Sentient Trading Simulator

AWS ec2 Preparation

To the target AWS ec2 apply the following changes:

Add the following export to ~/.profile:

export MONGODB=mongodb://<username>:<password>@<database_host>:<port>/admin?readPreference=primary
After saving the file, ensure that the above exports are applied:

$ source ~/.profile

Software Installation

It is assumed that Python +3.6.5 has already been installed on the target ec2. What follows are instructions to install the simulation software.

Install the software:

$ git clone https://github.com/1057405bcltd/simulator.git
Create a Python virtual environment in the same folder:

$ python3.7 -m venv simulator
Activate the virtual environment:

source ./simulator/bin/activate
$ cd simulator
Prepare the simulator environment:

$ source ./Scripts/project
Install the required Python packages:

$ pip install -r Python/requirements.txt

Configuration

Using whatever tool is available, create a document in the configurations collection in the sim database (documents must be uniquely named) that describes the details of the planned simulation.

ToDo: Describe the structure of configurations documents.

Running

Simulator execution is initiated as follows:

$ sim-load <config_name> $ sim <config_name>

where <config_name> refers to the name of a configurations document.

Output of a simulation run appears in the following collections:

simulations: Top-level description of the simulation. Investigation of the consequences of a simulation run normally will begin with this collection.
partitions: A partition is effectively a sub-simulation working within a time subframe of a wider simulation.
trades: Detailed simulation results recorded by the ids of the simulation and partition in which the simulation trade was created.

Cleaning

The following command clears out the results of all simulations (it does NOT, however, delete documents from the configurations collection).