EFFECTED USERS: Everyone who stores data, especially if you run a SMB
Hair of fire 4 of 5
SUMMARY: Data tampering is increasing, primarily though ransomware.
In a recent article in Forbes entitled, How Data-Centric Security Can Protect Data Lakes And Safeguard Innovation
Over the last 40 years, as the internet grew from a Department of Defense project to the behemoth it is today, the role of the application also changed. Instead of drawing from a single database as did the process-automation apps of yore, many of today’s apps aggregate data from multiple databases. Because databases remain siloed, aggregation requires building a data lake inside of a data warehouse. APIs with custom-coded security suck data out of the back end and into the lake. And that is where everything risks exposure.
I believe that we should not attempt to build another layer of protection around the lake — an API-driven process that often comes at a large cost — but in securing the data itself.
There are several components to data defending itself. Programming the data layer with fine-grained security logic shrinks potential attack surfaces. Such logic also bakes in authorizations so that users can access data in a fluid way, rather than controlling authorizations with middleware. Time-stamping data and making it traceable enables tracking and instant retrieval of data. This bypasses the need for manual replication, which in turn closes additional attack surfaces.
Here with me to speak about this today is the author of the article, Brian Platz.
Brian is the Co-CEO and Co-Chairman of Fluree, and was previously co-founder of SilkRoad technology.
What is a data lake?