Big Data Forecast: Cloudy, with Increasing Chances of Success (Part 1)

Categories: Cloud Data Warehouse

Today, public cloud is a compelling proposition for businesses and government organizations seeking to be more agile. Increasingly, self-service is seen as the most effective way to scale user access to data for analytics and operations. Cloud elasticity, combined with the right user applications, can reduce the friction of waiting for IT to fulfill requests and provision resources and data. As such, we’re seeing cloud-based big data growing exponentially for Cloudera customers and across the market as a whole.

For these organizations, it’s essential that they are also managing the associated risks and costs of going to a cloud environment. One of the most common fears is vendor lock-in and becoming too dependent on a single cloud service provider, especially with the ever-evolving nature of the market’s competitive landscape. Many cloud providers have their own “house brand” offerings, but often these were cobbled together through a mish-mash of OEM deals, hasty acquisitions, or open source code, and thus heavily modified as to have been permanently forked. The result is sub-optimal, fragmented, and interoperability is often very limited. The vast volumes of big data in these environments imposes some challenges, by magnifying the long-term costs associated with renting cloud storage and compute resources. Understanding the technical and financial implications is crucial before becoming locked into a particular environment.

Further, businesses and governments need to ensure proper control of sensitive data. Any particular cloud provider may offer a range of data management services, often composed of a miscellany of acquired and home-grown products– some ported to run in cloud, some native. This creates complications, as fragmented offerings lead to data silos and having to manage multiple distinct tools for security and governance. Simply finding what data lives where, and who has access, becomes less simple if there is no common control plane for the data.

Hybrid- and multi-cloud environments are also becoming the norm for many companies. These strategies carry the potential to help find the right mix of qualities, while balancing costs and preserving negotiation power. Some data sets and applications can be kept within the walls of the data center or country borders, while other applications can live in whichever cloud is most optimal for its own requirements. The solution is maintaining a standard data management approach across the variant data platforms.

Tune in for Part 2 of 3, coming soon…


Leave a Reply