Information Technology Navigator

Tips, Advice & Insights from Technology Pros

Cloud Backup – You Might as well Jump!

Posted by Matthew Brady

Thu, May 14, 2015

Cloud Backup

I chuckled to myself as I drove to work this morning. An ad came on the radio talking in clichés. “Synergy!” “Run it up the flag pole.” “Better safe than sorry.” And that was enough to send me into action and write this blog post.

The clichés and metaphors surrounding the cloud are seemingly endless. You read about “The cloud journey.” “The cloudy IT forecast,” and on and on.  It’s no secret that many organizations are looking for ways to take advantage of the benefits of the cloud. Moving to the cloud can reduce capital expenditures, eliminate over provisioning, lighten employee workloads, free IT staff to work on strategic initiatives, and help businesses become more agile.


No doubt cloud backup can do all those things. However, you have to change the way you look at storage and your business.

Get ready for it. Yup, you need a paradigm shift.

3 Paradigm Shifts Impacting Cloud Backup

Let’s look at three paradigm shifts needed to ensure your migration to cloud backup is successful today, and in the future.

1. You need to understand your data better

How much of your data changes daily? How many files do you have? What kinds of files are they? Are there backup or dump files on the network? Do you have someone’s Van Halen collection on your network? Is the company shared drive clogged with wedding pictures? Which data is actually required to be backed up?

Questions, questions, and more questions. What does that have to do with cloud backup? It’s pretty simple. If it is on your network, you are probably going to back it up unless you know it’s true value. Large files and large numbers of files all impact cloud backup performance. And it impacts something else too…cost. Cloud backup is a “pay by the drip” service. Whether you’re a fan of David Lee Roth (yes) or of Sammy Hagar (not so much), do you really want to clog up your nightly backup with Van Halen, and then keep paying for it month after month? 

No matter which backup approach you employ, you must deal with some key facts of life. Data has and will continue to grow at an alarming rate, as will the cost to store, manage and protect it - regardless of technologies utilized - unless you understand it and the associated value to the business.

In order to ensure your backups are optimized, from both a functionality and cost perspective, you have to understand how critical the data is to the business. This means knowing what data users are saving to the network. It also means understanding how applications are consuming storage space.

Understanding how your data lives and breathes from day-to-day, month-to-month, and year-to-year will help you be a better consumer of cloud backup services – or for any data protection solution.

"In order to ensure your backups are optimized, from a functionality and cost perspective, you have to understand how critical the data is to the business."

Getting back to the Van Halen example, rifling through your network to uncover music and movies might be a daunting task, but not excluding it from your backup is an all too common mistake that will cost you dearly. Dealing with backup files and database dump files at the beginning of your cloud backup journey (oops - there’s that cliché again) will save you money while reducing the headaches of quickly tracking down growing backup sets…and bills.

2. Not all your data is critical

In the traditional disk and tape backup world, we backed up E-V-E-R-Y-T-H-I-N-G. We bought licenses, media, and more storage than we needed. Data was data, and we made sure we had enough capacity to cover what we have now and hopefully what we’ll need in the future. With that in mind, backup administrators often treated all data the same – (think “better safe than sorry”). It was easy, and it fit the over allocation model to store all data as if it were critical. Files saved by a user, financial data, and the information store for the Exchange server were all stored with equal importance and backed up using the same retention guidelines.

A comprehensive data classification exercise will enable you to gain a better understanding of your business needs and your recovery time objectives for your young versus old data. In the end, you will have aligned the value of your data with the cost of protecting it. If you do it right, business-critical data will be fast and easy to recover and you’ll experience a potentially significant reduction in cost because you will only be paying for one, not multiple copies of backed up data.

"If you do it right, business-critical data will be fast and easy to recover"

Think about it. How many copies of the data you already have? Many organizations have more copies than they realize once you factor in snapshots and data replication solutions deployed for Disaster Recovery (DR) purposes.  So if your backup solution is not your primary DR or restore location, you likely can treat it differently there and save time and money.

Here’s another question for you. How many of your databases haven’t been purged or archived? If the first data entered stays there forever, do you need to keep more than 5-10 most recent copies of a database backed up? After all you’re never going to restore a ‘deleted’ record. You’ll only restore due to a disaster or corrupted database.

"How many of your databases haven’t been purged or archived?"

3. Data Policies are Key

Hopefully I’ve made the case for needing to understand your data so that it can be tiered based on type and use. In order to do that successfully, you need to develop and define internal policies that govern data. From there, you can create retention rules built around the policies governing the data.

So, what kind of data policies will you need? I am so glad you asked. Here are a few data policies that will make a cloud backup deployment smooth.

  1. Regulatory Compliance - Identify all regulatory requirements and the corresponding data that meets these requirements.
  2. Data Classification – Identify and differentiate data by its business purpose. This can include customer versus internal categories as well as public versus proprietary classifications.
  3. Retention Policy – After your data is segmented into categories, determine the minimum amount of time and number of restore points you’ll need for each category.

Don’t underestimate the importance of crafting a good cloud backup retention policy. Your policy will dictate what the restore points look like, so make sure the policy is specific to the data at hand while not being too broad. A poorly designed retention policy leads to two things. Neither of which is a pleasant conversation:

  1. You’re not keeping enough restore points. In other words, your RPO is not in line with the business’ needs. The consequence is you will not be able to restore a file if it is from a date outside the assigned retention policy.
  2. You’re backing up more data than you need, and as I’ve stated, that will unnecessarily increase your cost.

The Economics of Backup

Business and cost justifications for backup solutions of any kind are easy to make, especially cloud backup. It is a great solution for many organizations that are willing to invest the time to understand their data or consult with an expert to help them define cost-effective backup processes. 

You just have to keep in sight that the economics of cloud backup are different, or rather perceived differently. With traditional disk and tape backups, your investments are periodic and significant capital expenditures, combined with some media and license (remember, you’re accustomed to buying more than you need). Increasing costs with a traditional model are not realized on a month-to-month basis, but rather annually, at best, during budget and purchasing cycles.

"The economics of cloud backup are different, or rather perceived differently"

With cloud backup, you only pay for what you use. Your unit charge is the amount of data you store in the cloud, and you are billed every month. They say “perception is reality,” and when you see a monthly expense for protecting your data, your initial reaction may be, “Cloud backup is too expensive!” However, that may not be the case. A true total cost of ownership (TCO) comparison will likely support a cloud backup model, but due to the cost being “top-of-mind” every month and the need to move this expense from CapEx to OpEx requires (can you guess?)  Right - an attitude adjustment (sorry, I couldn’t resist).  But seriously, you will need to get Finance on board, educating them on why you no longer need to spend six and seven figures on hardware arrays and software licenses.

Key Takeaways

Of course the economics are important but understanding your data, treating your data differently based on its use and purpose, and crafting appropriate retention policies will have a positive impact on your initial costs as well as future costs as your data grows.

Another added benefit – a well-optimized cloud backup solution will free your team from mundane backup tasks and allow them to focus on new services delivery.

Don’t want to go it alone? Cloud backup providers have tools to analyze and asses your network so all these questions can be answered and accounted for, and used to define the most efficient and cost-effective backup processes for your company.

If you think cloud backup is worth considering, then my advice is “Go ahead, Jump!”

About the Author

Matthew Brady Daymark DPS specialistMatthew Brady is a Senior Consultant at Daymark Solutions and manages Data Protection Services (DPS).  Matthew specializes in cloud backup infrastructure and focuses on implementing cloud backup solutions for Daymark’s clients. Matthew is certified in VMWare, Cisco, Asigra, and Microsoft.