Virtustream Blog

What did we do before GPS or Global Positioning?


in Technology

Before GPS navigation

Explorers navigated by pure luck and chance. Eventually, navigation relied on the sun’s direction during the day, the moon by night, and later by stars with a technology invention called the sextant. Next came radio positioning and now satellites enabling GPS – a technology that is continually improving and can determine your exact location on the globe within inches. This leads me into several conversations I have been having both internally and externally with customers on what on-premises and off-premises actually mean.

It turns out, it can mean both, neither or a combination depending on the baseline of context. For GPS to work, satellites are in a geosynchronous orbit, meaning they never move in relation to the Earth’s rotation. GPS would be useless if they did. And, this is exactly what is happening in our conversations when the basis of on-premises changes based upon the reference point.

Let’s take a customer’s point of view. On-premises may be defined as their wholly owned data centers. But, what if they have a partner colocation? On-premises may be in their defined and contracted location because it is truly theirs. Another may be off-premises. The same location of the collocated center may now be referenced as off-premises, along with a public cloud location that is also off-premises. You can quickly see the discussion definition absolutely depends on your reference point not moving or changing definition during an exploration.

Case in point, I have recently approached my conversations with an establishment of terms. I define three locations like this:

               On-Premises-Private   A dedicated, customer owned and managed location housing their data and physical equipment.

               Off-Premises-Private   A dedicated, usually not customer owned location totally dedicated to their data and their physical equipment.

               Off-Premises-Shared   A shared physical location that is NOT customer owned or customer dedicated. This always has multi-tenancy security controls; however, all physical assets are leveraged and multi-tenant.

This brings us to the use case that defines the workloads.

  • A cloud service provider can provide both off-premises-private and off-premises-shared if some of the infrastructure is dedicated and some of it is multi-tenant.
  • Some of the pure play cloud providers only offer off-premises-shared environments. This is defined as no dedicated infrastructure, bandwidth and other service level guarantees. They may simply be estimates and objectives.
  • Others specialize only in providing services of on-premises-private.

Please don’t take away that I am discrediting any of the above. Just the contrary, I am an advocate for all. However, which one is the best? The answer is none of them, all of them or some of them. I subscribe to workload definition criteria as to where a workload should be housed. This is based on workload requirements, security, business criteria and many service level definitions that are too broad to cover here.

My advice is simply to understand your reference point when evaluating/discussing workloads and physical location criteria. Your reference point must never move and you will be assured you can explain where you are, where you have been and where your workloads are going.

Chris Gaudlip
CTO

Twitter @chrisgaudlip