I was on a course last week taking a trip down memory lane and re-learning all I knew about TSM; life has strange twists and turns but it made me think about back, recovery and tape.
Every year we have the conversation about tape and whether tape still has a place and I think I am now personally comfortable with answering 'Tape should no-longer have a place in backup and recovery!'
What has really changed my mind is the increasing growth of cloud-storage options which allow bulk-storage at a reasonable cost off-site; until recently, tape fulfilled this role but I think that we can do this now all in the cloud. I can think of no reason for most small businesses to do anything else; products like Mozy can support them and there is no longer any excuse for not having a robust back-up regime.
And the traditional back-up tools will need to start including options to back-up and restore from the Cloud.
Maybe I’m too old fashioned, but I’m still not comfortable relying on someone else’s continued existence for mine.
What sorts of tools are available for migrating data from one cloud to another? It’s pretty certain that cloud technology will exist for some time to come, but the odds of any one provider offering continued, uninterrupted service is somewhat less likely.
Also, I’m very curious about the insurance implications of this. I understand that a place like Iron Mountain has insurance to cover loss of data. How does the cloud compare? Who is “assuming the risk” at that point?
Matt, what if your cloud provider was Iron Mountain? With all the insurance to cover loss of data in place? Obviously you need to negotiate with your supplier and get the correct SLAs and legal protection in place.
Having your back-ups off-sited into the cloud also means that it is a lot easier to verify their integrity, no more generating pick-lists to recall tapes, load into silo and then test restore.
Migration, have you ever had to a do mass tape migration? I’ve done it once, it was a pain and of course we drift into the difference between long-term archive and back-up. Your back-up software should be able to migrate from one cloud to another; it is important that you are not tied to a single cloud provider.
Standards would be nice but actually I am not convinced that that they are entirely necessary, I can see all back-up/restore tools supporting the major cloud technologies in the future.
The one place where I think the public cloud will struggle for the time being is in the area of an active archive where there is a huge through-put requirement.
Martin
Someone senior (who now blogs, and it was not Barry) at EMC told me nearly 10 years ago that tape was dead. My answer to them, having just joined EMC from STK, is that tape was not dead, and won’t be for some time. I said the use of tape is going to continue to change, and move even further down the storage stack / pyramid.
I believe tape still has a place, though more and more disk based solutions will replace traditional tape solutions, and as you rightly say the growing cloud options are further expanding the alternative to tape.
From a personal perspective for several years, I have been using removalable disk drives to backup my home PC, and offsitting them to the mother-in-law ! But now I have just started to use Mozy, though I do need a faster broadband uplink.
As you say Martin there are many factors that need to be understood, to determine the best BC/DR option(s), and I agree physical tape is becoming less and less part of BC/DR solutions.
In my experience traditionally the cost of bandwidth, has been the biggest cost hurdle, though this continues to fall, along with the cost of disk storage, so making remote online backup solutions a viable cost effective solution for businesses, big and small, and with the cloud based solutions coming to market, customers need to consider these. Yes there are consideration, as with any solutions.
Martin,
For SMBs I agree with you 100% that the “cloud” or online backups essentially eliminates tapes for backup. For larger enterprises though the story is quite different.
Once you get into the space of measuring your daily backups in multiple terabytes, tape is still a very attractive solution for handling at least the off-site copies of data.
Until bandwidth is ubiquitous on a global basis and is a comparable cost per TB as tape, I believe tape will still be with us and a viable alternative.
Lets not kill it quite yet but use it smartly!
Bandwidth is indeed still a problem, however once you begin to do the entire TCO, the costs are closer than you may think.
Tape handling, offsiting tape, tape media, periodic testing, expiring tapes, tape migration. All of these add cost to the base cost of tape.
For example, someone decides to reduce a retention period on a back-up; if your medium is disk, the work is minimal; if it sits on tape, you either re-call the tapes and reload them into the library for re-use and we are talking about hundreds of tapes in some cases or you bin them. Neither which are especially attractive options.
Or an auditor wants access to your back-ups for data discovery; if your back-ups are on-line and in the cloud, you can give them access assuming the security model works. No more recalling tapes, copying tapes etc.
Your TCO will be different from mine but it’s worth another look.
Martin is right on with the TCO comment. I recently compared all-in TCO for long-term tape to long-term cloud archiving (golly, I wonder why?) and we found it was a positive ROI to move to the cloud.
This is not to say that disk or cloud storage is cheaper than tape. It says that backup to cloud is cheaper all-in than backup to tape with off-site storage. So if the problem is protecting data off-site for long periods and keeping the solution going for the forseeable future, cloud is cheaper. If the problem is writing some data on a tape and sticking it on a shelf, then tape will always be cheaper. But I think that NO ONE has that problem!
I agree with you Stephen. And while this isn’t the subject of the post, I personally think a lot of people overestimate the security and safety of tape. In a recent post I tell a story about a letter my father received from his credit card company, Chase, in which they admit that they actually lost one of the tapes, which contained names, addresses, and social security information of many customers. Insurance isn’t going help him out of that situation–it doesn’t even really help Chase! Something to keep in mind, in any case.
http://onlinestorageoptimization.com/index.php/chase-tape-fail/
Sunshine, I think that we are all agreed on one thing; before your data leaves your site, it needs to be encrypted whether to the cloud or physically leaving the building. Data-loss is far too common these days!
The unknown performance of restores as a condition of bandwidth availability following a large scale regional disater, such as an Earthquake, typhoon, etc. will keep tape viable for a long time – at least until massive data suckage from the backup site can be proved.
Firstly don’t build your primary data-centre in an area prone to earthquakes, flooding, typhoons, wars etc! It’s a stupid idea and you don’t really need close proximity to your data centre these days.
Secondly, try and build your secondary far enough away from your primary so that any event which affects both, you don’t care about any more because you are more than likely toast!
I used to believe that prior to any DR practise that you should randomly go up to people and tell them that they are dead or perhaps their loved ones are dead. I suspect that would be taking things too far but I did try to insist that any DR/BC plan should have a section in how to handle such a situation.
If you are in the situation where you are recovering business critical apps from tape; those apps which you really truly need to keep your business going, I suspect you have some more fundamental issues with your infrastructure architecture. Prison-time may actually be in your future if you are responsible.
Hmm, there should be a smiley somewhere in the above comment, add your own where you think they should go.
For TCO, do not forgot about time you need to manage all Tape, disk compare to let professional company do it with cloud storage.
I don’t think it will be too long before tape is eliminated by cloud computing. Companies already offer low cost online archiving for Inactive data such as Iron Mountain’s Virtual File Store. This is a way of storing redundant data in the cloud at a cost competetive to that of tape storage. I notice a few comments surrounding the reliability of cloud storage, Surely it has to be a more reliable way of backing up and restoring than tape? The bottom line is tapes fail (tried to restore one last week with no joy)…. If your data is stored in a data centre that is replicated to another.. surely this is as resilient as you can get. With regards to whether the big boys such as Iron Mountain and EMC being around in a few years.. They both have over 3 Billion dollars or recurring revenue through its existing contracts. Tape is on its way out out in my humble opinion.
One thing this has shown, people are still interested in tape and it maybe how you replace it but it still appears to generate a lot of interest!!
Martin, your backup site doesn’t have to be in the same neighborhood as the calamity to have bandwidth problems. For example, http://www.nytimes.com/1988/05/26/business/phone-system-feared-vulnerable-to-wider-disruptions-of-service.html?sec=&spon=&pagewanted=all
Many will feel that putting their backup data in the cloud will expose them to the unexpected shortcomings and budget-driven decisions of other companies that might not be driven by the same set of priorities.