Need help with cloudburst?
Click the “chat” button below for chat support from the developer who created it, or find similar developers for support.

About the developer

146 Stars 33 Forks Apache License 2.0 37 Commits 5 Opened issues


A stateful serverless platform

Services available


Need anything else?

Contributors list


Build Status codecov License

Cloudburst is a low-latency, stateful serverless programming framework built on top of the Anna KVS. Cloudburst enables users to execute compositions of functions at low latency, and the system builds on top of Anna in order to enable stateful computation. Cloudburst is co-deployed with the Anna caching system to achieve low-latency access to shared state, and the system relies on Anna's lattice data structures to resolve conflicting updates to shared state.

Getting Started

You can install Cloudburst's dependencies with

and use the bash scripts included in this repository to run the system locally. You can find the Cloudburst client in
. Full documentation on starting a cluster in local mode can be found here; documentation for the Cloudburst client can be found here. An example interaction is modeled below.
$ pip3 install -r requirements.txt
$ ./scripts/ n n
$ ./scripts/ n


is the main client interface; when running in local mode, all interaction between the client and server happens on
. Users can register functions and execute them. The executions return
s, which can be retrieved asynchronously via the
method. Users can also register DAGs (directed, acylic graphs) of functions, where results from one function will be passed to downstream functions.
>>> from cloudburst.client.client import CloudburstConnection
>>> local_cloud = CloudburstConnection('', '', local=True)
>>> cloud_sq = local_cloud.register(lambda _, x: x * x, 'square')
>>> cloud_sq(2).get()
>>> local_cloud.register_dag('dag', ['square'], [])
>>> local_cloud.call_dag('dag', { 'square': [2] }).get()

To run Anna and Cloudburst in cluster mode, you will need to use the cluster management setup, which can be found in the hydro-project/cluster repo. Instructions on how to use the cluster management tools can be found in that repo.


The Hydro Project is licensed under the Apache v2 License.

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.