Disaster recovery

What Is Disaster Recovery?

Disaster recovery (DR) is the process of restoring access to applications and data as well as functionality to IT infrastructure after disruptive events such as a fire, flood, system failure, human error, or ransomware attack. The goal is to be operational amidst the disaster and return systems to normal as soon as possible. Disaster recovery is closely related to business continuity known together as BCDR. Although the two terms are often used interchangeably, there is a key difference. Whereas disaster recovery restores technology systems as rapidly as possible, business continuity is focused on keeping the organization as a whole operational. As such, disaster recovery is an important part of business continuity strategies.

Why Is Disaster Recovery Important?

Data and the digital technologies that create, store, process, and analyze it are essential for the running of a business. When a disaster strikes, business and reputation can be significantly impacted due to operational failures and the non-availability of mission-critical IT systems. Businesses must get access to data and restore functionality to systems and infrastructure as swiftly as possible. This is where DR comes in.

A robust disaster recovery strategy is critical to help organizations:

• • Minimize downtime — Systems and networks can go down for many reasons. Businesses use disaster recovery plans to preempt damage by anticipating and preparing for such events. Having a second site such as a public cloud for running mission-critical applications is one way of preparing for a disaster, for example.
  • Minimize data loss — Businesses rely on disaster recovery sites and technologies to prevent data loss. Because data can be replicated to a secondary site, in many cases, instantly, there is near-zero data loss. Moreover, leading DR solutions provide the flexibility to restore critical applications and data from any point in time or location. This helps bolster business continuity stances.
  • Mitigate overall damage — The costs of downtime inevitably percolate to other segments of your business—customer satisfaction can be negatively affected, brand reputation can take a hit, and competitors can pounce. With a robust disaster recovery solution, you can avoid these beyond-IT costs, which also include lost revenues and loss of employee productivity and morale.

How Does Disaster Recovery Work?

Recovery from a disaster can be a manual or automated process, depending on the business operation. Typically, digital DR involves getting mission-critical business applications, databases, and other IT systems up and running as quickly as possible to reduce downtime and prevent data loss.

Most organizations create a formal action plan that outlines the who, what, and how IT recovery should take place. The order of resources to be brought back up within measured time objectives to complete the recovery are typically detailed in the runbook.

Organizations have options for restoring applications and data replicated and mirrored to a secondary site. For example, the alternate site may have company-owned servers with mirrored applications and data waiting to be activated as a failover option in case of a disaster. Other organizations may choose a DR as a service capability from a cloud provider.

In all cases, automated disaster recovery converges point-in-time snapshots, replication, and automated failover and failback orchestration.

What Are Disaster Recovery Plans?

A disaster recovery plan, sometimes also called a DR runbook, is a core element of any business continuity plan. Once the plan is written, it should be regularly tested and modified to ensure it remains operational.

Disaster recovery plans will typically include two key metrics to prioritize the bringing up of key applications and data based on their criticality. This is typically measured in minutes, hours, days, or weeks:

• Recovery time objectives (RTOs) — How much time IT estimates will elapse between the disaster and return to productivity.
• Recovery point objectives (RPOs) — The maximum acceptable time between an outage or data loss event and the last backup, snapshot, or data sync.

The most effective DR planning and plans will detail the people and processes responsible for bringing which systems up in what order and sequence to address system dependencies and minimize downtime. Teams that use automated DR solutions that orchestrate their DR runbooks and processes can respond quickly and failover when an incident occurs.

What Is Disaster Recovery as a Service?

Similar to self-managed DR, disaster recovery as a service (DRaaS for short) also offers an automated way for organizations to control their data recovery and application availability service-level agreements, but without the cost and complexity of deploying and operating the secondary site themselves. The solution gives organizations the ability to rapidly recover while taking benefits of the cloud.

With DRaaS, organizations can spin up on-demand, pay-as-you-go, cloud infrastructure only when it’s needed. That eliminates costly and hard-to-manage secondary data centers that sit idle for most of the time. Teams enjoy near-zero downtime and minimize data loss across many service-level agreements (SLAs) for a variety of applications using disaster recovery services.

How to Write a Disaster Recovery Plan?

To create a disaster recovery plan, or DR runbook, requires first conducting an assessment of all of the people, processes, and technologies involved in IT. Without knowing this information before an unexpected, negative event—whether that’s a hurricane, flood, ransomware attack, or human mistake—it’s impossible to get back up and running fast. The DR plan may or may not be a component of a larger business continuity plan for restoring additional operations. DR plans typically focus on restoring IT systems as rapidly as possible from downtime.

Your DR plan should outline and include:

• The leadership and IT team handling crisis response – this can be the same team that wrote the DR plan or not
• Advanced software to automate IT software and system recovery
• Blueprints and scripts, plus data and IT governance protocols for enterprise IT systems
• Support/help resources
• Safety protocols for ensuring systems are brought up properly
• Internal communications channels about what is coming back online and when

How Do You Set Up Disaster Recovery?

From boardrooms to backrooms, all employees have some responsibility for safeguarding their organization’s data. CIOs and other IT leaders typically take the lead in setting up disaster recovery plans and technologies by working with executives and teams to prioritize the data, applications, and IT infrastructure that needs to be protected. An important part of this process is defining what resources are mission-critical —or absolutely required to operate—versus business-critical which is important to have but will not disrupt revenue or safety. Another important element of the process is to determine the service-level agreements (SLAs) others across the business have for specific capabilities. This can help IT teams determine whether they want to have on-site recovery responsibilities or team with a service or cloud provider to recover data, apps, and infrastructure. For example, will they choose on-premises or a cloud option, such as DRaaS with AWS, Microsoft Azure, or Google Cloud? Today, disaster recovery in AWS, DR in Azure, and DR in Google Cloud are growing in popularity.

Once strategic protection decisions are made, teams looking to set up disaster recovery plans and services can discuss how to restore operations in more detail. This is where the DR runbook comes in as it includes information about the people, processes, and technology requirements for recovery. Yet a DR runbook cannot sit on a shelf, but rather it must be tested regularly to ensure it remains relevant. Ease of maintenance and testing of DR capabilities will be another important consideration at this point.

Disaster recovery plan testing is going through each of the many steps outlined in the runbook to ensure the organization’s disaster recovery plan doesn’t have any gaps or errors. Testing of the DR plan ensures IT systems can and will be restored in the most timely and effective manner possible should the worst-case scenario occur.

For some, a DR solution that unifies backup and automates DR in a single solution to reduce complexity and costs of separate point solutions will be highly attractive because it supports both on-prem and cloud workloads with near-zero downtime and data loss.

What Is the Cost of Disaster Recovery?

As with every IT initiative, disaster recovery service and solution costs vary. Depending on the plan to isolate data physically or virtually, recovery costs can involve physically retrieving information from an offsite location hundreds of miles away from the primary location. Depending on the scale of the data or inconsistency of weather in locations, some organizations may choose to set up one or more secondary sites, which often involves installing multiple instances of costly hardware and software to replicate and store an exact copy of production data—and keep it running 24/7 just in case. Recent technology advancements, such as cloud computing and next-gen data management are significantly reducing DR costs. This is good news for organizations because depending on the severity of it, downtime can be catastrophic for an organization.

The financial cost of disasters—such as cyberattacks are already in the billions of dollars and are projected to rise to $256 billion in the next decade—but those costs don’t include potential loss of revenue, customer loyalty or satisfaction, and employee productivity. Disasters happen and are much more costly to businesses that are unprepared—which are those without disaster recovery solutions.

What Is Disaster Recovery Testing?

Disaster recovery testing gives IT teams the confidence they can meet business recovery SLAs. Testing also helps confirm the meeting of internal and external compliance requirements. With the rise in cyber attacks, proven DR testing may also soon become a prerequisite to qualify for cyber insurance.

How to Test a Disaster Recovery Plan?

The testing of a disaster recovery plan and services can be automated or manual. However, comprehensive testing will cover these essential elements—people, processes, technology.

Teams conducting testing should ensure a full review of the roles responsible for recovery, documents outlining recovery, recovery time and point objectives (RTOs/RPOs) commitments, and more.

In terms of process, testing should also include a review of what happens and what is needed in terms of alerting, procedures, hardware, software, networking, data protection, backup and recovery snapshots, ransomware recovery, rollbacks, and more.

Testing should occur at least once per quarter with best-in-class organizations testing monthly.

What Are Five Major Elements of a Typical Disaster Recovery Plan?

Organizations are prepared for rapid recovery should the unexpected happen if they have these five elements of a disaster recovery plan in place already:

1. Identify a response team — Assign the right teams and staff roles to develop and execute the DR response plan.
2. Define service-level agreements (SLAs) and assess risks — As part of pre-planning and DR plan creation, document expected recovery objectives and assess the risks associated with recovery from a wide range of possibilities (e.g., ransomware movement within systems).
3. Document critical systems powering operations — Ensure everyone is aware of the systems needed to survive a disaster or threat in the short-, medium- and long-term periods of recovery.
4. Optionally also deploy a modern backup and data management solution — With a next-gen data management solution, you can deploy DR separately or converge it with backup for optimal efficiency, cost reduction and to eliminate complexity. Evaluate and choose a flexible, comprehensive multicloud data management platform such as Cohesity to automate backups, continuous data protection, and DR failover and failback orchestration across business-critical applications, service levels, and environments with near-zero downtime and no data loss.
5. Test and continually update the DR runbook at least quarterly — As operations change and data grows at exponential rates, ensure the disaster recovery plan or DR service remains operational so you can execute it flawlessly at a moment’s notice.

What Is the Best Method for Disaster Recovery?

There are multiple ways to implement disaster recovery, but we recommend looking for solutions that allow you to address a wide range of SLAs and recovery times while minimizing downtime, reduce overall system and operational complexity, and reduce costs by not having as much duplicate or idle secondary infrastructure. Also look for flexibility to allow you self-manage your DR deployment or have it managed for you, the cloud or a DRaaS model.

What Are the Types of Disaster Recovery?

Organizations typically architect disaster recovery sites to best meet their needs. The most popular options are:

• Data center to secondary data center (site to site) disaster recovery
• Data center to cloud disaster recovery (site to cloud)
• Disaster recovery as a service (DRaaS)

How to Build a Disaster Recovery Team?

Your disaster recovery team will typically be a subset of your business continuity team. The roles on that team include the CIO, IT resilience, crisis response, and security response roles.

Members of the team responsible for DR will typically be technical professionals with data center—compute, storage, networking, and cloud—responsibilities because the primary goal of a DR plan is to recover applications, data, and infrastructure quickly and completely.

What Are the Benefits of Disaster Recovery Software?

The most reliable disaster recovery or disaster recovery as a service (DRaaS) will enable organizations to:

• Simplify DR operations by replacing old, siloed point products with a unified solution that protects applications and data on-premises and in clouds.
• Protect against site failure with replication from one site to another, on-prem and in the cloud.
• Automate replication and save time with policy-based automation for backups and replication in a hybrid cloud environment.
• Support non-disruptive DR testing as well as audit trails and reporting.

Disaster Recovery and Cohesity

Disaster recovery can be complex and expensive. A business running hundreds of applications needs to tier these applications in terms of criticality, define separate policies, work with multiple vendors for each tier, and manage them all through disparate consoles. But Cohesity has introduced a solution that helps customers not only recover from a disaster almost instantly, it does so for every tier of application deployed. Complex, expensive, and fragmented solutions are a thing of the past with a unified and automated DR failover and failback orchestration solution.

Cohesity’s reliable disaster recovery and business continuity solution:

• Simplifies DR operations — Cohesity replaces the time-consuming management of organizations’ siloed, legacy DR point products with a unified policy framework in a single solution that protects applications and data — across tiers, service levels, and environments — both on-premises and in clouds.
• Protects against site failure — The disaster recovery solution from Cohesity gives businesses the ability to replicate from one site to another, on-prem and in the cloud, to guard against complete failures.
• Automates replication — Organizations can use Cohesity’s policy-based automation for backups and replication in a hybrid cloud environment.
• Allows for data reuse — The Cohesity DR solution lets organizations easily replicate data to an alternate location for other purposes, such as dev/test or analytics.
• Supports non-disruptive DR testing — Businesses reduce operational complexity and streamline compliance requirements with non-disruptive DR testing, audit trails, and reporting in the Cohesity solution.