Tech/Engineering

Simplifying data protection – a new approach

February 6, 2020 Stephen Manley, Chief Technologist

Simplicity has always been a challenge when it comes to protecting data. More than once I had conversations with my engineers that went like this:

“We just made Oracle backups 43% simpler.” 

I stared blankly at the team, “How do you put a percentage on simplicity, and how did you come up with exactly 43%?” 

The lead engineer confidently responded, “We built a configuration wizard.”

“…But …How did you measure it?…Why that number?…OK, let’s move on.”

Such was life in the “good old days” of backup software.

Data protection should be simple, but for decades it hasn’t been. The data protection industry accepted complexity because we have been constrained by the architecture and consumption model. Since the cloud changes everything, we finally have a chance to simplify data protection.

Simplicity should be job 1 for data protection

You buy data protection products to make sure you can recover from failures and meet compliance requirements. Data protection is not a Tesla; it’s a Honda. You need it to just work.

The complexity of data protection means it doesn’t “just work.” Management is so manual that companies have to hire large teams to manage the backup environment, and the workflows are so convoluted that even experts make mistakes. Setting up data protection for a new application takes so long that business owners often bypass the backup team, so they can get their application deployed quickly. The result of this complexity is that data protection is expensive, unreliable, and incomplete.

The biggest simplicity challenges for data protection

Backup solutions are complicated because protecting data is a difficult (and growing) conundrum. It’s not that companies don’t want to simplify, it’s that they can’t.

Three challenges drive the complexity of data protection:

  1. Scale: Data protection solutions struggle to scale with the growth of data, so there is no time to simplify or automate. To be simple, data protection must scale automatically.
  2. Evolve with the environment: Customers (and vendors) struggle to maintain an existing backup environment, while also adding support for new applications (NoSQL), infrastructure (Kubernetes), and environments (cloud, remote offices). To be simple, data protection must seamlessly and rapidly support new applications and environments.
  3. Keep up with requirements: Organizations now expect the data protection team to also provide legal searches, ransomware protection, and disaster recovery. Meanwhile, governments demand compliance with data privacy, data retention, and data immutability regulations. To be simple, data protection must transparently expand functionality to keep up with ever expanding requirements for that data.

Data protection is not easy. It truly is “change the tire on the burning car as it barrels down the freeway while being chased by gangsters…as your mother calls and asks why you don’t come home more often.” But that doesn’t mean we should give up.

Why isn’t the industry more focused on simplicity?

Backup industry leaders talk about virtually everything except simplifying data protection. You’ll hear about multi-cloud, data reuse, speeds and feeds, or new functionality. But not simplicity. If it’s such a big challenge, why isn’t simplicity first on everybody’s list?

The data protection industry does not talk about simplicity because we’ve reached the architectural limit of disk-centric backups. Fifteen years ago, data deduplication created a shift from tape-centric to disk-centric backup. First, deduplicated virtual tape libraries (VTL) eliminated tape management. That led to disk-centric backup solutions which use source deduplication and incremental-forever backups to make protection faster and more centralized. This new architecture simplified backups — for a while.

Individual vendors don’t talk about simplicity because their solutions are not simple. Furthermore, since the vendors all have the same disk-centric architecture, they are equally complex. In that context, you can almost convince yourself that a setup wizard makes you 43% simpler than the competition. Almost.

Simplicity needs a new approach to data protection

The only way to change the industry is to embrace a revolution that is even larger just than an architectural shift. The way data protection is consumed needs to change as well. SaaS data protection combines the architectural shift to cloud-centric data protection with the move to an on-demand SaaS consumption model.

The joint shift in architecture and consumption model lays the foundation for addressing the complexity of data protection:

  1. Scale: A cloud-native SaaS solution scales up and scales down resources, so you get the performance you need, but pay for only what you use. The architecture and consumption model combine to automatically and cost-efficiently scale with data growth.
  2. Evolve with the environment: A cloud-native SaaS solution can protect data anywhere in the world – from data-center to remote office to the cloud itself. Furthermore, the model enables customers to leverage new application support without affecting their existing application protection. The cloud-architecture and SaaS consumption model combine to help protection evolve with the environment.
  3. Keep up with requirements: With a cloud-native SaaS solution, customers can segment data by region (compliance), business unit (governance), or group (chargeback). It provides the ultimate air-gap for ransomware protection. If they need to add functionality, they will only be charged for what they use, when they use it. The cloud-native SaaS data protection provider can even add new functionality on-the-fly to meet their evolving business needs.

It is possible to simplify data protection. The cloud-native SaaS model transforms how we architect, purchase, and manage data protection. It’s the first time we can address the fundamental challenges of simplifying data protection.

Change is coming

People buy data protection products to protect their data. Unfortunately, the complexity of disk-centric backup solutions makes them expensive and unreliable. They struggle to support the growing data, evolving environments, and expanding customer requirements. As revolutionary as deduplicated disk was, we face a new generation of requirements. The only way to truly simplify data protection is to change both the existing architectural and consumption models.

Cloud-native SaaS data protection can simplify data protection. Cloud-centric architectures scale up and down in a way the industry has never seen. The SaaS model enables protection providers and customers to keep pace with the evolving environment and expanding requirements. This is our chance to stop chasing simplicity and get ahead.

Then, with all that power, Druva enables customers to finally expand beyond protection to data management and more. While the potential to be the Tesla of data protection is electrifying, Druva always starts with job #1 — make data protection simple.

Learn more about cloud data protection with the Cloud Data Protection for Dummies guide, and how Druva makes data protection simple.