Stratum: A Serverless Framework for Lifecycle Management of Machine Learning based Data Analytics Tasks
This addresses lifecycle management challenges for ML practitioners dealing with deployment inefficiencies, though it appears incremental as a framework integration.
The authors tackled the complexity of managing machine learning workflows across heterogeneous environments by proposing Stratum, a serverless platform that deploys, schedules, and dynamically manages data analytics tools across cloud-fog-edge systems.
With the proliferation of machine learning (ML) libraries and frameworks, and the programming languages that they use, along with operations of data loading, transformation, preparation and mining, ML model development is becoming a daunting task. Furthermore, with a plethora of cloud-based ML model development platforms, heterogeneity in hardware, increased focus on exploiting edge computing resources for low-latency prediction serving and often a lack of a complete understanding of resources required to execute ML workflows efficiently, ML model deployment demands expertise for managing the lifecycle of ML workflows efficiently and with minimal cost. To address these challenges, we propose an end-to-end data analytics, a serverless platform called Stratum. Stratum can deploy, schedule and dynamically manage data ingestion tools, live streaming apps, batch analytics tools, ML-as-a-service (for inference jobs), and visualization tools across the cloud-fog-edge spectrum. This paper describes the Stratum architecture highlighting the problems it resolves.