This is especially true when it comes to shiny new and exciting technology. In my previous blog I talked about how micro-services is making a big impact in the tech world and why we chose to adopt this as part of our multi-cloud strategy. Kubernetes (a.k.a K8s) plays an important role, delivering orchestration on a massive scale with elasticity and speed, self-healing apps through expendable resources.
However, moving over to Kubernetes is not a decision to be taken lightly. It is not as simple as a lift and shift project moving from bare metal infrastructure to a cloud provider, such as AWS, would be. This is a whole new paradigm. There are some key considerations beyond just the tech to think about, people and skills are just as important.
A deployment of Qlik Sense Enterprise on Kubernetes requires many and varied skillsets. Kubernetes, Docker/Container, Helm, Load balancing, Linux/UNIX, Identity providers and of course Qlik Sense are all skills that will be required. If your organisation lacks any of these then factor in training, time and/or recruitmentto fill the gaps or you could work with a 3rd party who can support you in these areas, such as Qlik Consulting and our strong partner ecosystem.
Here are 7 questions you should ask to ensure deploying Qlik Sense Enterprise on Kubernetes is the right fit for your organisation.
Is Kubernetes something you have already?
Moving to Kubernetes is an organisation level strategic decision, part of a wider IT and DataOps strategy and should not be made in the context of just one software product.
Kubernetes and container-based applications are not simply another platform like Windows vs. Linux. They are a different way of building, deploying and managing software.
If you say Kubernetes and your IT department says ‘kuber-whaties’? Then it’s probably not the right time. Do you really want your organisation to learn kubernetes on your business-critical analytics applications?
Do you still need a Windows deployment?
Does it make sense to move your products into Kubernetes? If your existing Qlik solution on Windows is mature and stable and it is not expecting any significant growth, would you really get value from moving it? There may be practical reasons to keep a Windows deployment even if it is not technically required.
Do you have dependencies on Windows?
If your current deployment of Qlik relies on APIs, ODBC drivers or any custom solutions that only run on Windows then it’s worth checking that the current set of APIs that are exposed for Qlik Sense on Kubernetes allow you to do the same things you do now.
Does your Infrastructure provider support Kubernetes?
The main cloud vendors such as AWS, Azure, IBM all provide kubernetes offerings, however if your running with a bare metal provider you may have to build and support the kubernetes clusters yourself and if you are utilizing dedicated static fixed hardware resources the benefits may be negated.
Do your managed service providers & solution providers support Kubernetes?
If you currently rely on 3rd party organisations, how mature are they in Kubernetes? As part of a new strategic direction you want the best people, with the correct skills, to look after your organisation and applications.
Does your environment have access to the internet?
At the time of writing deploying Qlik products to Kubernetes requires access to the internet for Qlik’s helm and docker repositories as well as licensing requiring an occasional connection.
Would you be better with a full Enterprise SaaS solution?
If any of this seems too daunting or represents a challenge, then it’s worth considering accessing Qlik Sense from our Qlik Cloud Services. Qlik’s fully managed Software as a Service provides full Enterprise SaaS capabilities with largely the same functionality as Qlik Sense Enterprise on Kubernetes. As it is managed and hosted by Qlik it means you don’t have to have any knowledge of Kubernetes or containers and can get up and running with next to nothing resources and IT skills.
This article was written by Adam Mayer and originally appeared here: https://blog.qlik.com/is-k8s-your-soul-matehttps://blog.qlik.com/5-technical-skills-every-successful-analyst-needs-to-have