floppy discs

Tell me if you've heard this story before: A growing “entourage” of people depend on me for technical support. It started when my father decided to become a computer whiz at the age of 70. Surprisingly, his plan went pretty well and I am happy to report that he now owns a smartphone and three laptops. 

Then things got complicated. He told his friends and they told their friends. Soon, 25 or so “TeamViewer buddies” were “eternally grateful” for my help and promised that this was “(cough) the last time they will bother me” with questions.

Separating Cloud Promises and Realities

A few weeks ago, I received a text from one of the entourage who owns a small business. Generally speaking, I draw the line at providing free support to businesses, but he promised a simple question. Should he upgrade his Windows 7 computers to 10? The business often shared large files amongst employees and customers, so I told him how well the new OneDrive sync client integration into Windows 10 was working. 

This got me thinking about how many people automatically upload files into OneDrive and how, as more users sign into Windows 10 with their work accounts, OneDrive will become the de-facto backup for users’ day-to-day files in the enterprise.

While OneDrive is not meant to be a backup or storage system, users will always go with what works and provides the path of least resistance. And as companies such as Microsoft, Google, Box and Dropbox succeed in pushing the stability, convenience and accessibility of the cloud, users treat these solutions as cheap, quick alternatives to traditional storage and backup. While this makes the companies selling these cloud platforms happy, it is a less than ideal situation in the enterprise. Administrators, IT directors and CIOs see their collaboration platforms transform into content preservation and compliance nightmares.

There's a significant difference between the perception and the reality of cloud solutions’ reliability and accessibility. The promise of 99 percent uptime and an “anywhere, anytime” access model has convinced many users and companies to trust that their data is both safe and accessible in the cloud. Depending on the platform, the uptime promises can be fairly accurate and considerably higher than what an internal or hosted datacenter provides. That said, the promise of additional uptime tends to overshadow a much more serious consideration: In the instances where the system does fail, productivity and accessibility are much harder, if not impossible to maintain.

The Cost of 5 Minutes Downtime

We’ve seen this before: A large cloud storage or computing provider experiences a problem in their datacenter — the cooling system goes offline, there’s a problem with an update, or the failsafe software has a bug. Suddenly individuals, companies and other large cloud service providers experience widespread slowdowns and/or outages. In the last three weeks we’ve seen the real life effects of that small percentage of downtime. 

Three weeks ago, Google suffered a five minute outage that cost it an estimated $545,000. The next week, we saw AWS go dark for 25 minutes, taking the Amazon.com website as well as Netflix, Vine, Instagram, IFTTT, Airbnb, Flipboard and many others along with it. For the US Amazon.com site alone, it is estimated that the outage could have cost Amazon’s retail arm as much as $1100 in net sales per second. That’s $1.65 million for an outage that lasted less than half an hour. A few days later, Microsoft’s Skype went down for 15 hours, completely removing users’ ability to communicate and collaborate.

At a time when even five minutes of downtime can cost half a million dollars, CIOs, IT directors and administrators should be extremely concerned about backup options. The entire company is pushing towards the cloud, placing more and more of the day to day operations in the hands of cloud service providers while the employees are dumping mission critical content into file sync and swap systems. This begs the question, “What are our backup options?” 

Backup Options

Realistically, there are two: OOTB (out of the box) and third party. OOTB usually focuses on disaster recovery, offering environment-wide backups for a finite amount of time that can be used to restore your environment en-masse. In the case of Office 365, the focus is on creating backups in the event of wide scale data loss and a restoration in the event of a hosting data center disaster. Bear in mind these 14-day backups can only be restored if you have network connectivity.

Out of the box is therefore limited, due to its focus on disaster recovery. These solutions are usually designed for catastrophic events, rather than the unspecified failure of one of the many moving parts companies rely on when investing in the cloud. The major outages of the last three weeks were not the result of catastrophic data loss, but the loss of access. A failed service or a botched configuration upgrade caused a chain reaction that ultimately overwhelmed the other services and knocked them offline. 

This is where third party options tend to excel. They reach across the siloed services in cloud offerings such as Office 365 and offer backup for more than just a single cloud vendor or SaaS offering. And unlike their out of the box counterparts, they account for the most basic of outages — loss of network connectivity — by creating accessible local backups in the event that you're unable to reach cloud services.

Aside from disaster recovery and maintaining content access, ensuring users have quick and easy access to needed documents is a priority. This is also one of the most time-consuming tasks. Determining what content is outdated, what content is in active use and what content should be archived for potential future use is too large a task for any one administrator, but is essential to maximizing ROI. 

Microsoft's OOTB options for the services within Office 365 vary widely in terms of both functionality and ease of use. For example, Microsoft now offers Exchange Online Archiving for an extra $3/per user per month, but this can be an expensive addition for larger companies and is (once again) limited by the ability to reach the cloud service. Office 365’s SharePoint Online, however, is left without a similar archiving option. With disjointed methods, multiple consoles to deal with and functionality that ranges from adequate to nonexistent, OOTB offerings are far from ideal when considering archiving options.

Third party tools have the benefit of being built from the ground-up to offer a full range of archiving services in addition to disaster recovery offerings. Third party tools have the advantage of being outside of the Microsoft stack, working across the many services to give you a unified backup. Archiving and extracting content from your environment is possible, which frees up space, reduces processing time and provides users with a cleaner workspace with more relevant content. Additionally, some third party tools offer data normalization, allowing your organization to store the backups locally as regular files that can be searched, accessed and restored to users, without the need for conversion or proprietary readers. Finally, third party tools have the added advantage of creating backups that are accessible even when your network connectivity has been interrupted.

While third party tools are good, it is ultimately up to us as CIOs, IT directors, administrators and stewards of our data to decide what backup will look like for our specific cloud implementation. The only universal certainty is that for moves to the cloud to succeed, our conception and implementation of backup has to be much further reaching than it is today.

Creative Commons Creative Commons Attribution-Share Alike 2.0 Generic License Title image by  sunxez