OCaml Distributed Data Processing
A distributed data processing system developed in OCaml
Todo
How to express arbitrary DAG? How to express loop? Apply function.
Interface to Irmin or HDFS to provide persistent storage
Test delay-bounded and error-bounded barrier
Split Context module into server and client two modules
Implement parameter.mli
Implement barrier control in parameter modules
Rename ... DataContext and ModelContext?
Implement Coordinate Descent in model parallel ...
Enhance Mapreduce engine, incorporate with owl.
Add techreport based on the barrier control.
How to compile & run it?
To compile and build the system, you do not have to install all the software yourself. You can simply pull a ready-made container to set up development environment.
docker pull ryanrhymes/actorThen you can start the container by
docker run -t -i ryanrhymes/actor:latest /bin/bashAfter the container starts, go to the home director, clone the git repository.
git clone https://github.com/ryanrhymes/actor.gitThen you can compile and build the system.
make oasis && make
Formed in 2009, the Archive Team (not to be confused with the archive.org Archive-It Team) is a rogue archivist collective dedicated to saving copies of rapidly dying or deleted websites for the sake of history and digital heritage. The group is 100% composed of volunteers and interested parties, and has expanded into a large amount of related projects for saving online and digital history.
