TorQ-Crypto Released

Blog Data Analytics & Engineering 6 Jul 2020

Data Intellect

Today we are excited to release our new TorQ-Crypto package! For those of you that have used TorQ, you will most likely be familiar with the TorQ Finance Starter Pack. TorQ-Crypto behaves in a similar manner to the starter pack and provides an example of how an application can be built and deployed on top of the TorQ framework. The key difference being that TorQ-Cypto will collect and store real time data.

The package will build and maintain a cryptocurrency database by subscribing to 5 exchanges. These subscriptions are fully configurable and can be easily extended by the user to include additional data sets. The data collected contains level 2 order book information which can be accessed through the inbuilt API, we have also included some functions to help analyse both real-time and historical data. The chart below was generated from our OHLC function to look at BTCUSDT trends throughout 2020.

Data Capture

The system contains a number of feed handler processes which collect real time cyptocurrency data through RESTful APIs. Each feed process collects bid and ask data for its subscribed symbols at a frequency and limit which is fully configurable by the user. We have included 5 feed processes which retrieve data from the following exchanges:

  • OKEX
  • DigiFinex
  • Huobi
  • ZB
  • Bluehelix /HBTC

The diagram below summarizes this process:

Data Analysis

We have included functions to help analyse the data which are located on the RDB and HDB processes, we recommend executing these through the gateway to avail of query routing and load balancing. These functions are focused around comparing data across exchanges, providing information on the order book, OHLC prices, top of book data and arbitrage opportunities. 

The following dashboard was generated using the arbitrage function on the 04/05/2020 to compare the best bid and ask prices across all five exchanges throughout the day:

The graph below used the order book function to depict market depth for ETHUSDT as of 3pm on the 01/05/2020. Full TorQ-Crypto documentation along with examples of using these functions can be found here.

TorQ and the TorQ-Crypto pack can be downloaded from the AquaQ Github so go ahead and take a look!

If you want to know more or are having data capture architecture issues, please get in touch.

Share this: