A MongoDB API compatible , high-performance, elastic, distributed document database.

EloqDoc is a high-performance, elastic, distributed transactional document database with MongoDB API compability. Built on top of Data Substrate, it leverages a decoupled storage and compute architecture to deliver fast scaling, ACID transaction support, and efficient resource utilization.

EloqDoc eliminates the need for sharding components like mongos in MongoDB, offering a simpler, more powerful distributed database experience. It’s ideal for workloads requiring rapid scaling, high write throughput, and flexible resource management.

EloqDoc is a fork of MongoDB 4.0.3 that replaces the WiredTiger storage engine with the Eloq storage engine. It is distributed under the GNU Affero General Public License (AGPL).

Explore EloqDoc website for more details.

👉 Use Cases: web applications, ducument store, content management systems — anywhere you need MongoDB API compatibility but demand distributed performance and elasticity.

Seamlessly integrates with MongoDB clients, drivers, and tools, enabling you to use existing MongoDB workflows with a distributed backend.

Supports multiple writers and fast distributed transactions, ensuring high concurrency and fault tolerance across a cluster without sharding complexity.

• Scales compute and memory 100x faster than traditional databases by avoiding data movement on disk.
• Scales storage independently, conserving CPU resources for compute-intensive tasks.
• Scales redo logs independently to optimize write throughput.

Delivers ACID transaction support with especially fast distributed transactions, making it suitable for mission-critical applications.

Operates as a distributed database without requiring a sharding coordinator (e.g., mongos), reducing operational complexity and overhead.

• Fast Scaling: Compute and memory scale independently without disk data movement, enabling rapid elasticity for dynamic workloads.
• Storage Flexibility: Storage scales separately from compute, optimizing resource allocation and reducing waste.
• Write Optimization: Independent redo log scaling boosts write throughput, ideal for high-velocity data ingestion.
• No Sharding Overhead: Distributes data natively across the cluster, eliminating the need for additional sharding components.

Download the EloqDoc tarball from the EloqData website.

Follow the instruction guide to set up and run EloqDoc on your local machine.

Follow these steps to build and run EloqDoc from source.

It is recommended to use our Docker image with pre-installed dependencies for a quick build and run of EloqDoc.

docker pull eloqdata/eloq-dev-ci-ubuntu2404:latest
docker run -it --name eloqdoc eloqdata/eloq-dev-ci-ubuntu2404
git clone https://github.com/eloqdata/eloqdoc.git
cd eloqdoc

Alternatively, you can also pull the source code in an existing Linux environment (currently, ubuntu2404 is preferred), and manually run the script to install dependencies on your local machine. Notice that this might take a while.

git clone https://github.com/eloqdata/eloqdoc.git
cd eloqdoc
bash scripts/install_dependency_ubuntu2404.sh

Fetch the code and initialize submodules:

git clone https://github.com/eloqdata/eloqdoc.git
git submodule update --init --recursive

First, specify an install path.

export INSTALL_PREFIX=/your/install/path/absolute
# Set data store type (required), ELOQDSS_ROCKSDB/ELOQDSS_ROCKSDB_CLOUD_S3
export WITH_DATA_STORE=ELOQDSS_ROCKSDB

Then, configure and build the Eloq engine and dependencies via CMake with Open Log Service enabled.

cmake -G "Unix Makefiles" \
      -S src/mongo/db/modules/eloq \
      -B src/mongo/db/modules/eloq/build \
      -DCMAKE_INSTALL_PREFIX=$INSTALL_PREFIX \
      -DWITH_DATA_STORE=$WITH_DATA_STORE
cmake --build src/mongo/db/modules/eloq/build -j8
cmake --install src/mongo/db/modules/eloq/build

Finally, build MongoDB via SCons with Open Log Service enabled.

pyenv global 2.7.18


python2 buildscripts/scons.py \
    MONGO_VERSION=4.0.3 \
    VARIANT_DIR=RelWithDebInfo \
    LIBPATH=/usr/local/lib \
    CXXFLAGS="-Wno-nonnull -Wno-class-memaccess -Wno-interference-size -Wno-redundant-move" \
    --build-dir=#build \
    --prefix=$INSTALL_PREFIX \
    --disable-warnings-as-errors \
    -j8 \
    install-core

For ELOQDSS_ROCKSDB_CLOUD_S3 data store type, EloqDoc uses an S3-compatible storage backend. For testing, deploy an S3 emulator.

wget https://dl.min.io/server/minio/release/linux-amd64/minio
chmod +x minio
./minio server ./data

Create a configuration file mongod.conf according to config/example.conf. Modify /home/eloq with your home path. The configuration file specifies $HOME/eloqdoc-cloud as deploy directory.

mkdir ~/eloqdoc-cloud && cd ~/eloqdoc-cloud
mkdir etc db logs
mv ~/mongod.conf etc/

export LD_PRELOAD=/usr/local/lib/libmimalloc.so:/usr/lib/libbrpc.so
export PATH=$INSTALL_PREFIX/bin:$PATH
mongod --config etc/mongod.conf

mongo --eval "db.t1.save({k: 1}); db.t1.find();"

Star This Repo ⭐ to Support Our Journey — Every Star Helps Us Reach More Developers!