Mike Daniels, Product Manager

In Part I of this blog entry, I talked about maximizing your data protection budget by considering consolidation and looking at disk-to-disk (D2D) or disk-to-disk-to-tape (D2D2T) solutions. In this article, we’ll look at deduplication and some budget savers versus budget busters.

Enter deduplication. When you start deduplicating data it can drastically decrease the budget requirement for disk due to reduced space needed. It reduces storage by eliminating redundant data. One unique instance of the data is retained, and redundant data is replaced with a pointer to the original unique data. So, less space, less money.

Further savings with disk is recognized with significantly faster recovery times. Remember cassette tapes and how cool it was when CDs entered the market?

If you wanted to hear song number three on your tape, you needed to hit

fast-forward and wait. When CDs showed up, you could enter any random track and get your song near-instantaneously. I don’t know any IT managers who don’t want to deliver this functionality for data restore to their teams. So, not only are you buying less disk with deduplication, but you speed up recovery producing further cost-savings.

Another disk benefit is reliability. There is no argument that disk is more reliable than tape. Anyone who uses tape will tell you it’s fragile. Humidity, temperature, number of uses will all negatively impact your chance of getting your data back. Are you confident you have proper guidelines regarding tape storage at your organization? What happens if the person whose job it is to swap tapes and take them off-site, leaves them in the car to run a quick errand, in the middle of summer, and you happen to be located in Arizona? Basically that tape is not usable. But you won’t know until you have to restore something. That’s an extreme example and unlikely that any IT manager is advising his team to manage their tapes in this fashion, but I’m sure you’ve heard of or experienced yourselves numerous accounts of data loss when recovering from tape.

I’m not saying you do away with tape backup. It’s a good solution for long-term storage, but it certainly is not cost-effective when it comes to reliable, fast recovery. If you can get the correct business unit managers supporting a disk-based purchase initiative, you can easily show tremendous cost-savings associated with productivity and reliability to your CFO.

I’ll leave you with one last cost-savings tip, choose one vendor with a suite of products to support your data protection requirements. You don’t want multiple finger-pointing vendors when you have questions, and it makes life much easier for your team to learn and stay trained on the applications.

Consider a couple of things when evaluating vendors: 1) how easy are the solutions to administer, 2) how responsive is support. I had an experience with a customer who was using a competitive product. While I was on site, he called the other vendor’s support to help resolve some issues. Meanwhile, he allowed us to do a proof of concept with our product. We installed, configured, and completed a backup before our competitor’s support team had even answered the customer’s call.

Make sure you understand every aspect of a vendor’s support program before you enter into a contract. This is overlooked far too often, in my experience, until you actually need support. This oversight can unexpectedly impact your budget and increase the overall cost of the product.

So here’s the thing: it’s not difficult to find cost savings when it comes to adequate and proper data protection. Just remember…

Mike Daniels, Product Manager

How do you protect an enterprise environment without an enterprise budget? I see this problem so frequently in customer data centers, trying to adequately protect data in a budget constrained environment.

Don’t despair. There are several things your team can do to maximize your budget.

First, consider consolidation. Do different departments or branch offices have their own protection solutions? Maybe some are on Mac, Windows, or Linux. Each likely has critical some critical data and applications that need special attention. If you haven’t already, consider consolidating requirements into a central location with heterogeneous protection. By centralizing, you recognize savings on economies of scale – one large environment versus small islands. You’ll have one maintenance cost, perhaps price breaks on standardized applications addressing multiple departments’ needs rather than each sourcing its own solution.

Many large companies have been doing charge backs for a long time. Why not you too? Departments were going to pay for the protection anyway, and now you could actually end up saving them money by charging a smaller fee than their previous individual payments. Not only is it an overall cost-savings measure, it addresses what should be your top concern: data being protected properly by the experts who really know how to do it.

Aside from consolidation, ensuring fast, reliable restores is one of your best budget-savers. One way to do this is to look at a disk-to-disk (D2D) or disk-to-disk-to-tape (D2D2T) solution. I know what those of you who haven’t already done this are thinking… new hardware, new applications, how is this possibly a cost-savings measure? Stay with me, because, ultimately, this could save you a lot of money.

In the past, disk-based backup has been a niche feature, because backing up to disk could take anywhere from ten to fifty times the amount of space versus server space. This was a big hindrance for people not moving to disk-based solutions faster.

In Part II of this blog entry, we will examine deduplication as it relates to maximizing your data protection budget, and we will discuss some budget savers versus budget busters.

Gary Parker, Sr. Product Marketing Manager

Data protection is no easy feat in the world of higher education. Recognizing retention periods for varying sets of data, determining required recovery times for each set, and mitigating data loss can be daunting tasks. Not to mention, you are often protecting a wide array of data while supporting thousands of users that range from those who are just sending personal emails to those conducting the most delicate of research projects And, when something goes wrong, it’s likely you’ll be inundated with a flood of complaints.

But instead of focusing on a reactive situation, let’s turn our attention to the objective of complaint prevention.

The very simple steps of a thorough data protection strategy are so often overlooked. Where does an IT professional supporting a higher education institution really need to start? It’s more than just understanding your data sets. Your team should really put the following in writing for each set and adhere to it:

1. Retention period
2. Recovery time objectives
3. Amount of acceptable data loss

Let’s take a quick look at retention periods for a typical academic computing environment. Thinking this through will help you get data to the correct, most cost-effective storage medium and determine your method for storing it on that medium. You will likely have at least some of the following data sets. As an example, I asked BakBone customer, Martin Frankhouse at University of Detroit Mercy, for an overview of his typical retention periods, knowing that these definitely vary from institution to institution.

1. Student Activities – 90 days
2. Student email – 90 days
3. Course content – 3 years
4. Catalogs – 10 years plus
5. Grades – Decades
6. Research Projects – Potentially decades

Recovery time objectives could actually run on the same scale as your retention periods. For example, student email may have a short retention period and likely have a short recovery time objective. It may not be important enough for long term storage, but you better be able to get it back up and running fast in the event of downtime.

On the other hand, you may need to archive grades for long periods of time and may also be able to tolerate longer recovery times should something happen to your database. All this being said, your recovery time objectives will necessitate different tools for different data sets; for example, NetVault: FASTRecover can bring up Exchange data in as few as 30 seconds while NetVault: Backup can manage the archiving of data to tape. And, regardless of your backup application, while tape is a relatively inexpensive storage medium, it certainly takes longer than recovering from an online disk solution.

Finally, let’s take a brief look at minimizing data loss. One area that can help you to sleep well at night is knowing you have high availability to critical data and applications in the event of a minor human error to a major campus disaster. As I said earlier, understanding acceptable data loss for each data set is an important step in creating your data protection strategy. This doesn’t necessarily mean that any data loss is acceptable. It’s up to the responsible parties at your institution to have a very solid understanding of what exactly is acceptable and to ensure that those accountable for mitigating the loss have policies and protection in place to meet your recovery point objectives (RPOs). For example, losing a certain amount of student email is unlikely to bring your institution to its knees; however, losing critical research data might put a critical grant in jeopardy.

So, what’s the real essence of top notch data protection for the higher education environment? Understanding that not all data can be treated the same and ensuring that your solutions can most cost-effectively achieve acceptable retention periods, absolutely meet required recovery times, and effectively restore data to specific points in time. And, let’s not forget one of our important objectives from earlier: accomplish complaint prevention!

Ray Lucchesi, President of Silverton Consulting

4 min 9 sec

In this Experts & Insights podcast, Matt Law, senior PMM at BakBone, talks to Storage Expert Ray Lucchesi of Silverton Consulting about recovery time objectives (RTOs) and recovery point objectives (RPOs). Ray discusses how the criticality of business activities associated with an application and its data determines RPO and RTO requirements.