dataClay ======== .. toctree:: :hidden: :caption: Getting Started user-guide main-concepts advanced-usage examples/index .. toctree:: :hidden: :caption: Deployment deployment/envvars deployment/management deployment/docker-deployment deployment/hpc-manual-deployment deployment/compile-redis .. toctree:: :hidden: :caption: API reference/index contrib/index .. toctree:: :hidden: :caption: Developers contributing dataClay is a distributed data store that enables applications to store and access objects in the same format as they have in memory and executes object methods within the data store. These two main features accelerate both the development of applications and their execution. Installation ------------ To install dataClay, you can use `pip `_: .. code-block:: console $ pip install dataclay Alternatively, you can obtain the latest source code from `GitHub `_: .. code-block:: console $ git clone https://github.com/bsc-dom/dataclay.git $ cd dataclay $ pip install . Usage ----- The :doc:`user-guide` is the primary resource for learning how to use the library and accomplish common tasks. For lower-level tweaking, refer to the :doc:`advanced-usage` guide. The :doc:`reference/index` documentation provides API-level documentation. Deployment ---------- The **DEPLOYMENT** section includes multiple scenarios. If you are not sure, you may want to start with the :doc:`deployment/docker-deployment`. License ------- dataClay is licensed under the BSD License. For more information, refer to the `LICENSE.txt `_. Contributing ------------ We welcome contributions to dataClay. Please see the :doc:`contributing` for more details.