Data Gravity is already here
Let's start with a question - when you hear the words "its virtual" do you envision the subject of the question as being unknown and potentially scary or enticing and you can't wait to hear more? This is question is as much philosophical as it is existential. You may be asking what does this have to do with gravity and the future of compute?
Basic primer - In the early days of compute, deployments or installations were physical by their nature, purpose built or bespoke with both hardware and software, and there was a one to one relationship between machine and application. It was a long time hence from the ubiquitous term of service we use today. New applications and infrastructure took weeks or months to roll-out, lots of signatures, and business model justifications. I for one, am glad that is the past. A service as we use it today, is the idea that a user experience, say a mobile phone app, or a web site, or even a smart TV, is being enabled or powered by a number of services that are glued together (integrated) via a superset of software controls (platform) running on a farm of machines, often many miles and milliseconds away.
Over the last half decade, we have seen significant changes in architectures and changes in software stack efficiencies, motivated by costs and more and more by environmental stewardship. Public and private cloud providers have invested billions in demonstrating and convincing the legacy generation of developers that hardware and applications are not inseparably connected. This philosophical and mental adoption, the idea that the infrastructure is disconnected from the application, we find the new relationship bounded only by imagination. Seceding the hardware from the application means that developers no longer have to worry about maximizing hardware resources and scale and can instead focusing on more important matters like performance, security, and enhancing the network communication language (application programing interface) API’s. This idea of modularity, many micro-services, each designed to do one thing, and do it well, all running alongside each other as peers or dependents in an environment, is often called a service-oriented architecture (SOA). The power of an SOA means that the platform and its services can run anywhere, and more and more is becoming out-sourced to an infrastructure (the physical bits) that is owned, managed, and maintained by a 3rd party company. This complex, but beautifully simple relationship is called “The Cloud.” The magic of the cloud is not that its hosted or managed, but it’s ability to transform the way software and creative minds work.
In the future, open source gives way to open systems, component ecologies, autonomous entities (AE); sometimes hardware, sometimes software, become the empowering tools with security controls and policies that will allow and wrap insecurities within small zones of access. Compromising one AE will not allow transitive access to its neighbor. Attacking or passing nefarious or even erroneous information between AE’s will either result in isolation or denials. Old school network practices like edge firewalls, VPN’s, and air gaps are not the future of security, and reliance on will become the anchor that impedes innovation. The baseline access method today is through a virtual network layer or virtual private cloud (VPC). The future of the VPC is the wild fire spread of pervasive computing that enables speed, performance, and interaction creating a data gravity well where data becomes the immovable object and stops chasing the sun, and the talent becomes nomadic. Public, Private, Hybrid, are just names that are used to comfort the wary and hinder the true ubiquitous capabilities of speed and performance. The virtualization of the desktop is already upending the paradigm allowing data and compute to remain secure, and latency has become the oppressor. New business models that move desktop applications and creativity tools to the “as a service” model and are charged on time use basis are driving the fixed seat model extinct.
Discrete object level security in the cloud is advancing as metadata and blockchain are the future of security in the cloud. Each and every data object should contain or conform cryptographically to the security standard that checks and authenticates access against an unlimited history file record that is updated continuously in real time as each object is touched. Access is granted based on source, credential, syntax, context, and pre-determined policies. Everything is logged and anything else alarmed. Audit logs and chains of custody will provide instantaneous alerting through logging and behavioristic API’s. Machine learning and newly evolving AI algorithms are tracking inbound and outbound data, exposing the unknown to the light.
In a world where AE security is known and trusted, such creates a potential for collaboration, and creativity by bringing the application to the data where actions are faster, cheaper, and verdant. As cloud services adoption increases because of price, flexibility, and popularity, data gravity will continue to create a data singularity surrounded by equally massive layers of compute. Already underway, such centralization is causing a tectonic shift in how and where creative and talented people live and work. Companies will be pulled into the gravitational field of the nucleus to compete and remain because of the talent and way innovation is evolving.