How to Design a Cloudy-Ready Disaster Recovery Infrastructure

Regardless of the emergencies that might come up every now and then, it is paramount for business processes to stay afloat for the good of the business. All kinds of things happen not only on cloud but even in the private circles and it is vital to stay prepared.

For the sake of your continuous business operations, it is crucial to build a surrounding that is able to support this. A pivotal part of this includes understanding that there is a probability of the cloud having down time or going down. To understand how detrimental this can be, take the example of Microsoft Azure where an SSL certificate expired taking down massive important public components of the cloud down with it. In the past, not many companies found it palatable to enroll in Disaster Recovery or even discuss matters pertaining to business continuity.

However, it is crucial to understand that these two; Business Continuity and Disaster Recovery are very different business objectives. However, they are reliant on each other and hence overlap. To date the ability of the cloud is not fully understood. However, it is not really about the functioning of the cloud but about the capabilities of Disaster Recovery.

Looking at the cloud as a Disaster Recovery or a business opportunity is something that every enterprise large or small is going to have to do at some point. Most organizations have already made the intelligent decision to move into the cloud. Some companies offer a recovery solution that seeks to protect organizations from loss of data, reputation and even revenue should an adverse situation occur. This is by helping them recover their IT resources in an efficient and effective manner. The point is to combine the benefits that a multi-tenant cloud platform has to offer in conjunction with the most recent recovery technology resulting into a testable, very reliable organization grade system of recovery.

There are a number of cloud technologies that have become key pillars in improving the IT redundancy business. It is vital to get a glimpse of these before we cover the execution aspect of a DRBC (Data Recovery and Business Continuity) strategy.

Some of these include:

Cloud-based Traffic Management

The ability to control over great distances using Global Server Load Balancing (GSLB) and Global Traffic Management (GTM) has positioned the cloud to becoming a real hub that can be used for different Disaster Recovery strategies.

Everything is Software-defined

With the increasing ability of using modern day networking controllers, using one physical switch port to create thousands of virtual connection points, controlling layer 4-7 traffic at better granular levels has led to more powerful and resilient cloud models that offer better storage computing and security.

Automation and Virtualization

Virtualization has allowed for back up of vast kinds of workloads and images, sharing and replication. Depending on what the organization needs, the images can then be span to global data centers. This mated with automation then makes seamless work of spanning the platforms to different infrastructures with better resource control.

New hardware and storage intelligence types

Even though cloud storage has come a long way, the ability to control storage regardless of whether it is on the more advanced open-source or the more traditional EMC platform has become more advanced. Some aspects like quick site to site replication and controller multi-tenancy make the cloud a pivotal piece of the Disaster Recovery process. The logical layer that the storage offers is equally powerful and the storage that is software designed allows other infrastructures like Virtual Machines (VMs) to scale dynamically.

With that said, here are some of the ways that DRBC can be applied in modern organizations.

First, it is imperative for every organization to conduct Business Impact Analysis (BIA) to identify effective recovery time objectives and an appropriate internal Disaster Recovery strategy.

There are a variety of cloud model options that administrators can look into that include:

Active or hot site configurations

In this case there is minimal downtime that occurs in the occurrence of an emergency. You will need a hot site that is constantly operational which is still a necessity but this can be a very pricey package.

Cold or warm site Active/Passive

Some downtime though not prolonged should be expected. However, this will largely depend on the BIA. This is more applicable to the public dynamic cloud model and is less expensive.

DRBC that is work-load based

In this case not the whole infrastructure requires to be recovered, only a certain criteria of services, databases and applications. Even though the databases, applications and other services are still stored in the cloud, in the event that an emergency occurs, they can be provided or mirrored live.

Recovery that is back-up based

Even though the downtime might not be a primary factor in this case, the workload or the application in question needs to be brought up quite fast.

These also work on the basis of replicating data, services and other applications to a back-up that is VM-based. The process is a bit slower depending on the contract and whether you are looking to have the entire workload or just specific parts.

 Recovery-as-a-Service (DRaaS or RaaS)

The best thing about this is the flexibility that it offers. You don’t have to worry about downtime in this environment. You can also specify the recovery of certain workloads or applications in case there is an emergency.

Creating a dynamic solution is not much of a challenge since there is already integration into virtual systems that are underlying. Furthermore, instead of charging you for the resource that you use, you get to pay for what you recover.

Keep in mind that there are a variety of services that are being offered by the different cloud providers and vendors. This is even better since the services are being offered to larger organizations of varying sizes. The cloud can be the best platform for your Disaster Recovery given its better replications, resilience and diverse avenues of  support.