Your first step changes everything. It can even change the hemisphere you find yourself in.
The same applies to your data’s journey into the cloud because the first few decisions you need to make affect every thing else related to command and control. So which way will you go?
- Will your first step be to the East where all of your backups converge as point in time snapshots that are monitored by an administrator; or to the West where your data decisions are made by users and files are constantly synchronized (copied) to multiple devices ?
- As for that second step, will your data go South into a vendor’s pre-built data stores; or North to data store(s) that you personally create, move, and delete with the cloud vendor(s) you choose ?
Before continuing it is important to note that every point of the Data Positioning Compass (DPC) leads to a viable destination. At issue is not the decision itself, but the alignment of those decisions with the desired outcomes.
For instance, if you want your employees to decide which files go into the cloud and you are comfortable with a copy of those files distributed to every participant’s device(s) then a sync process (West) is a good choice.
Just keep in mind that sync products started out as workgroup productivity enhancers and you should reflect on using them as your organization’s sole backup because the files are written to and deleted with the results mirrored in near real time.
Backup is different than sync. Backup data is no longer “in motion” once it is moved into an authoritative repository with controlled ingress and egress. The term backup also implies point in time snapshots for files, and as important, groups of files that are of a like generation.
If that is what you want, then your first step should be to go East with software that permits your administrator(s) to decide what is backed up and, critically, how long it is backed up. Most organizations will come to a point where they do NOT want to just keep everything forever and will want to impose policies and control distribution and access.
Whether you have decided to sync or backup, the next step is deciding whether you want to go North and create your own cloud data store or go South and use the data store bundled with your software.
The least expensive combination – and cost is a very important factor – is usually going to be in the South West quadrant with sync software and the data store provided by the same vendor. If you are ok automatically copying all your files to a mirror site but concerned about the implications of being in a “cloud crowd” then you might prefer to go into the North West quadrant and run sync software into a cloud data store that you control. That way there are no other organizations in your storage pool so you will not be affected by their activities.
Backup can also be into a software and data store combination from a single vendor in the South East quadrant, or you can pick the software and choose where you want to put your cloud data store(s) in the North East quadrant.
The primary benefit of using a single vendor for both the backup software and the data store is at the outset when you are just getting started because they are tightly coupled.
Separating the software and the data store results in a different set of benefits, primarily in the areas of flexibility and integration with the rest of your cloud operations. If you are going to be deploying production systems into your selected cloud(s) then it definitely makes sense to incorporate backup “Nodes” into your cloud computing pool so you can programmatically distribute data to your systems and the content networks.
Of course there are many other benefit and feature trade-offs between these options but this post is simply about identifying those first two critical steps and recognizing that they result in very different destinations.
So, to recap, the East-West decision is about the contents of the data store and how files are transmitted whereas the North-South decision is about the location and nature of the data store itself.
When viewed from a command and control standpoint; decoupling the method (software) from the repository yields higher command while structured backups of specific data sets governed by policies grant more control than simple synchronization.
These benefits are valued differently by each organization. The key is to understand that there is wide variation between “cloud backup” solutions and to start off in the best direction to meet all of your objectives.
Having said that, there are some real differences between the real world and the virtual world. Most notably, in the latter you can be two places at one time. This means that if your organization is deriving value from sync products that they themselves can be backed using a structured method. More on that next time.