The Reality of Cloud Adoption in 2016

Posted on

by James Keating III, Business Technology Architect, Evolving Solutions

Since I have been working with, in or around the cloud for a good portion of my career, first as a Data Center manager in placement decisions be it into the cloud or purposefully in colo sites, or in data centers I managed, to my work as a cloud architect, the one thing I am always asked about is use cases.  The question is usually either asked as a way to defend keeping all things in an owned data center, or by folks who have only heard of cloud in passing and can’t quite grasp what it really is.  My answer is usually around what the business goals are and what current problem areas exist.  This is because I don’t think cloud, just like all of the rest of IT, is a one size fits all.  Basically, it is a tool in the tool belt of IT, just like various SAN, NAS, Server and Software options are.  Not everyone fits running SAP, just like not everyone needs a Tier 1 storage array.  The goals of the company and skill sets are what make the use cases for or against cloud or anything in IT.

With all of that said/typed in the above paragraph, I would like to take some time to define the most common cloud use cases that I see being adopted by the various customers I am working with.  Most people I have asked seem to think development or “bursting” type use cases are what cloud is for, however I rarely see these as the use cases that really lead people into taking advantage of the cloud and saving money doing so.  The use case/s I usually see the most that benefit in terms of flexibility, definable scope and return on investment are Backup, Archive, Disaster Recovery and to a lesser but growing extent Data Analytics.   These may not seem like the ones that many people hear about when they see cloud marketing, or attend a cloud seminar, but they are indeed some of the most cost effective use cases for the power of the cloud.

I will use some costing numbers in the following examples, these are real numbers rounded up to show a high water mark so to speak.  I have also purposely chosen a smaller customer example rather than a large enterprise to start with to illustrate in smaller more understandable numbers why these use cases can be good financial decisions.  I am going to outline Disaster Recovery using cloud solutions for a small corporation that has the following items:

  • Total number of systems = 5
  • Data size = 2.5 TB
  • Backup software utilized = Veeam
  • RTO target = 48 hours
  • RPO target = 12 hours
  • MOL target = 100%
  • Replication manner = AWS storage Gateway

Just a quick level set on definitions:

  • RTO = Total outage window
  • RPO = Total data loss potential
  • MOL = Acceptable performance level of the DR location compared to production.

I know this example is very small when compared to large environments, but I have chosen it specifically because many small companies don’t have real disaster recovery abilities or real backups because they are deemed too expensive based upon the size of the overall IT budget.  In this example, the customer was able to utilize an AWS storage gateway and Veeam backup software to create a workable disaster recovery and backup plan using Amazon Web Services S3 storage and during a disaster EC2 compute.  This example does take human work to get the system up and running in a disaster but is much better than the original solution of backups written onsite to the same storage array as production with no ability to recover if the storage or facility was lost.

Actual cost figures for a disaster recovery test:

  • Actual RPO achieved = 8 hours
  • Actual RTO achieved = 36 hours
  • Actual MOL achieved = 100%
  • Monthly Spend for DR/backup data =  $200 to $300
  • Cost of DR test = $50
  • Cost of running 30 days at DR = $1500

So as you can see, for minimal investment, this customer was able to have a workable disaster recovery plan, because of the cloud and the ability to not have a full time disaster recovery site.

For those who are saying yes, but that is such a small environment, I will provide another cost example for a larger customer.  This one is more involved so I will not go into all the details, but just list out what was used and the real costs.

  • Total number of systems = 100
  • Data size = 96 TB
  • Backup software utilized = Veeam
  • RTO target = 4 hours
  • RPO target = 30 Minutes
  • MOL target = 100%
  • Replication manner = Zerto and Altavault

Results of DR test

  • Actual RPO Achieved = 12 Minutes
  • Actual RTO Achieved = 6 hours
  • Actual MOL Achieved = 100%
  • Monthly Spend for DR/backup data =  $2800
  • Cost of DR test = $1900
  • Cost of running 30 days at DR = $5500

In summary, it has been my experience that a good place to look at cloud in terms of financial benefit and increased resilience using the ability of the cloud is often DR, backup and archive.  If you would like to know more specifics, as with everything in IT, mileage will vary depending on the specifics of the actual data and software involved, feel free to  Evolving Solutions to discuss any of these use cases in greater detail.

______________________________

James Keating III is a Business Technology Architect for Evolving Solutions. James is a technology leader in the IT community and brings a combination of excellent technical skills and business acumen which allows him to guide customers in developing IT solutions to meet business requirements.