andmer/og-platform

The core OpenGamma Platform wrapper project

Welcome to the OpenGamma Platform!

OpenGamma's flagship technology for financial institutions, the OpenGamma Platform, is a comprehensive solution for financial analytics capable of satisfying the full range of front-office and risk requirements. It supports pre-trade ad-hoc calculations, near-real-time streaming analytics, batch/overnight risk calculations, and complex historical scenarios and stress tests in the same system.

Built as a completely open architecture, the OpenGamma Platform is designed so that every component can be individually used, or individually replaced, based on customer requirements. We do not believe in forklift upgrades, and we built the OpenGamma Platform so that they are never necessary: individual projects can use OpenGamma components when they provide a clear advantage, and later migrate additional portions of their infrastructure if and when time and resources permit.

Visit the developer website at http://developers.opengamma.com for more information, downloads, docs and more.

Building OpenGamma

Firstly you need to make sure Apache Ant and Git are installed and working. Our build uses Apache Ivy and our favoured IDE is Eclipse.

Maven files do exist, but they should not be relied on for building OpenGamma or running the tests.

Initializing the build

The OpenGamma Platform is open source software using the Apache License v2. The [company] (http://www.opengamma.com/) behind OpenGamma also offers support and some additional commercial components (The commercial components typically have dependencies with restrictive licensing incompatible with open source.)

The source code can be cloned using git from GitHub or downloaded as a source tarball from http://developers.opengamma.com.

git clone https://github.com/OpenGamma/OG-Platform.git

If you do not have access to the OpenGamma commercial components, then skip the rest of this section.

To download the commercial components, initialize the OpenGamma system with your username and password:

ant init

Then, download the additional commercial components:

ant clone-or-pull

This will do the initial clone of each project into the projects/ directory. When you run this subsequently, it does a git pull on each project, although you will probably need to edit the build.xml file to tell the git task which branch you want of each project.

Now, continue on to the next section.

Completing the build

The source code must be compiled before use. This will build multiple jar files and install them into your local Ivy repository, which is located in ~/.ivy2:

ant build

To run the unit tests (once the build is complete), use:

ant tests

The output is generated into build/test-reports/html.

Running the OpenGamma engine

The primary program in the OpenGamma platform is known as the "engine". For production, the engine is typically customized, however two example engine configurations are supplied, one with Bloomberg support and one using simulated market data.

To build and install the example engines run:

ant install

which will create a directory called opengamma/ in the project root. To start the OpenGamma server with one of the example engine configurations, follow the instructions in opengamma/README.txt.

Eclipse

Importing the projects into Eclipse requires following a very specific set of instructions to work properly. See the file ECLIPSE.txt for details.

Ivy Repositories and settings files

If you wish to set up a shared ivy repository, this can be specified as the IVY_SHARED_DEFAULT_ROOT environment variable. See the ivy settings files in common/. By default the shared repository is configured to be the repository/ directory in the root, which is where the dependencies go when you download a source tarball that includes them. If they are not present, Ivy will continue up the resolver chain until it finds the artifacts from the OpenGamma public Ivy repository, or fails if you do not have connectivity.

Each project's build.xml includes the common.xml file in common/ which defines most targets. Any tasks that need further customization can be customized by simply overriding that task (for an example, see the javadocs task in OG-Analytics/build.xml.

common.xml also includes common/build.properties and tests.properties files. All the database settings are in the .properties files under config/ in the appropriate project (e.g. OG-Examples or OG-BloombergExample)

More information

For more information go to http://developers.opengamma.com