Sparkle [spär′kəl]: a library for writing resilient analytics applications in Haskell that scale to thousands of nodes, using Spark and the rest of the Apache ecosystem under the hood.
This is an early tech preview, not production ready
The tl;dr using the hello app as an example on your local machine:
$ stack build hello
$ mvn -f sparkle -Dsparkle.app=sparkle-example-hello package
$ spark-submit --master 'local[1]' sparkle/target/sparkle-0.1.jar
Requirements:
- the Stack build tool;
- either, the Nix package manager,
- or, OpenJDK, Maven and Spark >= 1.6 installed from your distro.
To run a Spark application the process is as follows:
- create an application in the
apps/folder, in-repo or as a submodule; - add your app to
stack.yaml; - build the app;
- package your app into a deployable JAR container;
- submit it to a local or cluster deployment of Spark.
To build:
$ stack [--nix] build
You can optionally pass --nix to all Stack commands to ask Nix to
provision a local Spark and Maven in a local sandbox for good build
results reproducibility. Otherwise you'll need these installed through
your OS distribution's package manager for the next steps.
To package your app:
$ mvn -f sparkle -Dsparkle.app=<app-executable-name> package
or with
$ stack --nix exec -- mvn -f sparkle -Dsparkle.app=<app-executable-name> package
And finally, to run your application, say locally:
$ spark-submit --master 'local[1]' target/sparkle-0.1.jar
See here for other options, including lauching a whole cluster from scratch on EC2.
Copyright (c) 2015-2016 Tweag I/O Limited.
All rights reserved.
Sparkle is free software, and may be redistributed under the terms specified in the LICENSE file.
Sparkle is maintained by Tweag I/O.
Have questions? Need help? Tweet at @tweagio.