AI can only be effective if the data it is built upon, is trustworthy, accessible and compatible. The increased use of AI highlights weaknesses and limitations that have long existed in data systems, so enterprises must turn to new, modern strategies. Such agility requires a new information architecture, one that allows for seamless integration and operation across the entire data lifecycle.
Containerized architectures are key to this transformation. Today, 57% of organizations use containers with 89% expected by 2021.
IBM DataStage on IBM Cloud Pak for Data is the containerized version of IBM InfoSphere DataStage, based on a microservices architecture and optimized for Kubernetes. Through IBM Cloud Pak for Data,DataStage can run natively on Red Hat® OpenShift® the world's
leading container orchestration platform.
This paper demonstrates how IBM DataStage on Cloud Pak for Data provides:
• AI capabilities, built for AI projects
• Up to 50% lower cost of operations due to automatic failure resolution and automation of operational tasks
• 30% faster workload execution compared to traditional DataStage thanks to built-in workload balancing and best-in-breed parallel runtime
• 87% savings in development cost when using visual design
• Savings on data movement costs by bringing integration workloads to the data
• Pre-built integrations with data science, data warehouse and data virtualization services using a common UI