The goal of this project was to create an enhanced messaging infrastructure for QuantaVerse’s various services for their financial crime analysis model.
Sponsored By: QuantaVerse
Team Members
Aman Gangwani | Trevor Guidry | Christopher Magtibay | Nischay Pagarani | Harshin Shah | Kaiyuan Wang | Rui Zhang | | | | |
Project Poster
Click on any image to enlarge.
Watch the Project Video
Download the Project Summary
Project Summary
Overview
QuantaVerse is home to an artificial intelligence crime risk analysis model, which consists of various services, such as news article aggregation, criminality sentiment analysis, and article language translation. These services can be rather resource-intensive, proving costly for both QuantaVerse and its clients.
Objectives
The goal of this project was to create an enhanced messaging infrastructure for QuantaVerse’s various services for their financial crime analysis model. This messaging infrastructure would facilitate communication between QuantaVerse’s services while also serving as a storage space for services to store the files that they process. The system to be created was to consist of a message broker service and a data storage system.
Approach
-QuantaVerse provided the team a black box approach to the problem as well as specific requirements for the system to be built.
-The team researched different technologies for both the message broker service and the data management system.
-An evaluation matrix was created to select a technology for each service, and Apache Kafka and MongoDB were selected.
-Meetings were held with QuantaVerse to help create a codebase for the mock services provided by QuantaVerse.
-A MongoDB database with several collections for each of QuantaVerse’s mock services was created.
-A server was created to host both the database and the messaging service so that they could be run across machines.
-Both the database and the messaging service were containerized for easy deployment in the future.
-Both services were stress tested to determine whether they could fulfill the performance requirements set forward by QuantaVerse.
-The team held weekly progress meetings with QuantaVerse.
Outcomes
-The team’s system is successfully containerized and works across multiple machines.
-The team’s system successfully interacts with QuantaVerse’s mock services.
-The system can handle over 1000 messages and 200 files per minute.