---
title: "When Resilience Becomes a Cloud Service: The Future of Native Recovery | SHIFT 2026 On-Demand"
type: "WebPage"
language: "en-US"
url: "https://www.commvault.com/resources/video/when-resilience-becomes-a-cloud-service"
date: "2026-11-13T15:22:57-05:00"
modified: "2026-11-13T15:22:57-05:00"
description: "Explore the future of native cloud recovery and learn how resilience as a cloud service can simplify recovery and strengthen cyber resilience. Watch on demand."
image: "https://www.commvault.com/wp-content/uploads/2026/11/When-Resilience-Becomes-a-Cloud-Service-The-Future-of-Native-Recovery.png"
breadcrumbs:
  - name: "Commvault Home"
    url: "https://www.commvault.com/"
  - name: "Resources"
    url: "https://www.commvault.com/resources"
  - name: "Video"
    url: "https://www.commvault.com/resources/video"
  - name: "When resilience becomes a cloud service"
---

# When Resilience Becomes a Cloud Service • SHIFT On-Demand 2026

- [Home](/)
- Videos
- When Resilience Becomes a Cloud Service • SHIFT On-Demand 2026

ON-DEMAND SESSION

## When Resilience Becomes a Cloud Service: The Future of Native Recovery

Learn how application resilience goes beyond backup by protecting cloud configuration, dependencies, and workload data together for complete, tested recovery.

[Request a demo](/request-demo)

Play Video      ![Video thumbnail](https://img.youtube.com/vi/bHwSKyfHclI/maxresdefault.jpg)

---

- [Recovery Fundamentals](#fundamentals)
- [About This Session](#About)
- [Related Resources](#Resources)
- [FAQs](#FAQs)

[Back to all on-demand sessions](/shift#watch-now)

---

## Recovery Fundamentals

**Backup is Not Resilience**

Resilience restores data, configuration, and dependencies together.

**Protect Application Context**

Protection groups unite workload data and cloud configuration.

**Rebuild Cloud Dependencies**

Cloud Rewind reconstructs infrastructure before restoring application data.

**Test Recovery Continuously**

Testing proves recovery works before a disruption occurs.

**Recover in the Right Order**

Sequence data, configuration, and dependencies to restart business operations.

**Close Cloud Recovery Gaps**

Discover assets and restore the complete application.

---

## About this session

**Cloud recovery must restore more than data.**

Cloud applications depend on networking, identity, DNS, certificates, load balancers, service configuration, and data. Restoring files alone does not restore the business.

#### Backup Versus Resilience

A backup is a point-in-time copy of data. Resilience restores the system around that data, in the right order, while teams make decisions under pressure.

**Application resilience closes the gap between data protection and business recovery.**

Protection groups capture workload data and cloud configuration together as one application-focused recovery unit. `

#### Rebuild What Matters

Cloud Rewind reconstructs application infrastructure and dependencies before restoring the application data needed to bring services back online.

**Recovery confidence comes from testing.**

Plans can drift from production when recovery testing is disruptive or postponed.

#### Restore the Business

The goal is to restore the complete application and prove it works.

CONTINUOUS CORE RECOVERY

## CoreRecover

Unified, validated infrastructure recovery

[Explore solution](https://www.commvault.com/solutions/core-recover)

CONTINUOUS CLOUD RECOVERY

## CloudRecover

Recovery built for the cloud. Proven at scale. Ready for AI.

[Explore solution](https://www.commvault.com/solutions/cloud-recover)

---

## Related resources

SOLUTION BRIEF

## Microsoft Azure Cloud Application Resilience

Protect and recover entire cloud environments with Cloud Rewind.

[Read more about Microsoft Azure Cloud Application Resilience](https://www.commvault.com/resources/solution-brief/microsoft-azure-cloud-application-resilience)

eBOOK

## Building Cloud Resilience

Learn how to discover, protect, recover, and rebuild cloud environments.

[Download ebook about Building Cloud Resilience](https://www.commvault.com/resources/ebook/building-cloud-resilience)

eBOOK

## Applications Resilience Use Cases with Cloud Rewind

Explore application resilience use cases with Cloud Rewind.

[Read the ebook about Applications Resilience Use Cases with Cloud Rewind](https://www.commvault.com/resources/ebook/applications-resilience-use-cases-with-cloud-rewind)

---

## Frequently Asked Questions

What is the difference between backup and resilience?

Backup is a copy of data at a point in time. Resilience restores the surrounding system, access, dependencies, and business operations in the right sequence.

Why is infrastructure as code not enough?

Infrastructure as code describes what should be true. Recovery must also know what actually existed at a specific moment, including changes never captured in code.

What are protection groups?

Protection groups protect workload data and cloud configuration together as one application-focused recovery unit, avoiding disconnected recovery workflows.

How does cloud rewind rebuild applications?

Cloud Rewind uses protected configuration to reconstruct application infrastructure and dependencies, then restores protected data into the recovered environment.

Why should cloud recovery be tested?

Testing provides evidence that recovery works before an incident. It exposes missing dependencies, outdated plans, and gaps that data backup alone cannot solve.

---

## Transcript

View full transcript

Thank you for joining us.

I want to start this session with a question, and I would like you to sit down with it for a few seconds. And here’s the question. If your production cloud account disappeared like not not just the data, the whole account, how would you cover your business?

Most people start with the data. So they say we have backups, we have snapshots. And your file. Then a few seconds later someone says, wait, who is going to rebuild the network? How about the permissions? The DNS configuration. And that’s when the conversation goes quite what I want to walk through today is this gap that most cloud teams don’t have an answer for yet.

Let’s dive in.

The agenda for today’s session. We’ll spend the first part of the discussion identifying the problem. Why the confidence that most teams have about cloud recovery doesn’t match the reality. Then we will get precise about two words people use interchangeably. Backup and resilience. From there, we will go one level deeper into the difference between data resilience and application resilience, and where most current practices fall short.

Then I want to spend some time on where we think this is headed, our vision and the product to solve for the complete application Resilience directory. Then finally, let’s get real with a five minute demo

Let’s start with two numbers. Public cloud spend is growing about 33% a year. It’s on track to nearly triple to roughly half a trillion. More of your business is running in the cloud every quarter.

The second number, when something important goes down, it costs about $2 million an hour on an average. Now put those two numbers together. Most of your business is in the cloud, and every hour of their downtime keeps getting more expensive.

The stakes are only going in one direction here. So are the teams ready for that? Most of them think they are until they realize they don’t.

According to Gartner, 86% of operations leaders believe their disaster recovery capabilities either meet or exceed what they see was expected. Them. Sounds good, except only 27% have actually completed the three foundational elements of failure disaster Recovery program. So there is a massive confidence gap. Most leaders think they are covered, but most of them are back. That’s not a knock on anyone here.

It’s a structural problem. And here’s the reason why. The only way to know your recovery works. The statistic. But most die testing is disrupted. It takes people time and sometimes even includes downtime. So teams plan the drill, but they push it out there next quarter. Then another quarter. And then the plan in the paper slowly drifts away from what’s actually running in production.

If testing recovery is painful, people won’t do it. So it has to get easy. However, there’s even more. Underneath this gap. There’s something even more basic. 67% of organizations report blind spots in cloud asset inventories and configuration patching, meaning the stuff running in your application that isn’t fully documented or tracked. You can’t cover what you don’t know exists.

70% of organizations are poorly positioned for cloud disaster recovery. Roll out. Then there is scale. A single production appliance today can span 7 to 10 layers of dependencies. Networking, IAM, DNS certifications, load balancers, and worker, often across multiple accounts, subscriptions and legions. So when something breaks, someone has to reconstruct all of that. Usually from a script, a runbook.

Honestly speaking, somebody’s memory of how it is supposed to work.

So here’s a point that I want to learn before we move on. This isn’t really a tooling problem. It’s a definition problem. Most organizations don’t have a name for the specific thing they’re missing, and that’s what we are going to name next.

Let’s get precise about two words that get used interchangeably, but shouldn’t be backup and resilient. A backup is simple. It’s a copy of your data at a point in time that you can restore. That’s it. That’s a whole job. However, resilience is a bigger thing. It’s the ability to keep the business running, which means restoring not just the data, but the system surrounding the access to it and doing all of that in the right order under the pressure, while people are making decisions in real time.

Here’s the line that I want you to remember. A backup that can’t be trusted hasn’t been tested or gets restored in the wrong order. Does it actually give the business what it needs? You can have a perfectly good backup and still have a bad recovery. So resilient. He isn’t an insurance policy. You buy once. It’s the operational capability that you have to keep flowing.

Now, the natural reaction in this room is probably okay, but isn’t that what our infrastructure as code already gives us? Totally free question. I wonder actually directly, because it’s the crux of the whole talk back here.

I see tooling managers decide state it rewards by reapplying the code. It’s telling the environment. Here is what you supposed to look like. Not here is exactly what you look like at 2:14 p.m. before things look. Uncritically depends entirely on your state files and version history being intact, being current inaccurate. If someone made a change to the console and they never got captured in the code, that IAC tooling has no idea that existed.

It keeps no historical snapshot of what was actually it. So the line I would offer here is I think it is the cleanest way to say it as well. His code tells you what should be true, and recovery has to know what actually was. Most cloud teams have gotten reasonably good at data resilience snapshots, replication, versioning that Marcel is fairly mature.

But here is what’s not mature. The application resilience when an application actually goes down. What breaks isn’t just the data, it’s the networking. The key DNS certificates, load balancers, service configuration. Unfortunately, none of that leaves in a data backup. That’s the gap that we set out to close with, something we call as protection groups. So the ideas here protect data and configuration together captured as one.

Applying the protection to the application, not to a list of random disconnected resources and recover across boundaries across region subscriptions and connects. I’ll go deeper on each of these in a few minutes for now. Keep this picture in mind. The data in the middle and everything around it that also has to come back. Now let’s put this recovery lifecycle that you already know.

This lifecycle should look familiar by now. Detect a threat. Protect and create clean security points. Validate what’s safe to restore. Restore critical operations in the right sequence. Learn for the next time. That’s the discipline. Most organizations already apply to their data. It’s mature. And for the data itself, it mostly works. But notice the sixth stage here labeled. That’s not a data concept.

That’s the stage. This whole conversation has been circling. Detect protect validate restore and learn. All have well-worn playbooks. When you’re talking about databases, files and virtual machines and so on. However, rebuilding is what it takes to bring back everything we just looked at on the previous slide. The networking keyword load balancing and the configurations. The stuff that has no equivalent recovery point in a traditional backup.

So this is really about two life cycles overlaid on each other. One for data just mature and well-understood, and the other for the application configuration, which is newer and frankly, still being built out across the industry. Rebuild is where the second life cycle is. And what would and what would it take to make rebuild as mature as data and technology for the data?

That’s where we think this is headed. So where does this go today? Most organizations still treat data production and configuration protection as two separate problems. Different teams, different tools, different workflows. We believe that has to converge. The direction we are building toward is something we think of as protection groups. The idea that an application data and its cloud configuration get protected together as one unit instead of has two disconnected motions.

Three principles underneath this. First, protect together workload data and cloud configuration capture together with eight gapped point in time copies of both second aligned to the application protection organization around the application boundary. Not a file of disconnected day. So it’s six. Third become a class boundaries locally planned and executed across Legion subscriptions and tenants from a single workflow.

Instead of stitching several tools together within seven. That’s protection groups one recovery unit for the entire application. And here’s what that unlocks. Imagine protecting an entire application, the data and everything it depends on as a single unit at the scale of real enterprise state. Not a handful of pilot workloads. And imagine a recovery orchestrated to a one book that sequences itself.

Data first, then configuration, then dependencies the right order with the checks and approvals your team already requires, and so someone executing each of these steps by hand while the clock is running. That’s the real shift we are describing today. Not not a faster restart, but in a different relationship between an enterprise and its own. Going. The goal stops being reached all the resources and now becomes to restart the business.

That’s where we think this category is headed and it’s worth showing, not just describing. Let’s take a look at it.

Welcome to Commvault Cloud. In this demonstration we’ll show how Commvault Cloud brings application centric cloud configuration and workload data protection together through protection groups. Let’s start by creating a protection group for our open EMR healthcare application. We begin by selecting our Azure Cloud connection and giving the protection group a recognizable name. Next, we select the workload types that support our application, including databases, object storage, and cloud configuration.

We define the application boundary using a resource group based rule. This allows Commvault to identify the Azure resources belonging to the application.

Next, we assign a backup plan to define the protection schedule, retention and storage. After reviewing the configuration, we submit the protection group and initiate an on demand backup. With protection groups. Customers can bring cloud configuration and workload protection together under a common application construct. Rather than managing each workload independently. Now let’s look at application recovery. Starting with the cloud infrastructure within Commvault Cloud.

Cloud configuration powered by Cloud Rewind automatically creates a configuration group for our protected application. We select the open EMR configuration group to review the protected environment. Here we can see the application centric view of our Azure environment. Cloud configuration has discovered the App Service, Azure Key Vault MySQL flexible server configuration and supporting networking resources. These dependencies are critical because restoring application data alone is not enough to bring a business application back online.

Now we’ll initiate recovery of the application environment. We review the protected resources and select Restore. We specify a recovery prefix to distinguish the reconstructed resources from the original environment. We then map the source region to West U.S. two and submit the recovery. Cloud Rewind uses the protected configuration to reconstruct the Azure application infrastructure and its dependencies.

Let’s review the reconstructed environment in Azure. Here we can see our original application resources alongside the disaster recovery environment. The recovered environment includes the app service, networking and supporting Azure resources. This demonstrates the value of reconstructing the complete application infrastructure rather than manually rebuilding individual resources with the application infrastructure rebuilt. The next step is restoring the application data.

We return to Commvault Coud, navigate to databases and select the protected open EMR MySQL database. We select the appropriate recovery point and choose the reconstructed MySQL instance as our recovery destination. Commvault’s workload recovery capabilities. Restore the protected database into the rebuilt application environment. With both the cloud infrastructure and application data restored. Open EMR is now accessible. This is the foundation of application centric resilience and Commvault Cloud.

Protecting the complete application context, rebuilding its infrastructure, and recovering the data needed to restore business services.

Thank you all for sticking with me. Here’s one thing I would like you to take away. A good backup alone does not guarantee a complete recovery in cloud. Resilience means bringing back the whole application in the right order and proving it works. If you’d like to see this learning in your own environment, reach out to the county. Thanks again and have a great rest of the day.
