Loading...

Energy trading: order book

Case Study

Context

As the share of renewables grows in the energy mix, electrical supply and demand unpredictability puts massive stress on trading firms. In particular, building the order book becomes much more computationally intensive.

The order book is the list of buys and sells organized by price levels. It provides valuable trading information, which helps traders and improves market transparency.

In a nutshell, what Quasar delivered

  • Deployment of complete market and weather data capture, from start to finish, in less than a week
  • Instant order book rebuilding at any point in the day while using less cloud resources
  • Storage footprint divided by 10
  • Future proof solution thanks to built-in scalability

Challenge

Because of renewables, electricity production can suddenly drop or increase, creating sudden demand, which, in turn, puts stress on order book construction.

There can be an increase of at least one order of magnitude in the volume of orders, making the order book two orders of magnitude more computationally expensive.

Energy trading firms typically built an architecture relying on a relational database. That was sufficient to handle previous volumes. This approach does not work anymore regardless of the increase in computational power dedicated to the relational database. Building the order book is more and more expensive as the number of orders grows.

The solution

Quasar is connected directly to the market data provider and captures the data without a third-party tool. Capture is done with virtually no latency.

Quasar uses its unique suite of compression algorithms to compress the data faster than it arrives and use its micro-indexes to reduce future lookup and feature extraction time. This results in significantly reduced disk usage without impacting the queryability of the data.

Each instrument is stored as a timeseries a two dedicated tables (one for orders, one for trades). Weather data is organized by source.

Tables can be tagged at will, enabling a flexing querying mechanism based on those tags. Tags can be queried recursively. For example, obtaining all the instruments of an index or all the weather data of a specific region.

Queries use a SQL interface, dedicated API, or one of the numerous connectors available. For example, traders can load data directly into Tableau using the Quasar ODBC driver. Quants can load timeseries directly into Python Pandas using the Quasar Pandas API.

Quasar has a built-in order book computation module that enables sub-second order book rebuilding at any time of the day, even for deep order books. These order books are then accessible from a web browser using a dedicated dashboard built using the Quasar framework.

Benefits

  • Firms can keep up with the market and have linear scalability capabilities thanks to the clustering capabilities of Quasar, increasing their agility and competitiveness
  • Compatibility with all standard analytical tool suites for maximum flexibility
  • Faster time to a solution: Quasar abstracts away all data engineering complexity
  • Outstanding TCO thanks to the combination of data compression and rapid order book rebuilding, saving on disk storage, CPU power, and development time

Related Tags:
Top