Mojaloop Performance Testing Workstream

This repository contains the complete infrastructure, tooling, and test definitions used to perform end-to-end performance testing of Mojaloop at scale (hundreds to thousands of TPS).

The workspace is structured to clearly separate infrastructure provisioning, test execution, and test results, making it easy to reproduce, tune, and analyze performance runs.

---

Repository Structure

ML-PERF-WHITEPAPER-WS
├── infrastructure/
│   └── README.md
│       - Provisioning of cloud infrastructure
│       - Kubernetes clusters (MicroK8s)
│       - Mojaloop backend, switch, DFSPs
│       - k6 operator setup
│
├── performance-tests/
│   ├── src/
│   │   └── README.md
│   │       - k6 test implementation
│   │       - Helm chart for k6 Operator
│   │       - Test configuration and execution scripts
│   │
│   └── results/
│       - Test results (summaries, metrics, logs)
│       - Scenario-specific configuration overrides
│       - TPS-specific tuning references (e.g. 500 / 1000 / 2000 TPS)
│
├── docs/
│   - Supporting documentation and notes
│
├── README.md
└── LICENSE.md

---

High-level Workflow

1. Provision Infrastructure
   Use Terraform and Ansible to provision compute, networking, and Kubernetes clusters.
   See: infrastructure/README.md

2. Deploy Mojaloop Stack
   Deploy Mojaloop backend services, switch, DFSP simulators, monitoring, and security components.

3. Prepare Test Data
   Pre-register MSISDNs and verify Kafka, MySQL, and DFSP readiness.

4. Run Performance Tests
   Execute k6 tests via the k6 Operator from a dedicated k6 cluster.
   See: performance-tests/README.md

5. Analyze Results
   Review metrics, summaries, and scenario configurations under performance-tests/results.

---

Design Principles

• Reproducibility – Fully declarative infra and test configuration
• Scalability – Tested with multiple DFSPs and high TPS targets
• Isolation – Dedicated clusters for switch, DFSPs, and load generation
• Observability – Prometheus, Grafana, and Kafka UI support

---

License

This project is licensed under the terms of the LICENSE.md.