Mtell Operationalizes Apache Spark for the Industrial Internet of Things (IIoT)

Share Article

Extending the Mtell Machine Learning Platform for Extreme Performance, Simplifying Applications for IIoT, and Opening to Third Party Extensions

News Image
Customers can readily insert custom code, and our Spark integration provides a cost-effective method to scale to exceptionally large multi-site deployments.

SAN DIEGO (January 21, 2016) - Mtell announced today that the new release of its market leading predictive and prescriptive analytics platform, Mtell PreviseTM, has been revised to execute on and incorporate all elements of the open source Apache SparkTM project.

Apache Spark is an open source general framework used for large-scale data processing. It scales rapidly and easily among computing clusters, and performs in-memory computing to dramatically reduce execution times for real-time processing. Furthermore, it also permits computation of extremely large data sets. Spark is ideal for machine learning, where its intrinsic ability to cache datasets in memory greatly speeds up iterative processing of algorithms. In addition to performance improvements, Spark brings a reliable library of algorithms and consistent access to both streaming and batch data in one unified platform.

Mtell’s Industrialized Machine LearningTM is the cornerstone of IIoT prescriptive maintenance solutions that are improving contemporary methods of inspections and service. This simplification and approachability of big data, predictive analysis, and prescriptive methods, ensures that end users can solve maintenance problems that were unsolvable five years ago. Mtell Previse uses streaming sensor data that is autonomously processed by machine learning algorithms to focus on the origins of degradation. The early warnings allow problems to be fixed before they impact operations; thereby reducing maintenance efforts, extending the life of capital assets, and improving equipment uptime. Mtell maximizes operational effectiveness for customers in diverse markets that include transportation, oil and gas, mining, water/wastewater, and other asset-intensive industries.

Beginning today, data scientists and analysts who are familiar with R, Python, or Scala can enhance overall monitoring by deploying custom logic, calculations, and algorithms inside the Mtell Previse platform. The combination of Spark and Previse both leverage the agility of the Spark ecosystem for exploratory analysis, including Jupyter notebook capabilities, and the power of Previse as an end-to-end monitoring platform. The world’s largest industrial companies use the solution to monitor fleets of offshore drilling rigs, mining facilities, heavy haul trucks, locomotive engines, pumps, compressors, and static equipment.

“Engineers and data scientists at large enterprises want to perform exploratory analysis for advanced scenarios. Our approach enables them to concentrate on value creation, where the solution platform manages all the mechanics of data handling and live monitoring,” said Alex Bates, CTO of Mtell. “Customers can readily insert custom code, and our Spark integration provides a cost-effective method to scale to exceptionally large multi-site deployments.”

About Mtell

Founded in 2006, Mtell is a privately held company providing software solutions for managing the health of industrial equipment. Making machines smart, Mtell plays an important role in developing the Internet of Things. In addition to reducing risk to people safety, and the environment, Mtell is a significant contributor to equipment performance and profitability. Solutions are deployed globally in the transportation, oil and gas, mining, pharmaceutical, and wastewater industries. For more information, visit

Share article on social media or email:

View article via:

Pdf Print

Contact Author

Jillian Schwantz
+1 (619) 295-0022
Email >
since: 06/2013
Follow >
since: 06/2013
Like >

Follow us on
Visit website