June 11, 2019

284 words 2 mins read



A data-driven compute platform

repo name hydro-project/fluent
repo link https://github.com/hydro-project/fluent
language C++
size (curr.) 657 kB
stars (curr.) 1249
created 2018-07-24
license Apache License 2.0

NOTE: The Fluent project has been renamed to Hydro, and the project has been broken up into multiple repositories. You can find all of them in the Hydro Project organization on Github. This repository has been archived and will be deleted eventually.


Build Status codecov License

Fluent is a fully managed, data-first computation framework under development that U.C. Berkeley RISE Lab. There are two main components in Fluent.

The first is a key-value store based on prior open-source work from the RISE lab (fka Anna). The Fluent KVS is an elastic, cloud-native storage engine that uses coordination-avoiding techniques and asynchronous message passing to provide very low latency. You can find more information about running and using the Fluent KVS in the kvs directory.

The second component is a data-centric programming framework, built on top of the Anna KVS. The goal of the programming framework is to provide users a general-purpose API and runtime for executing programs on data stored in the Anna KVS. Users are able to submit arbitrary code or containers for execution, and we plan to support performance SLOs for function execution.

To get started with the full project, please see the Getting Started page. You can find the index of all the docs here.


If you run into any issues, please open an issue and make sure to include relevant information (e.g., stack traces) as well as operating system, dependency versions, etc.

If you are looking to contribute to the project, please look at our issues list, particularly those marked as good for beginners and help wanted.

comments powered by Disqus