Using Colocated Servers as an Effective Backup Plan

Disaster-Recovery-PlanningDisaster recovery is one of the scariest words in the IT world.

Did you know that 40% of IT professionals believe that their small business would permanently go out of business should they encounter a total loss of their data.

While disaster recovery is something that you will never want to go through in your career, the ugly truth is that your enterprise will encounter some sort of event in which data will need to be retrieved from backups. It’s almost inevitable.

Using Colocated Servers as a Backup Plan

Consider the following scenario:

Your organization has excess hardware that it would like to put to use. One of the best ways to ensure maximum uptime to send this hardware to a colocated datacenter and purpose this hardware as a dedicated backup infrastructure.

You may already have a colocated server workload for your organization and you may simply have extra space on your rack. If that’s the case, put your extra infrastructure to work for your business.

IMG_2024-X3Colocated vs. Cloud for Backups?

Every organization is different. Some businesses believe that they must be in complete control of their corporate data. When you give your data to a cloud backup organization, you put your entire data backup infrastructure in the hands of another entity.

What if your business finds itself needing a data backup retrieval. However, for some reason, the data backup the requires help of the cloud service provider. What if you need the help immediately? Can you count on your cloud provider to be there for the moment you need them the most?

By building your data backup infrastructure in a colocated datacenter, you can be sure that your organization has a warm standby server that is available for failover incase something bad happens to your network.

You can also setup file level recovery and rapidly move the data back to your production servers, should you need to restore data that is missing from your servers.

One of the things that businesses must realize is that when they use a cloud data syncing service, they are at the mercy of the bandwidth available between the cloud backup provider and your onsite infrastructure.

If your infrastructure is already setup in a  data center colocation scenario, you could colocate your workload in the same datacenter or in a nearby datacenter that is within the family of your colo provider. This type of setup could provide your network with fault tolerance.

Using the colocated infrastructure backup scenario, you can rapidly recover your data since the data won’t have to travel as far to reach your production environment. This can be advantageous for businesses who wish to be one step ahead of any potential problems that could arise from putting your data backup infrastructure in the hands of a 3rd party solution.

Recent Amazon EC2 Outages Boost the Appeal of Dedicated Servers

url51If you’re an IT decision maker, you’ve probably considered using Amazon Web Services’ EC2 virtual server offering. AWS gives your team the ability to spin up virtual servers on demand and pay for only what you use.

The Infrastructure as a Service (IaaS) model can provide advantages in certain deployment scenarios. On the other hand, hosting critical apps and services that keep your business flowing takes a tremendous amount of trust in an IaaS provider.

Can You Trust the Cloud?

At the end of the day, IaaS providers such as AWS must own up to their service level agreements when outages occur.

If you’re in a business that requires maximum uptime, wouldn’t it make sense to have the maximum amount of control over the infrastructure that houses your apps and data?

When you choose a service like AWS, you’re at their mercy should the right set of circumstances occur. Think about what happened to Google Cloud Platform back in August of 2015.

Lightning Strikes Google Cloud Platform; Causes Data Loss

A lightning strike that hit Google’s Western European data center caused a data loss that measured to be less than 0.000001%. While that may not sound awful, you’d probably be upset if your business happened to be impacted by the 0.000001% figure.

amazon-downAWS DynamoDB Outage Breaks Reddit, Netflix and More

In September of 2015, Amazon Web Services suffered an outage stemming from its DynamoDB service. This left users of popular apps such as Netflix, Tinder and Reddit without the ability to access these platforms.

Reviewing Popular IaaS Provider Uptime Numbers

In 2014, each of the “Big Three” cloud providers suffered some sort of downtime. Here are the numbers for each IaaS provider:

  • AWS – 6.01 Hours of Downtime in 2014
  • Google Compute Engine – 3.46 Hours of Downtime in 2014
  • Microsoft Azure – 42.94 hours of Downtime in 2014

Experiencing downtime that is totally out of your control can become frustrating as an IT professional.

The Great Debate: Colo vs. Cloud

While cloud can provide several strategic advantages to organizations that rapidly need to build and deploy apps, colocation provides IT decision makers with the ability to completely customize, administrate and have complete confidence that their workload will have the maximum uptime possible.

There’s a certain confidence in knowing exactly where a server sits in a cage and being able to trace that server back to the switch. With IaaS, businesses risk running their apps and storing their data in facility that they don’t control. For many IT decision makers, that’s a risk that their organization can’t afford to take.