Big data, which in most contexts refers to large data sets that are analyzed algorithmically to reveal patterns and relationships, is marching steadily toward ubiquity. According to a report published last year by Dresner Advisory Services, big data adoption in enterprise steeply climbed from 17% in 2015 to 59% in 2018. Concurrent with this trend, global market revenue projections have inched upward, with firms like Wikibon anticipating an increase from $42 billion last year to $103 billion in 2027.

Just because big data apps and systems are popular doesn’t mean they’re easy to manage, though. That’s where Unravel comes in. The Palo Alto, California-based company — the brainchild of CEO Kunal Agarwal, a Sun Microsystems veteran,¬† and Duke University computer science professor Shivnath Babu — offers a full-stack data operations platform that addresses everything from data ingestion and migration to processing and transforming. It’s an eminently successful one — annual recurring revenue grew 500% year-over-year in 2018 — and it’s poised to expand substantially in the coming year.

Unravel today announced that it’s raised $35 million in an oversubscribed series C financing led by Point72 Ventures, with participation from Harmony Partners, Menlo Ventures, GGV Capital, and M12 (Microsoft Ventures). The cash infusion comes after a $15 million series B round in January 2018 and a $7 million series B round in September 2016, and it brings Unravel’s total raised to $37.2 million.

“Every business is becoming a data business, and companies are relying on their data applications such as machine learning, [internet of things], and customer analytics for better business outcomes using technologies such as Spark, Kafka, and NoSQL,” said Agarwal. “We are making sure that these technologies are easy to operate and are high performing so that businesses can depend on them. We partner with our customers through their data journey and help them successfully run data apps on various systems whether on-premises or in the cloud.”

Unravel

Above: A graphic illustrating how Unravel monitors and optimizes systems and apps.

Image Credit: Unravel

Unravel’s endgame, Agarwal explained, is to reduce the complexity of delivering stable and reliable apps for engineers, developers, and network architects alike. Toward that end, the company’s eponymous platform tracks and triages performance issues across customers’ systems and the apps running on these systems, namely by applying optimization libraries and fixes to failures, bottlenecks, and resource-wasting apps.

Its agentless, low-overhead microsensor design — which enables Unravel to be deployed on-premises, in the cloud, or in hybrid cloud environments such as Cloudera, Google Cloud, Oracle DBA, Quoble, Amazon Web Services, Hortonworks, Microsoft’s Azure, Cassandra, Databricks, and Mapr — offers per-cluster and per-node visibility into code, configurations, containers, resource constraints, and schedulers. It maps dependencies among apps, services, storage (both hot and cold), and users in a single dashboard with metrics and visualizations, and it leverages machine learning to provide plain-language configuration recommendations that take into account future app needs.

For instance, in a Spark framework, Unravel keeps tabs on status, duration, data I/O, stages, partitioning, garbage collection, and more, along with lowdowns, failures, killed jobs, and resource consumption. Unravel automatically resolves errors and auto-tunes Spark apps, and it serves up “context-sensitive” suggestions covering topics like RDD caching, CPU resource contention, and container resource utilization. Plus, it shows representations of SQL query plans and insights into how and when they are executed, and streamlines tasks like spawning processes, running queries, moving services to other queues, updating external databases, and killing apps that threaten the performance of other apps.

Perhaps best of all, Unravel boasts a robust API that integrates with workflow engines, messaging systems, and other DevOps solutions like Spark, Kafka, Hadoop, Tez, Slack, Pagerdubut, Apache HBase, and more to deliver proactive alerts about unexpected performance degradations. And for customers transitioning to the cloud, Unravel has an extensive set of app migration tools that identify the best app candidates, reveal the seasonality and the ideal time of day to take advantage of the best services prices, and measure baseline performance pre- and post-move.

“CIOs in our network told us story after story of traditional application monitoring tools failing in a big data context because those tools were designed for the world of the past. And we didn’t just hear this problem from third parties, we were seeing it at Point72 as well,” said Point72 chief market intelligence officer Matthew Granade. “This new architecture requires a different product, one built from the ground up to focus on the unique challenges posed by big data applications. Unravel is poised to capture this emerging big-data APM market.”

Unravel’s current customers include Kaiser Permanente, Autodesk, YP.com, Adobe, Deutsche Bank, Wayfair, and Neustar, plus software startups, Fortune 100 financial services, airlines, supermarket retailers, multinational telecom groups and telecom providers, and global banks. Data Elite Ventures and AppDynamics founder Jyoti Bansal are among the company’s previous investors.

Sign up for Funding Daily: Get the latest news in your inbox every weekday.