Prerequisites

• PostgreSQL
• psycopg2

Clone the Repository

Clone this repository to your local machine.

Install PostgreSQL 16 or higher and Development Tools

Ensure PostgreSQL and necessary development tools are installed

sudo apt-get update
sudo apt-get install postgresql postgresql-contrib postgresql-server-dev-all build-essential

sudo apt install libpq-dev

Install and Setup pgvector

The pgvector repository is already included in this project.

Initial compilation and installation:

chmod +x compile_pgvector.sh
./compile_pgvector.sh

This will compile pgvector in debug mode (-g -O0) which is useful for development and debugging.

After modifying pgvector source code: Simply run the compile script again:

./compile_pgvector.sh

Then restart PostgreSQL to load the updated extension:

sudo service postgresql restart

Setup Database

Start PostgreSQL service:

sudo service postgresql start

Create database user and database with pgvector extension:

chmod +x setup_db.sh
./setup_db.sh

This script will:

• Create user xx (with no password, or configure your own in the script)
• Create database rbacdatabase_treebase
• Install pgvector extension

Note: You may need to modify setup_db.sh to match your desired database name, username, and password. Make sure to update config.json accordingly.

Setup Python Environment

Create a virtual environment and install dependencies:

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

Setup embedding model

python -m spacy download en_core_web_md

Download Dataset

Download the dataset to {project directory}/dataset/:

• Cohere Wikipedia 22-12 Dataset

mkdir dataset
cd dataset
sudo apt-get install git-lfs
git lfs install
git clone https://huggingface.co/datasets/Cohere/wikipedia-22-12

• SIFT10M features (Fu et al.)
  • Download SIFT10M.tar.gz and place it in the directory pointed to by dataset_path (defaults to /data).
  • The loader extracts SIFT10M/SIFT10Mfeatures.mat automatically on first run, or you can run
    tar -xf SIFT10M.tar.gz SIFT10M/SIFT10Mfeatures.mat.

Configure database config file

Edit config.json in project root directory to match your database setup:

{
    "dbname": "rbacdatabase_treebase",
    "user": "x",
    "password": "123",
    "host": "localhost",
    "port": "5432",
    "dataset_path": "/data",
    "use_gpu_groundtruth": false
}

Configuration Options:

• use_gpu_groundtruth:
  • false (recommended): Use PostgreSQL for ground truth computation. Slower but no setup required.
  • true: Use FAISS GPU for ground truth computation. First run is slow (builds indexes), subsequent runs are much faster.

Note: If you used setup_db.sh, the default configuration is username xx with no password and database rbacdatabase_treebase. Adjust these values as needed.

Optional: Install FAISS for GPU-accelerated Ground Truth

For faster ground truth computation (especially for repeated testing), install FAISS:

# Create a conda environment with FAISS GPU support
conda create -n faiss_env python=3.11
conda activate faiss_env
conda install -c pytorch faiss-gpu

# Install other dependencies
pip install -r requirements.txt

Then set "use_gpu_groundtruth": true in config.json.

Performance comparison:

• PostgreSQL mode: Consistent speed, no index overhead
• FAISS mode: First run builds role-level indexes (slow), subsequent runs use cached indexes (very fast)

Prepare Data

cd basic_benchmark
# Load the default Wikipedia dataset (1M rows by default)
python3 common_prepare_pipeline.py --dataset wikipedia-22-12

# Example: load the SIFT 1M benchmark vectors (load-number 0 loads the entire file)
python3 common_prepare_pipeline.py --dataset sift-128-euclidean --load-number 0

# Example: load the 10M SIFT feature matrix (auto-extracts SIFT10Mfeatures.mat if needed)
python3 common_prepare_pipeline.py --dataset sift10m --load-number 0

# Flags:
#   --dataset       One of {wikipedia-22-12, arxiv, sift-128-euclidean, sift10m}
#   --load-number   Number of rows to ingest (0 or negative means “all remaining rows”)
#   --start-row     Offset within the dataset before loading
#   --num-threads   Worker processes used for ingestion (defaults to CPU count)

go to controller directory
# prepare main tables(user/role/permission) index
python3 initialize_main_tables.py

Generate Permission

go to services/rbac_generator diretory

# Taking treebased as an example
python3 store_tree_based_rbac_generate_data.py

Initilize partition and prepare for queries

go to basic_benchmark
# initialize role partition
python3 initialize_role_partition_tables.py --index_type hnsw

# (optional) initialize user partition
python3 initialize_combination_role_partition_tables.py --index_type hnsw

# generate queries
python3 generate_queries.py --num_queries 1000 --topk 100 --num_threads 4

# compute ground truth (pointer benchmark & tests share cache)
python3 compute_ground_truth.py

Ground Truth Caching:

• Ground truth results are automatically cached in basic_benchmark/ground_truth_cache.json
• Subsequent test runs with the same queries will load from cache (instant)
• Cache is automatically cleared when regenerating queries with generate_queries.py
• To manually clear cache: rm basic_benchmark/ground_truth_cache.json

### Initilize dynamic partition
```sh
# initilize dynamic
go to controller/dynamic_partition/hnsw
# if needed, delete parameter_hnsw.json from hnsw directory to regenerate parameters
python3 AnonySys_dynamic_partition.py --storage 2.0 --recall 0.95

Run(HNSW index)
```sh
go to basic_benchmark directory

# example:
python test_all.py --algorithm RLS --efs 500
python test_all.py --algorithm ROLE --efs 20
python test_all.py --algorithm USER --efs 20
python test_all.py --algorithm AnonySys --efs 40

Run(ACORN index)

go to acorn_benchmark directory

#modify efs value from main.cpp
build C++ project and run

Before building, create acorn_benchmark/config.json to point benchmarks at the shared index location:

{
    "index_storage_path": "/pgsql_data/acorn/"
}

Make sure /pgsql_data/acorn/ exists ahead of time; both ACORN and dynamic-partition indexes are persisted there.