I’m finally getting back to this train of thought. To recap…
How long can your business survive without your data? How long can your business survive if your people can’t access your business via the internet?
What kinds of data have to be “always on” for your customers and your business? How will you make sure that those data and applications are highly available? What has to be done to your computing, storage and networking infrastructure to keep your business running? Back in the twentieth century, the answers were quite a bit different, and if you didn’t have a million dollar budget, you couldn’t really make sure that your business was accessible even if the power went out at corporate headquarters.
So, what about the Cloud? Can you do disaster recovery to the Cloud, so that if your datacenter is cut off, you can failover to a virtual infrastructure, if only for a short while? Is this possible? Is it affordable? If it is, and you can do it and it works, then…why are you in the business of building and running a datacenter at your headquarters?
“Disaster Recovery In The Cloud”!
Like many things “Cloud”, this is a concept and not a product. The “productization” of Cloud DR is proceeding apace, though. Here are some of the things to keep in mind as you try to figure out a DR strategy in the 21st century.
- Cloud Computing platforms, Infrastructure as a Service (Iaas) is the new Managed Hosting for data center operations. This technology allows the data center operator economies of scale for server and networking deployment for customers - a few guys are provisioning for customers on demand without running around slapping servers and switches into racks.
- Cloud Computing Portals, done right, allow the customer to provision their own servers, based on their own builds (hardening,patches, security, apps), without requiring datacenter staff to actually be “remote hands”.
- Cloud Computing provides network, computing and storage ON DEMAND, meaning not only “standing up” servers, but also adding capacity in all three respects.
- Cloud Computing cannot host all critical applications, depending on the platform: AS400, Solaris, AIX and HPUX all do not fit into any Cloud platform that I know of, although they all have their own virtualization strategy (zones, LPARs, workloads, etc.).
- Multi-tenant storage and computing platforms raise security concerns.
So where does this leave the cloud as a DR or even a primary datacenter “location”?
You, the business representative, have to figure out what applications are suitable for the Cloud computing platform, or Managed Hosting/Collocation infrastructure. Nothing has changed in that regard. The only thing that has changed is the method and speed of provisioning certain aspects of your business technology platform.
A true Business Continuity solution will take into account much more than a DR strategy (typically confined to recovering the IT infrastructure and applications). Your “DR In The Cloud” strategy will need to confine itself to the Windows and Linux applications and data that are supported in the Cloud platform of your choice. Typically, this includes the basic IT infrastructure and many business applications, so this may work just fine for you: AD, Exchange, SQL (in most cases), Web and middle-ware, file serving. As was pointed out to me at a recent speaking engagement, enterprises have much more complex requirements (e.g. mainframe, Exadata, etc.) and more resources, so they are excused from class now….
Ok, now you commercial and SMB companies, now that those big guys have left the room, this solution is FOR YOU. As I said before, major replication and HA technologies have been, and still are, out of reach and scope for many companies, and this new kind of solution is for you – not Them.
How can you recover your servers in the cloud? There are many technologies that are now within reach, including host and virtual machine replication (not storage array replication, $$$).
Next week, a run-down of those applications and tools that enable you to achieve a recovery time objective of half a day or less, and a data loss of about the same. I promise.
