"Platforms” are all the rage in software positioning and messaging. And recently, a new platform has become the “platform du jour” – driven by the urgency felt by enterprises as they struggle to manage an increasing amount of data and an increasing number of data formats all generated from an increasingly number of applications on an increasingly diverse mix of infrastructure – the “data platform.”
Of course, when you look under the hood, many of these so-called data platforms are more likely to just be vendors’ old point solutions, able to only manage one piece of the larger data management puzzle, either because they only manage a specific subset of data – on premise data, virtualization data, or particular types of application data – or fail to provide comprehensive data management capabilities – only replication, but not data governance.
If we hope to correctly assess whether someone actually has a real data management platform, we have to first determine what the fundamental needs that enterprises have when it comes to managing data, and what are the core capabilities they need from a platform if they hope to meet these needs.
We believe that there are five fundamentals to managing data – you need to be able to know it, federate it, mobilize it, govern it, and use it. These fundamentals provide you with the power to implement a holistic data management strategy – one that does not just allow you to ensure that your data is fully protected, but also ensure that you can activate this data for business value.
Stemming from those fundamentals, we think there are six key capabilities that define a modern "data" platform.
- Open Data Access – Data must be ingestible and consumable via industry standard protocols, like ReST, File system, Storage or APIs, not just a proprietary GUI or connectors – no matter how easy these proprietary methods are to use. A modern data platform must be truly independent and future proof so organizations have confidence they can get to their data whenever and however they choose.
- Virtual Data Consolidation – Data is splashed across locations, infrastructures, delivery methods, and silos. It is impossible to practically bring all an organization’s data together to one location, one format, or one silo. A modern data platform must virtually unite disparate data locations and formats by providing consistent management, operations and navigation of data sets, as well as making data sets portable between infrastructures.
- Deep, Adaptable Data Indexing – Metadata is the foundation for more intelligent control of data. Without intelligent control, you are relegated to basic movement of blocks or containers – a wasteful and less effective strategy. A modern data platform’s metadata handling must go way beyond simple ACLs, but into content, faceting, classification, and natural language processing. It must programmatically accept dynamic, custom attributes because a modern data platform must be able to get smarter and smarter about the data.
- Comprehensive Data Security – More than just encryption, a modern data platform must provide the authentication and authorization for individual data objects to prevent data leakage or loss. Data security must handle complex and changing permissions, roles and responsibilities, and tightly integrate with an organization’s existing directory and security services.
- Lifecycle Data Services – More than any other capability, lifecycle data services offer an opportunity for significant cost savings, risk reduction and operational simplicity. A modern data platform should transparently orchestrate and automate the lifecycle, copy management, compliance and governance of data across infrastructures, application types, formats, containers, locations, even SaaS. A data platform is the natural software layer to control the underlying storage resources it uses. It can optimize hardware utilization and performance to the data lifecycle, thus minimizing cost, by aligning redundancy, copies, tiering, security, and cost to the data profile, access and usage. But a modern data platform shouldn’t lock you into specific hardware in order to get the lifecycle and copy data capabilities you need.
- Data Value Delivery – Any platform’s mission is to provide value to its users. Data value is more than just analytics or simple data visualization, rather it is the ability to match information with the user’s needs. For example, data value can be delivered by using GPS coordinates to link third party contractor’s work with safety issues, or matching personalized patient data with distributed healthcare information, even correlate vendor invoices with emails about subsequent terms and conditions. The greater accuracy a modern data platform can match information to user needs, the greater its ability to deliver value.
At its essence, a modern data platform primary role is to deliver value to its users. To deliver value, a data platform must understand data at a very deep and granular level. Without this level of data knowledge, it’s impossible for a modern data platform to do its job – to protect and activate your data – simply, reliably, securely, and cost-effectively.