Back up your systems with 138 USB sticks

I get solicited daily from companies promising me the holy grail of data protection. Without fail each one of them touts some whiz-bang technology that promises me I will be able to sleep well at night knowing that my data can be safely recovered should the server fairies decide to go on strike late one evening. leaving my systems without their needed supply of magic server fairy dust. You know it will happen super early Monday morning too. Nothing says, “Hello Monday!” like waking up to an inbox full of emails about servers that have gone down hard before you have had a chance to partake of your blessed bean juice. Darn those server fairies! 

There are so many compelling products on the market that it is really tempting to let price creep up to the top of the decision analysis “Most Important” list. After all, saving money will make the CFO happy. Everyone wants a happy CFO. A happy CFO means better holiday parties, bigger raises, and better coffee in the break room. Hey, maybe price does deserve to be at the top of that list after all.

I’m going to let you in on a little secret, just between you and me. I’ve been at this game a while now and even though a 12-year-old could school me at Fortnite, I still have some geek cred to my name. I’ve been around the desktop a few times if you know what I mean. A lot has changed since computers started to become more common in the workplace, a lot of things except one. Do you know what that one thing is? Its OK if you don’t because I’m going to do you a BIG favor and pass on the single irrefutable truth of computing that I have learned over the course of my life. That truth is, that eventually, a computer (be it a laptop, a desktop, a server or any other modern computing device) is going to die. That’s right, you read that correctly. Read it again, we will wait for you to get over the shock.

Yes, dear reader, the device that you are reading this article on will, some day, quit working. It will happen when you least expect it. All those family photos, gone. All your spreadsheets, calculations, sophisticated modeling showing how much money the company could save by switching from paper towels to a blow-dryer-based system in the restrooms, gone. The database with all your client records, gone. Your company’s financial data, gone. All of it is gone, and it’s not coming back unless you were smart enough to convince your CFO to purchase the whiz-bang backup system from that ad you saw while you were watching Fail Army during the Monday morning meeting. You know, that backup system that was picked because it was the lowest price.

Folks, when your magic server fairies decide to go on strike, and you run out of magic server fairy dust, the last thing you want to do is rely on a decision, who’s most weighted factor was simply price. Imagine planning to ask the love of your life to marry you at dinner tonight. This could be one of the most important nights of your life. If all goes well, you live happily ever after. If it goes wrong... well, what is understood need not be discussed if you catch my drift. Would you get reservations at Ruth's Chris Steakhouse, and dine on prime rib and fine wine and propose over sharing a slice of cheesecake? Or would you opt for the two cheeseburger value meal from McDonald’s and pop the question over an ice cream cone? Yeah, yeah, with true love it wouldn't matter if its Kobe beef or chicken nuggets, I know, my wife makes me watch those movies too. My point is would you feel more secure knowing your data was backed up by the equivalent of a value menu cheeseburger based marriage proposal?

Now I know someone reading this is currently thinking “This guy doesn’t know what he’s talking about. I backup all my data to a Hooli box that utilizes the latest inside-out compression algorithm, and it’s the cheapest system ever.” I’ll be honest with you. Sometimes the cheapest option is the right option. You heard me correctly. Sometimes the best way to backup data is with the cheapest option. A few of you might be sitting there, a little confused. Don’t be, I’m about explain. 

There is nearly an unlimited amount of ways to backup your data. Truth is, at its most rudimentary form a backup is simply a copy of a file. Ideally that copy would be stored somewhere else, in case ninja’s break into your office amidst a Kill-Bill style battle royal and destroy everything in sight, but it doesn’t have to be. Under some circumstances a copy may be all you need, and a copy can be as cheap as writing it to a USB flash drive. In fact, you could copy your company data to 138 flash drives to ensure that you have plenty of copies in case one gets lost, or corrupted, and it would still be cheaper than some of the systems out there offering the latest and greatest solid state disk arrays with super-cloud-lightning-fire-shock-and-awe-crypto-locker-proof-active-passive-image-based-recovery doodads.

Back to reality for a minute. Data protection is important. Price should not be the primary motivator when determining how to backup data, but as I said before, the cheapest option could be the best option. The trick is in being able to figure out IF it’s the best option for your needs, and to do that, you need to remove price from that top spot on the “Most Important” list and replace it with a Risk Profile. “What is a risk profile” you ask? Great question.

A risk profile consists of an understanding of your objectives, not if, but when those server fairies go on strike. In plain English, how much pain can you and your company endure before your clients start to get angry and look elsewhere? Some companies can endure longer amounts of downtime, some can’t. Determining how long you can be down before it starts to cost you money is critical to ensuring that you get the right product for the job.

As a consultant I was brought in to help recover a small manufacturing company who lost their main file system. Their existing provider had spent a week trying to get things recovered from their backup tapes already before I got the call. As frustrated as they were at being down a week, the company hadn’t really suffered yet because they still had access to email, and as it happened, their production facility didn’t need real time access to their shop floor system due to complicated manufacturing that took days and sometimes weeks between database entries. A company like this may not be interested in paying for a product that gives them near instant recovery capabilities because even at a week of downtime, their risk profile doesn’t demand it.

I have also worked with another company who has sub-fifteen-minute recovery obligations to its clients. If their systems are down for more than 15 minutes, heads are going to roll. Simply put, their risk profile is vastly different than the previous companies and is going to require a more substantial investment from the CFO.

Earlier I mentioned that a happy CFO means bigger bonuses, better holiday parties, and better coffee, right? Well at least in an ideal world it would, but what if that CFO is presiding over a company who has sub-fifteen-minute client obligations and you give him a backup/disaster recovery system that is going to take a week to get all the data off of a tape or from that really cool and cheap, unlimited storage, cloud vendor? Is he/she going to be happy when the clients call and cancel their contracts? At that point, how much money you saved on that 138 USB drive backup solution won’t matter because all the CFO is going to be concerned with is how much money you have cost the company by losing clients.

In both of the preceding examples, there are multiple ways to back up the data and provide for system recovery. The risk profile helps you choose the correct product within your budget, but it also shows you if your budget is not in line with what you actually need. Your risk profile should take into consideration not just how long you can go before you incur financial losses, but how much down time is going to cost you as well. Put a dollar amount to it. It’s one thing to say that we will start losing money after 4 hours of down time, it’s quite another to show how much money a company is losing due to an idle workforce and potential contractual penalties.  

Your risk profile can also help you determine if you need to pay for a product that can replicate the data offsite so that you can recover in the cloud, like if you had a fire or flood, or if you just need to have an extra copy that could be downloaded if need be. Maybe your risk profile will show that you don't need any cloud-based backup storage at all. The point is, if you don't develop a risk profile and understand your tolerance for downtime you could be overpaying for a backup solution or under buying and setting yourself up for lots of pain.

It’s important to present the solution using language that the CFO will respond to. CFOs tend to be pretty analytical in nature. Include a spreadsheet and graph in your risk profile to show just how much it is going to cost the company as an outage progresses. If the CFO doesn’t respond positivly to your risk profile, then go back to the drawing board and try to rewrite it. This time try and add some pretty pictures, or maybe a Dilbert cartoon about IT stuff. If that doesn’t work, make sure you get a label maker so you can clearly label all those USB sticks.