April 3, 2018 | Written by: Preetam Kumar
Categorized: Data Analytics | Data Science
Share this post:
The IBM Analytics Engine team is excited to announce the General Availability (GA) of IBM Analytics Engine, the next generation of IBM’s Apache Spark and Apache Hadoop cloud service in the London DC.
Sign up for free trial today
Delivering an agile architecture for data science
IBM Analytics Engine is a cloud-based service that enables data scientists to rapidly provision, manage, run and retire Apache Hadoop and Apache Spark clusters. The solution is designed to solve the key pain points that organizations currently experience as they try to build up their big data analytics capabilities.
Spin up Spark and Hadoop clusters in minutes
The primary motivation of IBM Analytics Engine is to give your business a quick and simple way to deploy analytics applications.
As a data scientist or data engineer, you can log into the intuitive web interface from anywhere, and can spin up a Spark cluster or a Hadoop cluster with Hive, HBase, and other Hadoop ecosystem components within minutes.
Meanwhile, for application developers, IBM Analytics Engine also makes it easy to embed Hadoop or Spark capabilities into your code. The solution’s REST APIs and command line interface make it possible for your applications to provision and manage clusters programmatically. As a result, it’s easy to operationalize your apps with predictive modeling capabilities and integrate new levels of insight into your everyday operations.
Increasing flexibility by separating compute from storage
Unlike a traditional Hadoop architecture, IBM Analytics Engine keeps compute and storage infrastructure separate. The data is stored in IBM’s Cloud Object Storage Service , and the Hadoop and Spark clusters connect to the object storage repository when they need to access it.
This separation of concerns gives you much greater flexibility and reliability. Because the clusters themselves are not involved in data storage, you can spin up a cluster environment for the duration of a single job, and delete it on completion—with no risk of data loss. Moreover, you can write a configuration script once and pass it to IBM Analytics Engine as you create new clusters, to apply the exact same configuration as your previous clusters.
Only pay for what you use
Separating compute from storage also makes the solution more cost-effective than a permanent Hadoop cluster. With IBM Analytics Engine, you no longer have to keep a compute cluster running if you are not using it. Instead, you can deprovision it as soon as your job is finished, and spin up a new cluster the next time you need one.
As a result, IBM’s flexible cloud delivery model only requires you to pay for the compute resources that you are actually using—there is no longer any need for your organization to bear the costs of a fixed Hadoop infrastructure that may spend much of its time idle.
Take the next step
IBM Analytics Engine offers you the power to revolutionize your approach to big data analytics, and deliver a fast return on investment to your business. To find out how, please click here for more information.
Sign up for free trial today