Platform 1.0.0 Release


#1

We’re delighted to be able to announce the available of the 1.0.0 version of the OpenGamma platform! I’d suggest starting by reading my blog post about it here. You can get the source and binary packages from the 1.0.0 download section of the developer site, and to help you get started, we’ve added a new Quick Start guide on the documentation website.

Note: the Windows installer packages mentioned for R and Examples will be released in the next few hours.
Update: Windows installers now available with and without built-in R support. We wanted to add support for the latest version of R that came out over the weekend (2.15.0) before pushing them up to the download site.

Some of the highlights are:

First release of R integration module

  • Allows execution of completely custom scenarios controlled from R
  • Market data can be manipulated at level of individual 'ticker' or in tensor form (matrix/vector etc)
  • Custom securities and portfolios can be created from R or existing portfolios referenced

Bloomberg module now open source

  • Complete example system using data populated from a Bloomberg Terminal or Managed B-PIPE/SAPI instance
  • Automatic reference data loading for exchange traded securities
  • Time series loading and updating
  • OpenGamma Live Data adapter for real-time streaming data

HTML5 Web UI

  • Market data snapshots supported as a data source
  • Time series viewing through the UI (see PnL series in dep graph for Historical VaR for example)
  • Position detailed information pop-up, with link to full security info in analytics view
  • Multiple live data sources including combined data sources
  • Fall-back to time series for missing data
  • Single level re-aggregation of portfolio on-the-fly
  • First uses of new 'push REST' support in notifying web client of updates asynchronously
  • Quantity in separate column, other small formatting changes
  • Portfolio breadcrumbs
  • Configuration form improvements
  • Integration of Security Time Series data into Security views

Packaging

  • Windows installer (.msi) for Examples
  • Automatic Bloomberg terminal detection with installation of appropriate example set
  • OpenGamma server installed as a Windows service
  • Windows installer (.msi) for R support
  • Windows installer (.msi) for Examples & R together
  • All installers are signed with a digital certificate

Database Masters

  • New asset classes in security master
  • Re-written batch database
  • Arbitrary attributes on portfolios, positions, trade and securities allow for custom meta-data
  • Hidden portfolios
  • Performance enhancements (esp. time series)
  • Support for PostgreSQL 10

Aggregation

  • Support for aggregating by currencies (currency + normalized currency pairs)
  • Support for ordering/sorting of positions
  • Multi-aggregator aggregates several ways into single portfolio
  • Option to aggregate where possible when some values are missing

Complete overhaul of the configuration system

  • Much easier deployment and maintenance using distributed component management system

Examples

  • With access to Bloomberg terminal
    • Arbitrary exchange traded positions can be loaded from a CSV
    • Several example portfolios containing OTC instruments are generated during installation
  • Without access to a Bloomberg terminal
    • We still support a synthetic data simulator (live and historical)
    • We show a wider range of asset classes (e.g. CMS/FX/Swaptions etc).

Data Import/Export

  • Standard import format for security/portfolio data
  • Framework for custom importers

Improved Build System

  • Single-step build
  • Easier deployment
  • Better ant target names

Analytics

  • Externally provided sensitivities
  • Yield curve sensitivity mapping
  • Separate Yield Curve/Credit/All sensitivities buckets
  • DV01, CS01, Historical VaR

Documentation

  • Improved documentation, especially for Analytics

#2

A big thanks to the entire opengamma team for their combined effort on making this awesome product. I tried the example and bloomberg example and they look just fine. I especially loved the part where bloomberg integration was quite seamless.

I have a couple of questions on version 1.0.0

  • Which branch should i download to work with 1.0.0 , When i tried master , there were couple of compilation issues. - https://gist.github.com/2297860
  • I could find the batch tab removed from the UI. Can i know what is planned here.

Please find time to answer them.


#3

The branch you should use for now is either dev/stable or dev/v1.0.x, which will be tracking each other for the time being. We’ve actually decided to stop committing directly into master and start working from an different branch, and only updating master when it’s stable. This means users like yourself can follow very near the bleeding edge (we’ll be merging changes every 2-3 weeks), but avoid occasional compilation errors.

With respect to the missing batch tab, in the form it was, it wasn’t actually a useful view so I decided to remove it. The release includes a new streaming batch mode for the engine, and a new database schema, all of which can be accessed programmatically. We intend to incorporate the ability to view previously run batches and a console to run new batch jobs into the new analyics UI we’re working on. I’ll keep you updated on progress we make there. We’ll also add some scripts soon so batches can be run more easily from the command line, in which case you should be able to view results using standard OLAP cube protection tools e.g, Mondrian.

Hope that helps.

Jim


#4

Oh, and glad to hear you liked the Bloomberg integration! Something else that might be of interest is the new upload button on the portfolios tab that lets you upload a really simple CSV file containing Bloomber tickers, CUSIPs or ISINs, quantities and optionally, trade date, price (premium) and counterparty. There’s a link to an example template CSV file on the upload form you can download, edit in Excel and re-upload. It should load all the required security definitions, build your portfolio and positions and load underlying price time series. If you’re wondering what to enter in the ‘Data Provider’ box, try CMPL (which is Bloomberg’s London Composite provider). You’ll still need to create a view definition, but that’s relatively easy once you have a portfolio in the system.


#5