These days, hybrid is a fact of life, and with it the need for hybrid portability. If you’ve ever traveled between continents for work, you’ve certainly had to bring along your corporate laptop containing all your work materials. To make it function in the other country you bring some physical components, like a power cable with the correct plug and perhaps also a power supply that can adjust the voltage. You might have briefly pondered transferring your data onto a second laptop instead of taking your own, but it’s more efficient to do your work on a machine that you know has everything you need easily accessed. There’s also no guarantee that a new laptop would have the required specific security and governance your organization requires.
Following the rush for organizations to move all workloads exclusively to the cloud, many organizations found they needed to repatriate workloads to on-prem storage due to privacy regulations, and cost. It turns out that moving on-prem workloads to the cloud with the same process is very, very costly – and having costs rise rather than fall was certainly the opposite of the intended effect for businesses. Besides this, re-doing anything has the potential for errors to creep in, and many on-prem workloads simply shouldn’t move for security concerns.
This thought is echoed by IDC which sees the industry trending toward workloads first instead of heading straight to the cloud.
“Core infrastructure isn’t going anywhere. Given the application gravity today and governance complexities, the core data center remains the starting point for hybrid and multi-cloud excellence as it still hosts key workloads now and in the near future…For 95% of European organizations, enabling a multi-cloud environment is a priority, with 48% citing it as “very important” or “extremely important,” the IDC European 2022 Multicloud Survey states.
Here are three examples of organizations that unlocked increased value from hybrid data models:
- A large African telecommunications company was able to build up its analytics once and then deploy them on whatever infrastructure was available in the specific country of the African continent.
- An American Technology company was moving everything to the cloud to combine workloads and for flexibility and auto-scaling ability. They used Cloudera Distributed Hadoop (CDH) on-prem in two data centers and moved to CDP on Azure. The result was that Apache Spark jobs are now running faster on CDP, and they can run at 25% of full scale on weekdays, scaled out four times for weekends.
- A European Multinational Insurance company deployed CDP both on-prem and on Azure public cloud with Cloudera Data Science Workbench (CDSW) and Cloudera Machine Learning (CML). The result was teams had the same user experience between on-prem processes with CDSW and in the cloud with CML – no interruptions.
It’s clear that we need to use the cloud more smartly. It’s oversimplified to assert that all new workloads will be in the cloud while all old ones stay on-prem. Instead, let’s take a holistic approach. What workloads do we have? Where are they best placed?
Think of your data and the analytics you plan to run on it. Where are those analytics best deployed? There’s a mix of elements at play: performance, cost, regionality, ESG (carbon footprint), and more. This mix differs based on individual departments and teams within the same organization – and across multiple organizations, these factors differ further.
And, well, they’re subject to change. Imperatives today are different from weeks, months, and years ago. Look at what the pandemic threw at us (looking at you, remote work) and how quickly a change was needed.
Going forward, the speed of change will only increase. The bottom line is you need to be able to keep up with the changes. And that means you need portability, and you need to be able to move data and run analytics without redeveloping them. But here’s the thing. Portability only works if you are fully portable: not just workloads and data but everything that goes along with it. That means security, governance, and metadata.
With the right approach, hybrid portability can be a game-changer for businesses, enabling them to thrive in a fast-paced, interconnected world. We have some good news for you, though. Cloudera Data Platform (CDP) is the only hybrid data platform that works both on-prem and in any cloud environment that lets you move between them without redevelopment or re-architecture. And it’s fully integrated for security and governance with SDX, Cloudera’s built-in security layer for stronger governance, consistent risk mitigation, and continuous innovation with data. It’s one tool that covers your full data lifecycle, wherever it’s located, because, as we all know, hybrid is here to stay.
Learn more about CDP.
