ONEx's mission is to simplify the process of experimenting with distributed systems by proposing a set of standardized models and APIs for Fabric configuration, Data Flow and Chaos Engineering.
Imagine an environment to experiment easily with different Data Center fabric configurations without learning a NOS, and replicating a complex communication pattern (for example, a distributed machine learning training job) without having to become an expert at ML.
Imagine running 1000s of such experiments quickly with different fabric, workload and transport settings and different packet loss or buffer load impairments.
Automated, repeatable, experiments with distributed systems - made easy with ONEx.
Define the experiment once, run against multiple implementations
