Hosting Trading Algorithms on Google Compute Engine
Summary
The document outlines using Google Compute Engine virtual machines to run trading software, with R and Zorro as examples, and connecting the system to a broker through Interactive Brokers Gateway. It frames cloud hosting as a way to avoid maintaining local infrastructure, adjust computing resources as needed, and connect the trading process with cloud storage, data, and analytics services.
The main limitation it identifies is latency: a cloud instance may be unsuitable for latency-sensitive strategies, even though the author considers this less likely to be decisive for many independent traders. The setup is organized as a sequence covering account creation, selecting Compute Engine, creating a virtual machine, configuring access, connecting and testing, and stopping the instance when finished. A Windows permissions issue is also noted. The text is only a high-level walkthrough: its pricing section contains no figures, and the actual configuration, security practices, broker connectivity details, and operational safeguards are not developed.
Key ideas
- A cloud virtual machine can host trading software and connect to a broker gateway.
- Cloud hosting reduces infrastructure maintenance and allows computing resources to be adjusted.
- Latency requirements may make cloud hosting unsuitable for some trading algorithms.
- The setup covers creating an instance, setting access, connecting, testing, and stopping it when idle.
- The walkthrough gives limited detail on pricing, security, and production operations.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.