The power of machine learning, data science and big data is no longer considered a hype or fad. It’s a reality and it’s impacting the bottom lines of businesses. Data has emerged as a real competitive edge and disruptive force for enterprises. Companies that make the most of data by using machine learning and data science will win and outlast in this digital world.
IBM and Hortonworks expand partnership to help your business
IBM recently extended its partnership with Hortonworks to better help businesses accelerate data-driven decision making. Hortonworks is the leading industry and only pure open source Hadoop platform. The IBM and Hortonworks partnership brings the innovation of the open source community to enterprise customers without the worry of vendor lock-in, but while also enabling world-class enterprise support and access to the most efficient, high performance compute and storage platforms for big data and cognitive analytics.
The IBM Hortonworks partnership aims to help business in four ways:
IBM Big SQL is a hybrid SQL engine for Hadoop that enables you to easily query big data across the enterprise.Big SQL allows organizations to access data across Hadoop and relational databases, whether they’re on the cloud, on premises, or both, using a single database connection. With its federation capabilities using IBM Fluid Query, Big SQL can virtualize data from many different data stores including Hive, HBase, Spark, Db2, Oracle, SQL Server, Netezza, Informix, Teradata, WebHDFS and object-based storage.
Further, Big SQL federation capabilities help data professionals complete data discovery and analysis across repositories more quickly. It enables you to bring use cases from analysis of time-series data to geospatial data to your business professionals quickly.
Key business use cases supported with IBM Big SQL include
- Self-service analytics
- Trusted data lake
- Data warehouse optimization, modernization and offload
- SQL business Apps connectivity
2. Accelerate innovation with data science
IBM Data Science Experience (DSX), together with Hortonworks Data Platform (HDP), accelerates adoption of data science, machine learning and artificial intelligence for businesses and enterprises. It enables data scientists and data engineers to learn, collaborate and productionize the machine learning models at scale. Data professionals can choose from their preferred language — from R, Python to Scala on Spark and Hadoop clusters — to do predictive analytics on unstructured and structured data.
Key business use cases supported with IBM Data Science Experience include
- Real-time personalized recommendation engine
- Behavioral customer segmentation
- Next best action solutions
- Churn analytics
- Cyber security
- Fraud detection and anomaly detection
3. Enhance unified governance
Unified governance is an important element providing self-service analytics access to your business in a governed way facilitating compliance standards and helping detect data breaches. The IBM Unified Governance stack, which includes IBM BigIntegrate, IBM BigQuality and IBM Information Governance Catalog, natively runs on Hadoop to govern and manage unstructured and structured data in seamless ways.
Key business use cases supported include
4. Choose infrastructure that’s purpose-built to deliver more data, faster and more efficiently
Built for data-intensive workloads, IBM Power Systems deliver up to 70 percent more HDP queries per hour, with a 40 percent on-average reduction in query response times, than equivalent x86-based solutions1. That means that data-driven business decisions can be made faster, using far fewer system resources. In addition, Hortonworks support for Power Systems means data scientists can easily tap into GPU-accelerated deep learning training at scale with the industry-leading IBM PowerAI solution.
Platform optimization is further enabled by IBM Spectrum Scale, the first software-defined storage solution to be certified with Hortonworks Data Platform (HDP). This provides a high-performance parallel file system for managing data at scale, with the ability to perform data archive and analytics in place. In addition HDP is now certified with IBM Elastic Storage Server, a pre-integrated hardware and software solution that enables massive infrastructure reduction versus typical scale-out x86-based solutions.
All four of the above capabilities are important pillars to fully realize the vision of data-driven organizations and to propel your journey towards becoming a successful cognitive business powered by artificial intelligence and machine learning.
Bring your data to warp speed with IBM at DataWorks Summit in Sydney
Faster time to value for machine learning and data science use cases are key to today’s digital business success. Ranging from a personalized recommendation engine to improved sales to next best action use cases that improve efficiencies of customer service, companies are deploying data science and machine learning to detect anomalies at scale and to efficiently manage fraud detection and cyber security. Using your data to its full potential is foundational to becoming a cognitive and digital business.
Learn more about the innovations IBM and Hortonworks are enabling at DataWorks Summit 2017, to be held Sept 19 to Sept 21 in Sydney, Australia.
1Performance results based on IBM testing of 10 queries (simple, medium, complex) with varying runtimes running against a 10TB DB. Tests run on 10 x IBM Power Systems S822LC for Big Data (20C/40T), 2 X POWER8 2.92GHz, 256 GB memory, RHEL 7.2, HDP 2.5.3, compared to published x86/Hortonworks results running on 10 x AWS d2.8xlarge EC2 nodes running HDP 2.5; details at https://hortonworks.com/blog/apache-hive-going-memory-computing/. Individual results may vary. Data as of February 28, 2017.)