Cloud backup differs from tape backup in many ways, with an inherent ability to immediately back up customers’ data off-site and place the responsibility of managing the backup process on the provider instead of the end user. These differences translate to advantages storage resellers looking to sell cloud backup services can capitalize on.
In this interview, Rachel Dines, senior analyst at Forrester Research, discusses the pros and cons of cloud backup compared with tape backup. Find out which type of customers are good candidates for cloud backups, what concerns exist around disaster recovery, and how resellers can help their customers overcome reticence to move to the cloud for backup.
Download for later:
- Internet Explorer: Right Click > Save Target As
- Firefox: Right Click > Save Link As
SearchStorageChannel.com: How does cloud backup differ from tape backup in terms of its main pros and cons for IT organizations?
Dines: There’s a big difference between tape backup and cloud backup. First of all, when you're dealing with cloud backup, you’re immediately getting your backups off-site, which is a really critical part of a lot of companies’ disaster recovery plans. When you’re backing up to tape, you of course have to back up locally to tape, then take the tapes out, encrypt them, ship them, [and] get them to another location. So the off-siting, in terms of tape versus cloud, is a very, very big difference.
The other thing is, in a cloud backup world, usually you have a provider that’s actually managing the process of running the backups, scheduling the backups, making sure that everything is completing properly, error checking, things like that. So a lot of the burden of the day-to-day task of managing the backups in the cloud world is offset from an end user.
When you’re looking at tape, it depends on the scenario, but oftentimes in the tape backup world, end users will have a little bit more burden and responsibility set on them. Beyond that, tape is notoriously slow. When you have to do a restore from tape, it can take quite a bit of time, especially if that tape has been already taken off-site and you need to first recall it. [You must] wait a day or two for the tape to … be shipped back, and then start restoring the data, which is fairly slow, versus restoring from the cloud -- which, depending on the amount of data, is usually still faster than tape. [When] restoring from the cloud, you have to restore over the network, but because it’s from disk, it’s still significantly faster.
SearchStorageChannel.com: For storage resellers looking to sell cloud services to customers, what kinds of IT organizations are good candidates for cloud backup?
Dines: There are definitely a few different types of IT organizations that I would target for cloud backup services. Organizations that just don’t have the IT staff to manage the backup can be a really good potential customer for saying, “Hey, send this to the cloud.” Organizations that have notoriously and historically struggled with tape [are also good candidates]. There is a lot of anti-tape sentiment because things go wrong with tape backups -- the tapes are not recoverable, physically handling and managing the tapes can be very tricky -- so organizations that haven’t yet moved much to disk are actually great candidates for cloud backup because there’s such a big difference from tape versus cloud.
Beyond that, for smaller organizations, it can be a great use case for companies that have a lot trouble, a lot of headache over their backups -- which … is a lot of companies. As you get into larger organizations, thinking more on the midmarket and the small and [medium-sized] enterprise space, a lot of organizations that are good to target would be the ones that have branch offices and are running local tape, or even local disk backups at the branches. I hear from a lot of the companies I work with that backing up and protecting data at branches is a huge headache, and going to the cloud with that is actually a really good use case.
As you move up-market, looking for large organizations, targeting islands of storage … at the branches and remote offices, is a really good way to go about this.
SearchStorageChannel.com: What are the concerns around data recovery with cloud storage?
Dines: One of the big concerns I always hear is around having enough bandwidth to recover large amounts of data. So if you had a very large outage or even maybe lost a whole site, and you needed to recover a very big data set from the cloud, given the state of networks and bandwidth today, it would be very difficult and very time-consuming to do a large restore from the cloud. Usually what I recommend is VARs can offer the opportunity to do a quick ship of an appliance to customers that need to restore a lot of data. So if you get a call, and a customer says, “Hey, I need to restore 5 TB right now,” the fastest way to do that is to load it up onto removable media or some kind of backup appliance and ship it to them overnight. That is usually the best approach for large amounts of data.
In terms of data recovery, [regarding whether the data can become corrupted or unusable], I’ve not seen many issues in that space. That’s actually much more of a concern in the tape arena. In cloud, having usable data -- data that is not corrupted, and [there are] consistent points in time that you can restore from -- cloud is actually very good for that. The main issue of course is trying to squeeze a large amount of data over a fairly thin-pipe bandwidth.
SearchStorageChannel.com: How can resellers overcome organizations’ reluctance in moving to cloud backup?
Dines: A lot of the reluctance I see from end users not wanting to move to cloud backup is just inertia in this space -- people getting attached to the status quo and not necessarily seeing anything wrong with what they’re doing. [Resellers could point] out that they could be doing backups much more efficiently, they could be freeing up resources to have people spend time on more strategic initiatives in the organization. There’s also oftentimes a lot of concern over security and over bandwidth. First of all, [explain] about how most cloud backup solutions work with incremental forever [and] deduplication -- they’re actually extremely WAN-efficient on a day-to-day backup basis.
Most customers assume [that] if [they’re] protecting 1 TB, [they’re] sending 1 TB every day. That’s absolutely not true. That’s not really the way any cloud backups work. They’ve got a lot of differentials and different types of compression algorithms, so you’re sending very small amounts of data. So explaining the bandwidth-efficient part, and then whatever you’re doing from a security perspective to make sure that their data is secure and separate from all your other customers, is going to help ease fears around the security and compliance piece.
I do think that a lot of the reluctance today has been from people not having time to think through fully this potential change. Once they take a look at their backup environments and say, “Hey, this is very much out of date,” and look at new potential options, it’s not as hard of a sell as you might think. A lot of companies really do want to move off of tape and move away from legacy backup solutions, but they haven’t had time to research all the options yet.
This was first published in May 2012