Don't Backup? Is it really all or nothing?

[Samantha Morris]

"Backing-up everything is a lazy option which requires little thought from you and your users but it is terribly inefficient and probably is major cause of failures in your back-up environment." -- Martin Glassborow

Interesting proposition, isn't it? I grabbed this from one of his latest blog posts titled "Don't Backup" where he defaults to two backup policies -- Backup everything or nothing at all.

What if we had users tell us what the value of their data was and what they needed backups for? Would this help take some of the onus off backup pros come recovery time?

We need to back up, goes without saying -- but what if we decided to start with nothing and then determine what's actually really important? Or perhaps we need to backup everything. Is it better to be safe than sorry?

[eTegrity]

OMG I can not Stress it enough DIFFERNT DATA HAS DIFFERENT VALUE! and it should be treated as such, "just backup everything" is like everyone is the same, we are not thank God. We reccomend the use of a Lan discovery see http://steveferman.wordpress.com/201...fferent-value/ for a deper explanation but as a business owner I can tell you there are many reasons to understand what data you have and treat it differently.

Steve

[mclarin]

If for no other reason this is a good approach from a sales perspective. Gets people thinking... how does our data get organized? How would I decide if it is important or not?

The time and resources required to prioritize the "backed up vs.not backed up" is an expensive processes itself.

[eTegrity]

Well we use a tool that comes with our product called a Lan Discovery, it wil tell you when the last time a fiel has been accessed, how big it is, how many duplicates on the network and across the enterprise not just on one machine or server, even the amount of data change ratio that happens on a daily basis. If you would like more information please email at steven.ferman@etegrity.com and I will send you an information sheet. This tool is invaluable in our day to day cleint management of thier data and more importantly in the management of thier RTO (Recovery Time Objective) you know what matters the Recovery not the Backup.

Steve

[Samantha Morris]

Hey Steve,

Would you be able to share some of the key content of that information sheet on this thread? I think it might be useful and help move the conversation along.

[mclarin]

To clarify, to spend time considering the value of data by IT departments and backup admins is one thing and obviously needs to be done. But if this question is suggesting to push the responsibility down the organizational hierarchy further than is generally done in the industry, there are soft costs to consider. Such as educating employees where to save to have something backed up. Educating employees on the logistics of backup and how that works. It is not their daily job and thus a foreign concept albeit simple to those that deal with it on a regular basis.

[outofthewoods]

Well we use a tool that comes with our product called a Lan Discovery, it wil tell you when the last time a fiel has been accessed, how big it is, how many duplicates on the network and across the enterprise not just on one machine or server, even the amount of data change ratio that happens on a daily basis. If you would like more information please email at steven.ferman@etegrity.com and I will send you an information sheet. This tool is invaluable in our day to day cleint management of thier data and more importantly in the management of thier RTO (Recovery Time Objective) you know what matters the Recovery not the Backup.

Steve

Steve makes a lot of very good points. It's not about the backup, it's kind of about the recovery, but what it's really about is the RESTORE! With a true data protection solution, you want the data backed up and secure, you also want to know the data can be recovered, and lastly, the data you recover can be restored. Everyone has heard of BUR...let's all update our vernacular to BURR (backup recovery and restore)

[Samantha Morris]

I think we're moving away from the context of the original question, let's take software/solutions/tools out of the equation.

How do we determine what to back up? And can our users help us make that determination. If we move some of the responsibility onto employees, how do we provide them with the information so they have a clear understanding of exactly is being backed up and what is not.

Is this "Don't Backup" policy even realistic? If it is, how can we work on getting there.

[eTegrity]

I would have to agree with Samantha and Out of the woods, but Samantha you are right BUR or BURR have 0 play if you do not understand what data you have and how you need it restored. Below are some things that we get from our Lan Discovery and these are invaluable in determining what you have and how it should be managed so you can meet or beat your RTO what ever that time frame is. Below is information on what the Lan discovery does. I wanted to post the actual flyer but not possible here, hope this helps.


LAN discovery mines critical statistics about the network and the following reports are generated:
♦ Summary shows the network shares that have been scanned.
♦ Share Usage Report shows number of files and share sizes and describes the amount of duplicate files.
♦ Largest Files Report will uncover large files that could be junk (backup files, iso, movies, music etc.).
♦ Largest Duplicates show files like windows service packs that were downloaded but never removed. These files could be
flagged for removal or exclusion to ensure a faster restore.
♦ Ownership is a trending report used to monitor how many
files a user has created.
♦ File Distribution shows the top 10 files on a network,potential files that could be excluded from the backup job or
stored at Tier II. (See Example)
♦ Partition Size shows the selected shares and their utilization.
♦ Access Report displays the age of files and the percentage of data that is older than a week/month/year, helping to decide
which data is best suited for Tier I or Tier II. (See Example)
♦ Growth and Modified Files establish a storage trend to determine end user data growth and help with sizing
bandwidth for daily change ratios.
♦ SQL Server Sizes Reports how large the SQL databases are.
♦ Exchange Server Size Report is unique in that it also tells the total size of .pst’s on the network, allowing you to
establish a policy as to how the .pst’s should be backed up and if they are best suited for Tier I or Tier II.

LAN Discovery empowers you, enhancing your value proposition to the market! Utilizing its information will set you apart as an MSP by being better equipped to help clients avoid needless down time, lost customers, damaged reputations and all other crippling results of an inefficient
data recovery plan.

Steve

[smokey]

Is this "Don't Backup" policy even realistic? If it is, how can we work on getting there.

Yes, this is very realistic for some situations. If a server can be built from scratch with all services in place in less time than it takes to restore, then it does not make sense to do backups. Alternatively, if there a SAN is in use and there is some kind of replication or snapshot capabilities on the SAN then that could be used in place of a backup from within the host. It depends on the situation.