The Future of Cloud-based Analytics (Part 3)

As the market moves toward cloud-based big data and analytics, three qualities emerge as vital for success. While many services will get some traction without meeting all three goals, they will also disappoint users and cause perpetual headaches for IT. At Cloudera, we see these undisputable attributes to be:

  • Easy – Certainly no one goes out looking for a harder way to do their job. Cloud IaaS facilitates resource self-service provisioning, eliminating the hassles of procurement and deployment on-premises. Cloud PaaS takes this a step further and allows users to focus directly on building data pipelines, training machine learning models, developing analytics applications — all the value creation efforts, vs the infrastructure operations. End-user focused tools accelerate daily tasks like job submission, performance tuning, and workload analytics. Intelligent defaults and built-in logic eliminate much of the guess work. The net result is much improved productivity for data engineers, data scientists, and analysts.
  • Unified – Conceptually, cloud sounds like a single place to host diverse, data-intensive functions. In practice, many services end up reproducing the silos that existed on-premises. A far superior approach is to truly consolidate data in one, persistent object store, and then bring different applications and workloads to bear against that set. Fragmented services lead to fragmented controls, when in actuality, what people really want is a common platform and control plane to manage everything, even across hybrid- and multi-cloud deployments. An advantageous side benefit of a unified approach is lower total cost of ownership, stemming from eliminating redundant data storage, leveraging transient compute, and simplifying management overhead.
  • Enterprise-grade – Perhaps this goes without saying, but enterprises need cloud to be every bit as robust as their traditional approaches. To be acceptable, cloud analytics platforms must meet or exceed corporate requirements around security, governance, and management. Central control for role-based access, authentication, authorization, encryption, keys – these are all must haves to pass audits and show compliance. The ability to discover and define metadata definitions for the business is a critical enabler for self-service functions. Not least, businesses will want a platform that has been proven out in the market by their most demanding peers.

What we repetitively see is that these goals are harder to meet than vendor marketing might suggest. Make sure any cloud-based analytics service meets these criteria. Ultimately, Cloudera’s goal is to efficiently deliver machine learning and advanced analytics capabilities that leverage the power of big data. These qualities are at the heart of our differentiation. Strategize on how the right approach and right technology can make data in the cloud your most important asset. We’re here to help.

The post The Future of Cloud-based Analytics (Part 3) appeared first on Cloudera VISION.

Leave a Comment

Your email address will not be published. Required fields are marked *