Software

Threat Detection and Analysis Solution Uses Data Science, Big Data and Self-Learning

Our client develops, manufactures, and sells networking hardware and other products related to the communication and information technology industry worldwide. Additionally, the company offers security products including cloud, email, endpoint, web, and network security; advanced malware protection; and next generation intrusion prevention systems.

The client recently purchased a security startup that pinpoints attacks before they can disclose and corrupt sensitive data. The solution analyzes web traffic, analyzes endpoint and network data, and then uses machine learning to identify malicious activity.

Business Challenge

Our client's initial goal was to scan and detect threats in the ecosystem using simple network snapshots and data science. However, the client did not have the expertise in-house and was looking for a team of professionals to help address the goal and take on the challenge of dealing with the technology’s beta versions and its relatively new functionality.

Project Description

The client consulted with SoftServe to assist with its security tool, threat detection, and analysis solution that included hardcore data science, big data, and self-learning. Its main features tracked and detected abnormalities based on a snapshot of network traffic.

The SoftServe team focused on configuration, building the infrastructure, and monitoring. Additionally, the team led the migration from the on-premise model to the AWS cloud and container world. And the service of migration from the Monoliths architecture to Microservices.

SoftServe team—following Scrum—consisted of three DevOps engineers, one team lead (TL), and one project manager (PM). The team communicated regularly with daily sync-ups, weekly planning meetings, weekly demos, and HipChat.

The requirements were described by the technical project owner (PO) and the scope was divided into phases. The SoftSerServe team worked closely with the PO to translate the requirements into technical tasks. The migration phases consisted of:

  • Planning
  • Proof of concepts (PoCs)
  • Production in a cloud
  • Traffic switch

Technology Stack

Boto3 IntelliJ IDEA Kubernetes Heapster PostgreSQL
Python Git Docker Prometheus AWS
Bash TeamCity Ingress Controller Jira Fluentd
Terraform BitBucket Splunk ElasticSearch Metrics api
Ruby ELK Cloudwatch Confluence Nginx

Value Delivered

The SoftServe team provided the client with the needed expertise and optimized the processes to meet the goals of the collaboration. Additionally, the trust and collaboration with the client’s developer teams resulted in achieving the desired results in a short time period.

Download PDF