Skip to content

Portfolio example

Portfolio Evaluation using QuantLib

This example demonstrates the use of the QuantLib with its SWIG Python bindings.

Components: - /examples/workloads/python/quant_lib/portfolio_pricing_client.py client application that generates a portfolio of trades using simple portfolio generator. Trades then being split into individual tasks and sent to the HTC-Grid for computation. Once all tasks are completed, the client application merges results together to determine total value of the portfolio.

  • /examples/workloads/python/quant_lib/portfolio_pricing_engine.py compute engine that receives a list of trades to evaluate (could be entire portfolio or just a single trade). The engine uses QuantLib to evaluate the value of the portfolio.

Deployment

  1. Follow all the steps in the main Readme file until you reach section "Build HTC artifacts". In this section you will need to modify the make command replacing the happy-path with the python-quant-lib-path as follows: 
    make python-quant-lib-path TAG=$TAG REGION=$HTCGRID_REGION
    This will apply the following changes:
  2. prepare python runtime environment for the lambda functions

  3. generate 2 sample yaml files that will be used to deploy testing client containers.

  4. After make is completed, please run 

    make apply-python-runtime  TAG=$TAG REGION=$HTCGRID_REGION

Follow all the remaining steps as is in the main readme file.

Running the example

Two default configurations are provided. The first configuration submits a portfolio containing a single trade.

kubectl apply -f ./generated/portfolio-pricing-single-trade.yaml

The second configuration submits a portfolio containing multiple trades.

kubectl apply -f ./generated/portfolio-pricing-book.yaml

Refer to the corresponding yaml files to change the configuration of the client application and refer to the help of the client application to identify all options.

Uninstalling-Installing and destroying HTC grid

The destruction time is about 15 min.

  1. To remove the grid resources run the following command: 

    make destroy-python-runtime TAG=$TAG REGION=$HTCGRID_REGION

  2. To remove the images from the ECR repository execute 

    make destroy-images TAG=$TAG REGION=$HTCGRID_REGION

  3. Finally, this will leave the 3 only resources that you can clean manually, the S3 buckets. You can remove the folders using the following command 
    make delete-grid-state TAG=$TAG REGION=$HTCGRID_REGION