Unravel Data, a company that has raised a total of $23M in three rounds of funding delivering an Application Performance Management (APM) platform designed for Big Data applications, has introduced APM for streaming applications.
This new feature would enable enterprises to improve performance and reliability of their Internet of Things (IoT), real-time, and other streaming applications. Fortune 500 companies would rely heavily on real-time applications to deliver up-to-the-second analytics and the best user experience. Unravel’s latest innovation would help these mission-critical applications maintain peak performance.
“Unravel’s latest release makes huge strides towards giving enterprises the best performance, predictability, and reliability for all their streaming applications,” said Dr. Shivnath Babu, Chief Technology Officer, Unravel Data. “Unravel leverages recent advances in machine learning and AI to automate root-cause analysis and resolution by applying these techniques to the full-stack monitoring data available for streaming applications. This monitoring data includes metrics and logs from applications, from systems like Kafka, Spark Streaming, and HBase, as well as from on-premises and cloud infrastructure.”
Stream processing powers some of the most critical undertakings today such as algorithmic trading, autonomous cars, and health monitoring. These streaming applications process real-time data feeds from IoT devices, financial transactions, social media comments, or from database update events. Systems like Kafka, Spark Streaming, and HBase have emerged as critical components of the Big Data stack to support these applications. These systems would provide a unified and high-performance architecture for processing real-time data feeds.
Bottlenecks Streaming Applications
While this architecture has made it quite easy to create streaming applications, guaranteeing the performance and reliability of these applications would be extremely challenging. If a streaming application starts to lag behind in processing data in real-time, then diagnosing the root cause would take considerable time and effort with current tools. The root cause could be attributed to a number of intertwined factors, making it hard and time consuming to pinpoint the exact cause. For example, the root cause may be an application problem (e.g., poor data partitioning in Spark Streaming) or a system problem (e.g., suboptimal configuration of Kafka), or an infrastructure problem (e.g., resource contention in the cloud).
Unravel can automatically pinpoint the cause of bottlenecks, slowdowns, and failures in streaming applications. Furthermore, Unravel would provide automatic fixes for these issues, thereby reducing the amount of time and resources needed to firefight such ongoing problems.
Unravel Data will unveil the streaming applications features at Strata Data Conference San Jose on March 6th.
