Should Backups and DR be separate Systems?

Backup and DR can be separate systems, but do not have to be. Modern advances in backup and DR technology make it possible to have a single integrated system to satisfy the needs of both, and I can think of no reason to not use such a system.

Serving backup and DR needs with a single system is nothing new, as we were doing this over 20 years ago. When I joined the IT industry in 1993, backup and DR used the same storage target: tape. You backed up your data center to tape every day, you put those tapes in a case, and your off-site vaulting company picked them up and took them away. Best practices were to have a tape copy on-site and off-site; on-site tapes were used for day-to-day operational restores, and off-site tapes were hopefully never used. They were for disaster recovery, and you hoped that day would never come.

This was before the ubiquity of RAID, so the loss of a single disk drive could mean the loss of the operating system, the application, and even all of the data. This is why we became good at what is now called bare-metal restore. We were regularly testing our ability to restore one or more servers just because of small-scale disasters that would happen inside the data center. Then in a real disaster we simply used the same system for DR that we had been using for operational recovery.

Then backups and DR diverged when that same company switched to replication for DR and started using RAID to protect all valuable data. Backups became used only to restore from operator error and malware attacks or the occasional double drive failure. Even if we did test our backups, they had nothing in common with the replicated copies of the data stored off-site. The point is there was nothing magical about having DR and backup needs served by multiple applications; it’s just something that happened as technology changed.

Today it is possible, once again, to have a single system that stores a copy of data on-site and off-site and can serve as both a backup and a disaster recovery system – and I think this is a good thing. It’s less expensive to have a single system serve multiple purposes.  More importantly, It means that no matter where your test your recoveries, you’re testing both the backup and DR systems. Such systems also support automated testing of backups, something which wasn’t really possible back in the day.

Therefore, I think that it is perfectly fine to serve backup and recovery needs from a single system. In fact, it may actually have a higher level of data protection than doing the opposite.

Watch On Demand

W. Curtis Preston (aka Mr. Backup) is an expert in backup & recovery systems; a space he has been working in since 1993. He has written three books on the subject, Backup & Recovery, Using SANs and NAS, and Unix Backup & Recovery. Mr. Preston is a writer and has spoken at hundreds of seminars and conferences around the world. Preston’s mission is to arm today’s IT managers with truly unbiased information about today’s storage industry and its products.

Tagged with: , , , , , , , ,
Posted in Blog
One comment on “Should Backups and DR be separate Systems?
  1. Tim Wessels says:

    Well, vendors of object-based storage(OBS) software promoted backup/restore and disaster recovery as foundation use cases for object storage. Having been actively learning about object-based storage and observing the relatively small group of vendors in this space since 2012.

    The “startup” vendors in this market include Amplidata (HGST), Cleversafe (IBM), Caringo, Ceph (Red Hat), Cloudian, Scality, and SwiftStack. Amplidata raised $44M before it was acquired by HGST. Cleversafe received approx. $128M in total funding before being acquired by IBM for $1.3B. Caringo, which is largely self-funded, recently announced a series B funding round of $8.8M. Ceph (InkTank) was acquired by Red Hat for $175M several years ago. Cloudian just announced a series D funding round of $41M. SwiftStack is a newer entrant and has raised $23.6M in funding. Scality raised $57M in a series D funding round just over a year ago. The age of these “startups” ranges from just a few years to over ten years.

    My question is with so much attention given to object-based storage, why do most of them have a customer base of fewer than 100 customers? Early funding rounds usually are applied to development and later rounds are usually applied to marketing and sales, but with something has “hot” as object storage emerging in an era of unprecedented data growth, why don’t they each have many hundreds or thousands of customers?

    I think a partial answer is most OBS vendors have been conditioned to think scale-out is all that matters. Mr. Lecat from Scality has commented that he thinks the market for PB scale OBS deployments worldwide is approx. 20K customers. That’s a very small number. But suppose the problem is really how do you scale down to address a market where hundreds of thousands or millions of customers need tens or hundreds of TBs? Cleversafe (IBM) says their initial comfort zone for new customers is 500TB. Amplidata (HGST) is looking at 500TB to PB+ for starters. Cloudian can run starting with 10TB of usable storage. Scale down can be done by some OBS vendors but they don’t really talk about it much.

    My point is that OBS vendors supporting backup/restore and disaster recovery use cases for OBS storage may be able to serve hundreds of thouands and possibly millions of customers if they can easily and affordably scale down to meet the requirements for these use cases. Cloudian can do this today, so it will be interesting to see what they have to say in your upcoming presentation.

Comments are closed.

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 22,227 other followers

Blog Stats
%d bloggers like this: