.txt (id   feat (dim = 5)   class):
0 1 0 1 1 1 0
1 0 1 1 0 1 1
2 0 0 0 1 1 0

edges.txt (src   dst)
0   1
1   2
0   2

3: Move these two files to their directory "/mnt/data/cora"

4: Set the number of workers and servers in "python/context/context.py"

5: Run 1 worker and 1 server as an example

Run "python/example/distgnn/dist_start.py" with "--role_id=server,0"
Run "python/example/distgnn/dist_start.py" with "--role_id=worker,0"

The other settings are shown at the beginning "argument list".