Time Series Analytics with Spark on Hadoop

Time series analysis is used in a variety of applications, including industrial equipment maintenance, user behavior analysis and high frequency stock trading. With the advent of the Internet of Things, time series data has become much more commonplace. Devices and sensors of all kinds are continuously transmitting data points, often with an associated time component, allowing end users to glean new trends, forecast seasonality, or calculate aggregated statistical averages. Application developers and architects can leverage the Time Series Analytics solution to quickly develop and deploy real-time alerting and dashboarding applications on time-series data. With this solution, you can:

  • Develop alerting/monitoring systems that capture newly ingested time-sensitive data
  • Build real-time dashboards, allowing you to leverage new aggregated insights in near real time
  • Deploy on the industry’s best Hadoop solution provided by the MapR Converged Data Platform, which features the full Apache Spark stack

What's included


Software
1-year subscription of
MapR software

Professional Services
Custom engagement
based on your use case

Certification
MapR Hadoop
certification credits

The solution template includes workflows, parsers, and deployment models that help you build time-series applications that incorporate data generated from the Internet of Things. Installation and configuration of the MapR cluster is included within the scope of this Quick Start Solution.

Key solution capabilities

  • Low-Latency NoSQL Database: Store and process large volumes of time-series data from disparate data sources in MapR-DB—a top-ranked NoSQL datastore. OpenTSDB deployment over MapR-DB is also an option, depending on the use case.
  • Rapid Application Development with Apache Spark: Integrate Hadoop and Apache Spark capabilities to perform in-memory transformations and aggregations over large-scale, real-time data.
  • Real-time Dashboarding: Deploy real-time, web-based dashboards based on aggregated data. You also have the option to deploy search capabilities depending on the use case.

Time-Series Analytics Template

Key MapR differentiators

  • Leading Key-Value Store: MapR-DB has been recognized as the top-ranked Key-Value data store that provides high throughput and low latencies along with enterprise-grade resiliency—so you can process high ingest rates reliably.
  • Scalability: MapR is the only Hadoop distribution that scales all the way to a trillion files and tables without compromising performance.
  • Full stack support for Apache Spark: MapR is the only distribution that supports the full Apache Spark stack, giving you maximum flexibility to build applications rapidly and deploy complex ETL pipelines and transformations on your data.
  • Search and discovery: Indexing and search capabilities that integrate with MapR help aggregate various customer touchpoints and provide a user interface that delivers rich insights.

Talk to a Time Series Analytics Expert
Engage one of our subject matter experts to see how you can get started

Contact us

 

Learn More
EBOOK

Time Series Databases: New Ways to Store and Access Data
Download

BLOG

The Ultimate 3-Minute Guide to Time Series Data and OpenTSDB
Read