Platform manages oil and gas real-time and historical sensor data
April 16, 2015
California companies Mtelligence and MapR Technologies have jointly announced a big data platform called Mtell Reservoir that combines the MapR Distribution including Hadoop, Mtell Previse Software and Open TSDB (time-series database) software technology. The system ingests and analyses real-time sensor and historical data alongside maintenance data that are generated from industrial equipment for oil rigs, chemical plants, mining, water and wastewater plants.
Designed for data centre user needs, Mtell Reservoir is an enterprise historian that distributes disk access and CPU processing across MapR clusters of computers to provide orders of magnitude improvements over contemporary plant historians. It has proven loading of over 100 million data points per second on four servers, with performance that scales linearly with the number of servers.
“With a predictable, low-cost scaling methodology, MapR offers a top performing Hadoop distribution that can deliver the performance levels required when it comes to handling massive amounts of sensor and maintenance data,” said Mike Brooks, president and COO of Mtell. “We discovered that other Hadoop distributions require far more hardware to accomplish what Mtell has deployed.”
The platform enables subject matter experts within organisations to perform remote monitoring and analysis from a central repository. There they can act on volumes of data retrieved from many assets at many locations to enable new levels of predictive maintenance. With Mtell Previse, the system proactively learns patterns of normal and errant behaviour across fleets of equipment to provide warnings of minor degradation. Early problem mitigation can prevent equipment failure thus increasing net product output at any plant. The platform also enables entirely new insights into machine and process operations efficiency, quality and utilisation.
“Mtell holds a unique position in the oil and gas space as one of the only companies with an advanced machine learning platform for predictive maintenance,” said Ted Dunning, chief application architect at MapR Technologies. “Their expertise in the oil and gas space has been invaluable and played a key role in the success of applying the MapR Distribution in this demanding environment for reliably ingesting and analysing data.”
With contemporary approaches, it takes a long time to load large datasets, forcing unacceptable delays before analysis can start. The combined MapR and Mtell platform reduces loading time, while also enabling the ingestion and analysis of high-speed, real-time sensor data streams. This big data method for manufacturing industries leverages Open TSDB and MapR-DB (the MapR in-Hadoop NoSQL database), which allows more data to be acquired and accessed with Hadoop and uses less hardware for lower total cost of ownership. MapR-DB further decreases costs by simplifying the administration of large databases.