Democratising Your Data – Part Two

By Carl Ferrer – April 2020

Head Of Technology – FWD View

 

 

Democratising your Data while remaining within compliance

There are some impressively smart technologies available that allow the democratisation of your data, without creating more security holes than letting Edward Snowden borrow your laptop…

For instance, data catalogues provide a backbone to democratising data thus, giving people the ability to see it exists as their initial step. Furthermore, having the ability to search an existing data catalogue for a simple- or complex- business term, swiftly refining and then pushing the data you have sourced into your preferred analytics tool is a great example of personal empowerment. Simply put, you can allow users to locate any data in your enterprise, but also maintain autonomy by only allowing them pre-specified permissions. You can even use this platform to request access to data you would like to use, but don’t yet have the required permissions. 

Virtualisation Technology is a good place to start. Take Delphix for example. Delphix quickly delivers production-like data in minutes while still maintaining the security of sensitive data. By allowing a developer or analyst the ability to pick a point in time and self serve their data at super fast speed is their killer feature. We are talking click provision, pop the kettle on, brew made and it’s ready and waiting. That fast! Gone are the days of waiting on DBAs or data owners producing copies that can take days. Users can now access the data they require, by themselves.

And if that’s not enough, the masking of sensitive data is applied as standard, automatically. So, even though we are giving the user access to the data, you as the data owner, can still restrict what users can see by automatically masking the data to a standard before they can view it. Thus, not exposing sensitive data. Your InfoSec team will be happy to hear this. 

Data management platforms are a more advanced solution for your enterprise. Incredibly powerful at leveraging your data and making it work for you, this is a more ‘belt and braces’ approach to democratising data. Ideally you need a platform that can cover distributed environments like Big Data, OnPrem and of course your favourite cloud vendor. It should give fine-grained, dynamic access control via Role Based Access Control (RBAC) AND Attribute Based Access Control (ABAC). Don’t limit yourself to just RBAC, it won’t cut it. Provide data obfuscation and be API driven so you can leverage all your end user tools and avoid any lock in. Data management platforms is a growing space, but I have my eyes on Okera our newest partner, as starting to become a leader here. The With some great support of the features highlighted above and being so incredibly well thought out in how to integrate into your Infrastructure. An area where they lead is their ability to dynamically protect the Data on the fly by masking or tokenisation. Providing views of the data, where the sensitive data can be removed or masked at the cell level, is a very elegant solution indeed.

Leveraging the power of Cloud can also help considerably. Allowing you to spin up data as users need it and then shrinking your footprint once they are done can be cost effective and very agile. Though this has some pitfalls. Namely costs and Cloud lock-in. This is a whole discussion in itself but ensuring that you have some controls and have considered the cost of the environment when fully utilised is important. Snowflake have a good model here to allow you to have a Data Warehouse that is flexible, agile and supports data sharing. Though Snowflake does not support Masking at present, this means you still need an Okera or Delphix to secure your data. 

 

Democratising your data is not all about technology;

 

Arguably the most important aspect is still empowering your people. It can be a jarring change to go from hoarding all your Data to sharing it like you have gone out of business. Quoting a famous film “Build it, and they will come” comes to mind here. 

However, if you make access to your data: 

* Simple

* Engaging 

* Clean and Accurate and;

* Inclusive

People will want to be a part of that change. Technology enables this to a point, but the buy in from the C level execs is paramount. Having a clear and simple strategy so everyone can be a part of it without questioning they are doing the right thing. And, adoption will suffer if users get stuck in a dead end and unable to get to what they need. If something is difficult, the majority will try to work an easier path around it. So, my last piece of advice is this; don’t try to attempt to do everything at once. 

People have a natural instinct to crave leadership, and therefore them needing to be persuaded towards democratised data. However, using a multi-tiered approach enables users based on their confidence level to use the tools appropriate for them. 

Using the Data Lake analogy; more confident users can swim further away from the safety of shore to find the depths of data they need, while the casual swimmers can stay near the shoreline and enjoy a wider view looking at the surface. 

 
To learn more about FWD View click HERE