Powered by Elsen

The Elsen nPlatform in action

Elsen helps world-class financial organizations build solutions that help them or their clients overcome their most difficult data problems.



35x time savings over a standard industry workflow

Elsen partnered with Thomson Reuters to offer its hedge fund and investment management clients the functionality they needed to effortlessly harness and use tremendous amounts of time-series data to test investment strategies.

The application developed – QA Point Powered by Elsen – is intuitive and easy-to-use, leading to an average outcome of 35X time savings on a typical user workflow, based on an industry norm.


Making massive bond calculations more than 250x faster

To test a client’s high-performance computing requirements for a generic bond portfolio, Elsen developed an application on the Elsen nPlatform that calculated the portfolio value for 100,000 bonds with equal portfolio weights. Each bond had its own Floating Spread, Coupon Payment Frequency and Time to Maturity.

The application also calculated 1 million reference and discount rate curves, where the curves were defined as Base rate + an Adjustment, where the Adjustments were provided by client (in production, these are generated through a Monte Carlo process).


over 250x faster

Than CPU Only

Average time to process 1 curve: 14 ms

  • Average time for I/O for 1 curve: 1 ms
  • Performed using single thread of i5 with a Nvidia Tesla K40c (GPU) and SSD

Total time to process 1M curves: 3.6 hours

  • Estimated 250x faster than single-threaded CPU-only performance
  • Performed using single thread of i5 with a Nvidia Tesla K40c (GPU) and SSD


Elsen achieved 30x speed-up in VWAP calculations

Rather than running volume-weighted average price (VWAP) computations over nights and weekends to populate the in-memory database for quick retrieval throughout the week, an investment firm wanted to be able to compute VWAP calculations at any time to eliminate the need to store pre-computed values.

Elsen developed an application on the Elsen nPlatform that retrieved aggregated data for a given security between two set timestamps. For example, the application should be able to compute the total volume and VWAP price for ticker IBM between 9:30:45.1000 AM and 11:45:15.3000 AM.

The client’s request: “If we make 50,000 calls for such data, we want to implement a process where each core on a GPU would compute this linear algebra in a parallel fashion.”


The client’s metric for success was at least 30x speed up for 50,000 simultaneous VWAP calculations with random double timestamps across 500 securities. As the number of calculations increases, the GPU’s performance becomes accretively more efficient when compared to the CPU.