Humans and computers have evolved in very similar ways… although this isn’t obvious at first glance.
The human brain has massive computational power and has access to large data sets through all the senses. However, it’s a low bandwidth, high latency decision system. To compensate for this, humans have evolved into creating an extension of this computing fabric called the spinal cord that is closer to the edge. It offloads the brain with high bandwidth, low latency specialized computing. In this way, models created in the central brain are eventually programmed into the spinal cord for more efficient behavior, and long term survival.
When a child learns to walk, learning models and processing needs are being pushed down from the central brain into the spinal cord. Same thing happens when one learns to ride a bike or drive a car.
Finally, at the edge, there is a large surface area of peripheral nerves leading to the skin. This is where the bandwidth of data is massive, and decisions are very time sensitive: Is it a spider or your shirt on your back? The primary job is to do as much local processing as practical to relieve and abstract the higher layers from a deluge of information.
Billions of years of evolution cannot be wrong.
This is how computers have evolved as well. We see time-sensitive processing done at the edges, specialized models for processing done at the mid tier, and complex and intelligent processing happening at a central place.
There is a central brain, the CPU, that has access to large data sets, runs complex algorithms, and is the most powerful component in the system. Nonetheless, it needs to offload certain mundane tasks to daughterboards like NICs, RAID controllers, and specialized graphics accelerators. Finally, the peripherals connected via USB do highly specialized computing.
Two things to learn from the above examples:
- As we get closer to the source of data, the information bandwidth gets higher, the intelligence needed to make decisions gets lower, and the decisions get quicker.
- A single fabric is important (be it silicon, or be it neurons) to allow what used to happen at a higher layer to be dynamically programmed at a lower layer either via learning or via evolution.
And Now…the Cloud Computing Nervous System
We are now undergoing a similar evolution in Cloud Computing. The large public cloud providers are taking the place of a central brain where a lot of data is accumulated; these clouds are differentiating based on both the richness of data sets and the superiority of machine learning algorithms. This central cloud will continuously process data and create opportunities for a new “dispersed cloud” at the edge to locally analyze, act, and summarize. Such a dispersed cloud that acts like a spinal cord will reach out to important hubs in the world for low latency, high bandwidth connectivity. A “system of nerves” is finally evolving into a Nervous System.
A lot of what is happening in the central clouds today will inevitably move down to the spinal cord or the edge as the level of intelligence evolves to a much higher level in the central cloud. This is the reason we need a common fabric across the central clouds and the edge. The rapidly expanding world of IoT that is getting hooked into this dispersed cloud is only accelerating this evolution.
Laying the foundations for the next era of computing
I am thrilled to see two great innovation engines, Google and Nutanix, come together. While Nutanix has proven its success in creating the next-gen web-scale infrastructure that spans data centers from HQ to a remote office/branch office, all the way to the edge, Google has proven its success in creating large state-of-the-art computing infrastructures that excel at machine learning and data and analytics at scale.
The fabric of both these infrastructures have many commonalities. They are both built with common software-defined open-source components, and fundamental tenets of modern distributed computing such as: always-on with no single point of failure, fail-in-place with distributed everything, optimizing operations with machine learning, API-first and all intelligence in software.
When such a common fabric comes together, it creates a very powerful computing paradigm with the following properties:
It’s critical to have a unified management of all the applications running in a dispersed cloud including those running on the edge. Nutanix Calm will provide for intentfulrepresentation and operations of applications across a dispersed multi-cloud environment. This includes a single pane of glass for managing both private and public cloud. Traditional and cloud-native applications can be provisioned into Google Cloud Platform (GCP) or on-premise Nutanix cloud environments with a single click, and migrated between the two cloud environments using Nutanix Xi Cloud Services.
At the same time, Nutanix is enabling a 1-click deployment of Google’s Kubernetes fabric into the private Nutanix-based cloud. With Nutanix Calm, the user can launch, upgrade, troubleshoot, and scale the Kubernetes platform for modern Mode 2 apps on the Nutanix infrastructure. Atop this platform, users can launch Kubernetes pods or Helm charts with a single command leveraging persistent storage capabilities provided by Nutanix Acropolis in the underlying infrastructure. This brings notions of 1-click lifecycle, auto-tiering, data locality, QoS, encryption, replication factor, snapshots, backups, DR, etc. invisibly to Mode 2 apps on-premises.
Once the fabric in the central cloud and the dispersed cloud is materially the same, the central cloud is enabled to push some of its work to the dispersed cloud via learned models and packaged apps. Nutanix Enterprise Cloud OS has demonstrated the ability to run at any-scale, by providing support for large deployments of thousands of nodes to 2-node and 1-node deployments for remote deployments from ROBO to on oil rigs, barges, and remote outposts. In fact, Nutanix Enterprise Cloud OS can be embedded even in an Intel NUC on a drone.
This unique capability of extending a single compute fabric all the way from the edge to the central Data-center, places Nutanix in a unique position in the industry. Joining hands with Google to extend technologies like TensorFlow allow machine learning, insights from training models, edge-processing and analytics to be seamlessly pushed to the edge on top of Nutanix’s ubiquitous Cloud OS. The Nutanix Enterprise Cloud OS itself has been at the forefront of bringing efficiency to data center operations via MapReduce, data tiering, ILM, versioning, etc. since its inception. While these transactional capabilities are now table stakes, the Nutanix Enterprise Cloud OS has been evolving to bring ML and analytics (with X-Fit) and aims to bring TensorFlow for cognitive needs in large data-centers.
As we witness this interesting revolution in cloud computing, there is still a lot to learn from how things have evolved in the past. As Jean-Baptiste Alphonse Karr rightfully observed in 1849, “plus ça change, plus c’est la même chose”— “the more things change, the more they stay the same”.
Note: The integration of Nutanix Calm and Google Cloud Platform will be available in the first quarter of calendar year 2018. Other features are in development, and pricing details will be announced closer to the release.
Disclaimer: This blog may contains links to external websites that are not part of Nutanix.com. Nutanix does not control these sites and disclaims all responsibility for the content or accuracy of any external site. Our decision to link to an external site should not be considered an endorsement of any content on such site. Please send us a note through the comments below if you have specific feedback on the external links in this blog.
This blog includes forward-looking statements, including but not limited to statements concerning our plans and expectations relating to product features and technology that are under development or in process and capabilities of such product features and technology, our plans to introduce product features in future releases, strategic partnerships that are in process, product performance, competitive position and potential market opportunities. These forward-looking statements are not historical facts, and instead are based on our current expectations, estimates, opinions and beliefs. The accuracy of such forward-looking statements depends upon future events, and involves risks, uncertainties and other factors beyond our control that may cause these statements to be inaccurate and cause our actual results, performance or achievements to differ materially and adversely from those anticipated or implied by such statements, including, among others: failure to develop, or unexpected difficulties or delays in developing, new product features or technology on a timely or cost-effective basis; delays in or lack of customer or market acceptance of our new product features or technology; failure to form, or delays in the formation of, new strategic partnerships and the possibility that we may not receive anticipated results from forming such strategic partnerships; the introduction, or acceleration of adoption of, competing solutions, including public cloud infrastructure; a shift in industry or competitive dynamics or customer demand; and other risks detailed in our Form 10-Q for the fiscal quarter ended April 30, 2017, filed with the Securities and Exchange Commission. These forward-looking statements speak only as of the date of this presentation and, except as required by law, we assume no obligation to update forward-looking statements to reflect actual results or subsequent events or circumstances.© 2017 Nutanix, Inc. All rights reserved. Nutanix, the Enterprise Cloud Platform, the Nutanix logo, Xi and Nutanix Calm are registered trademarks or trademarks of Nutanix, Inc. in the United States and other countries. Google and Google Cloud Platform are registered trademarks or trademarks of Google Inc. All other brand and product names mentioned herein are for identification purposes only and are the property of their respective holder(s).