Möchten Sie diesen Beitrag in Deutsch zu lesen? Lesen Sie die Deutsch-Version hier.
Note: This article also appears on Talkin’ Cloud.
There are a lot of options when it comes to cloud backup and even more philosophies that govern how these cloud backups actually work. Some are easy and user friendly but offer limited recovery options, while others are more complex and offer flexibility and nearly unlimited recovery options. The purpose of this article isn’t to say which is best because there’s no hard and fast answer to that. Instead, we’ll look at the advantages and disadvantages of various options so that you can make informed decisions and lead your clients toward a decision that fits their budgets and particular business requirements.
A quick note on cloud
Before we get too far, note that redundancy is the key word when it comes to backup and recovery. Having your data in only one place is never the best option, which means that if you’re all in on the cloud, you may lose Internet access, which means losing access to everything.
Furthermore, there are situations where the cloud itself goes down or where your data can end up being accidentally deleted by the vendor. If you are all in on the cloud (meaning, all of your servers are running in the cloud), it’s possible to backup systems in the cloud to a local location, depending on how your systems are set up and which services you’re using (one of our MSP partners has an interesting story about how he does cloud to local backups for a client that uses Amazon Web Services).
Whether you’re talking local first, cloud second or cloud first, local second, the essential thing here is to have two backup copies, which is something you’ll notice is missing from a few of the options below.
Archival in the cloud
Archiving can be extremely inexpensive, particularly with services like Amazon Glacier. This is great for data that needs to be saved for long periods and isn’t likely to be accessed. Note that it’s cheap because it can take several hours to access data in Glacier—that’s not ideal for a recovery scenario. Still, it’s a decent option if you have hyper-sensitive stuff that you need “just in case,” but not for keeping business critical information. Remember that often times this archival data is stored in one place, not two.
File and folder in the cloud
There’s a new file and folder backup service seemingly every day. In fact, if you have a Microsoft or Google account, you probably have it by default. The advantages of these are that they’re dead simple to use, often free for quite a lot of storage space, and they’re great for things like collaboration or for accessing files and folders remotely. The disadvantages are three-fold. First is that with anything cloud-based, you lose all access if your Internet connection goes down. Second is that access is an advantage and a disadvantage where security is concerned. You might be creating opportunities for criminals to access sensitive information if it’s stored using one of these services—all it can take is someone losing a smartphone. Third is that you’re limited to recovering files and folders. For some this is fine, but if a quick recovery is what you’re after, this isn’t a good option. You can’t use the files and folders if you have to rebuild a system to use them. File and folder can be an option for desktops, but for a server it won’t get you far.
Image-based backups stored in a regular cloud
This is where data protection really starts to happen. Suppose you take a full image-based backup of a critical server, you keep one backup onsite, and you replicate the other to a cloud. You could replicate to one of the big names used to store files (in this case, your backup is really just another file), or you can use your own data center or co-location facility. This gives you a copy you can use to recover onsite quickly, or if your business is destroyed, a second one that is safe in a data center elsewhere. This means your data is (thankfully) safe, but until you can get back to actually using the data (or the systems that create, use, and modify that data), business isn’t happening. And that leads us to the next one.
Image-based backups stored in a recovery cloud
This option works the same as the one above with a few distinctions. You may wonder: what is a recovery cloud exactly? Well, it’s just what it sounds like: a cloud designed with recovery in mind. Suppose you’re using a regular cloud for backup. If your company had a site-wide disaster and any local systems and data are destroyed, what kind of options do you have to get back to work immediately? You can order new equipment, rebuild systems, and finally download the data you had stored in the cloud, but how long does that take and how much does that kind of downtime cost your business? This varies depending on size of business, but for a small one this costs an average of $8,581 per hour. Now multiply that by 8 hours a day and multiply that by how many days it takes to recover and what do you have? A sunken business.
Now let’s imagine your cloud is built for recovery. Assuming you carefully planned for a site-destroying event, you would be equipped to spin up VMs of equipment you backed up to the cloud in a matter of minutes, and you’d be able to do it from anywhere—basically any laptop or desktop with an Internet connection. That downtime we talked about starts to disappear pretty quickly if you think about it that way—particularly if the company in question gets its money through ecommerce.
Of course, these are just a few scenarios and site-wide disasters aren’t common in every area or for every business. You’re much more likely to run into an issue like hardware failure, user error, malware, or what have you. For situations like those, a recovery cloud is still advantageous because you can keep essential equipment running from the cloud if it goes down at a local site. This means that if one piece of equipment goes down your whole workforce doesn’t go down with it.
There are different costs and advantages associated with various cloud backup options. The real thing to think about is how well each option serves each client, which is often a matter of balancing budget and functionality. Just be sure each client has a backup onsite and offsite, and that any downtime they would potentially have from a failure isn’t more than they can tolerate.
Photo credit: J bizzle via Flickr