ParallelM, one of the fastest growing new companies in machine learning operationalization, today announced the launch of ParallelM MCenter™, the first software solution for operationalizing machine learning (ML) and deep learning across the enterprise, in Europe.
Operationalizing machine learning poses a significant challenge, as current operations solutions have limited ability to tackle the unique intricacies of ML behavior patterns when running in production. Prevailing workarounds tend to be manual and brittle, inhibiting ML service from scaling, and delaying the desired benefits of ML to the business.
That’s why, in order to be successful in deploying AI and ML across the enterprise, organizations are starting to adopt MLOps (a compound of “machine learning” and “operationalization”) techniques. MLOps is a unique combination of technologies that take ML code and safely deploy and manage it in production and practices for collaboration and communication between an organization’s academic (e.g., data scientists, data engineers, and developers) and economic (e.g., IT, operations, governance teams, and business analysts) professionals as machine learning algorithms are operationalized and put into production.
MCenter delivers a unique approach for MLOps, addressing ML production issues head-on by automating ML-optimized continuous deployment and integration, ensuring the quality of live ML applications, and empowering data science and operations teams with innovative visualization and collaboration facilities to manage the ML applications over time. Using MCenter, business teams can mitigate risk, ensure compliance, assess, and optimize the ROI of their AI initiatives. By providing a single, unified software solution for the full ML production lifecycle, MCenter enables enterprises to move confidently into the critical phase of realizing and scaling ML business value.
“We have seen immense interest from companies worldwide in using AI and ML in recent years. However, the actual number of companies that have been able to successfully deploy AI and ML and see its true benefits is quite small,” said Sivan Metzger, CEO of ParallelM. “In order to put ML into production, companies must be able to bring new technologies and processes together. Our solution helps companies accomplish this by automating the core elements and functions required for scaling out the live delivery of ML applications.”
MCenter is powered by ParallelM’s MetaMap™ technology, a proprietary, patent-pending approach that provides a comprehensive, logical representation of any ML-driven business application. The MetaMap manages the relationship between the ML models, pipelines, events, predictions, policies and dataflows, while abstracting away the underlying composite of model dependencies and compute infrastructure. This enables operations teams to organize their efforts around managing ML business services rather than individual components within the ML stack.
MCenter can be deployed on cloud, on premise, or in hybrid scenarios and works across compute platforms such as Apache Spark™, TensorFlow™, Apache Flink®, and PyTorch. MCenter also integrates with leading data science and AI developer platforms.