Is your organization really using hybrid analytics?

Is your organization really using hybrid analytics?

By Drew Clarke, VP of Cloud, Qlik

The simple definition of hybrid cloud is a computing environment that uses a mix of on-premise, private cloud, and/or public cloud infrastructure to deliver services, with orchestration between the platforms.  Hybrid cloud joins multiple clouds – or on-premise installations with cloud-based installations. Under that general definition, many vendors will claim “hybrid cloud analytics” in their marketing verbiage. However, you have to get into a little more of the details to better understand what value you are getting.

For example, being able to publish an analytical application (or sheet for some) from an on-premise installation to a cloud offering could be valuable, but it’s not hybrid cloud analytics.

There are a few key ingredients that we use to define our approach for hybrid cloud analytics:

Transparency – First off, a hybrid cloud analytics solution is completely transparent to the end users of where the data resides, and where the analysis happens. Any user should be able to access their environment, from any device, and choose what data and/or applications they want to view and interact with, regardless of where it sits and runs. You should have universal hub which represents everything available to a user based on their role and security permissions, not the location of where things reside and run.

Location Enforcement – For many reasons, customers may choose that data sources - and/or the analysis that is run against those data sources - stay in a particular environment.  Typically, they want to restrict data and/or the analytical applications with that data to an on-premise environment, behind their own firewall. This may be due to several reasons including but not limited to industry regulations or data that represents a company’s most secret competitive weapon. Regardless, we believe a properly governed solution enables organizations to define rules around where data and/or the analysis on that data can be stored or run. For example, you might choose that a particular data source is too sensitive to allow outside of the firewall, and that any analytical application that uses that data source should also therefore run behind the firewall. By simply designating this in a management console, you can create enforcement rules on where things can and will reside based on that data set.

Orchestrated Entitlement – One obvious need is for organizations to be able to easily manage entitlements and licensing for their user base across the hybrid cloud solution. This is a basic element of “orchestration between the platforms”.

Bi-Directional Migration – If the whole point is to enable customers to choose where data and analysis should occur, based on their own criteria, then the solution must allow for bi-directional migration to/from one infrastructure environment to another in the hybrid cloud deployment.

Single Management Console – A hybrid cloud analytical solution should be managed as one, seamless environment across infrastructure boundaries, so it should be managed via a single console.

With that in mind, when thinking about your long-term hybrid cloud analytics strategy, ask your teams and vendors these questions:
-          Can the user access data from any device based on their role and security permissions and not be limited by restrictions of where data resides?
-          Can we define rules around where data and/or the analysis on that data can be stored or run?
-          Can we create enforcement rules on where things can and will reside based on the sensitivity and security of that dataset?
-          Can we manage user entitlements and licensing between the platforms?
-          Can we migrate data to and from one infrastructure environment to another?
-          Can we manage all our data in one seamless environment across cloud and on premise infrastructure?

In essence, don’t be fooled by marketing speak. To ascertain if your organization is using hybrid analytics, be sure to raise relevant questions and use the above checklist of Transparency, Location enforcement, Orchestrated entitlement, Bi-directional migration, and Single management console.

For the LATEST tech updates,
FOLLOW us on our Twitter
LIKE us on our FaceBook
SUBSCRIBE to us on our YouTube Channel!
    Blogger Comment
    Facebook Comment