What Is an Autonomous Network?
Automation itself, and the idea that technologies could be self-provisioning, self-diagnosing, and self-healing, has been around for some time. But with advances in Artificial Intelligence (AI) and cloud technologies, such fanciful notions are quickly becoming realities.
Nowadays, most of us use AI-enabled apps when we ask Apple’s Siri or Amazon’s Alexa for help with a task. And even streaming services like Netflix help us pick movies and TV shows using AI.
In the world of terrestrial and submarine networks, AI and the elements that make up the cloud are converging to enable the next level of networking.
Those elements include:
- Virtually boundless compute resources: Modern CPUs can be clustered to offer massive, almost unlimited parallel processing power.
- Essentially unlimited storage: Storing a single, massive set of data across multiple processing platforms—within the same data center or across numerous physically separated data centers—is not only feasible, it’s now common.
- Dynamic connections: Agile, on-demand bandwidth—provisioned with the precise connection needed depending on the type of data in transit—is a critical component of an autonomous network.
- Open source software: Open source solutions like Hadoop are at the center of AI transforming networks.
- Big data: For an autonomous network, the gathering and analysis of data from network sensors leads to better detection of patterns and anomalies.
- Sensors: Embedded sensors, and the data they produce, form the foundation of an autonomous network infrastructure
Ciena has been on the leading edge of the development for autonomous networking, releasing the first virtual control plane in 2008. But, even though it’s a significant advance, Ciena sees autonomous networking as too restrictive and too rigid. Our customers have been telling us for years that they need to maintain strict control of their network and that automation alone is not enough.
So Ciena has started to champion a broader, more holistic concept: the Adaptive Network, which is geared toward providing a network that can grow and adapt with a company as its business needs and markets change.
An autonomous network runs with minimal to no human intervention—able to configure, monitor, and maintain itself independently.
The Adaptive Network includes three important layers:
- Programmable infrastructure: This includes the network’s physical and virtual elements, as well as the telemetry gathered from them.
- Analytics and intelligence: The programmable infrastructure produces significant amounts of data. Some of it is big data that indicate trends the network learns and adjusts to over time. Big data can inform the network how to adjust in the long term, which traffic patterns to look out for, and which parts of the network could be vulnerable.
- Software control and automation: Effective automation of network tasks as defined by the network provider, such as loading access controllers and provisioning routers, can eliminate human error and keep the network running at peak performance.
Ciena’s unmatched product portfolio supports the Adaptive Network in a number of important ways. Critical components include:
- WaveLogic Ai: The next generation of Ciena’s industry-leading WaveLogic coherent technology fundamentally changes how optical networks are built and managed. WaveLogic Ai enables tunable capacity, from 100G to 400G, in 50G increments—giving you previously unattainable control over your network.
- WaveLogic Photonics: Ciena’s fully instrumented, intelligent photonic system includes WaveLogic coherent optics and flexible line elements that, combined with embedded and discrete software tools, offer superior automation, control, and visibility of optical networks.
- Ciena’s optical and packet platforms: Ciena has an array of converged optical and packet platforms built on flexible platform architectures that are indispensable in the autonomous network and provide efficient matching of client services to line capacity.
- Blue Planet Analytics (BPA): BPA is a core element of the Blue Planet intelligent automation platform. It normalizes real-time and historical data collected from multiple sources across the network and also seamlessly integrates with third-party big data cluster systems.
- Blue Planet Manage, Control and Plan (MCP) domain controller: Ciena brings the power of software-defined programmability to your next-gen Ciena network and service operations. MCP eliminates the manual, time-consuming, error-prone steps between multiple, separate management tools and provides the Blue Planet foundation from which you can evolve your operations to drive closed-loop intelligent automation across multi-vendor multi-domains.
Ciena’s 25 years of experience connecting the world makes it the perfect partner to deliver the Adaptive Network.
- Unmatched network experience: With more than 1,300 customers worldwide, Ciena supports 80 percent of the world’s largest network providers. Ciena has deployed over 160 million kilometers of coherent optical networks.
- Network provider services: Ciena has designed services specifically to help providers evolve their networks. Ciena’s approach supports the complete network lifecycle, with consulting, solution practices, and services oriented around the needs of providers.
- Partners: Ciena augments its value with a vibrant partner program that evolves and develops offerings and expertise to enable all aspects of the Adaptive Network.
- Open APIs and modern data models: At both the hardware and software layers, Ciena enables better real-time network telemetry and measurement using APIs. Ciena’s APIs provide faster, easier, and multi-platform/multi-vendor application development, easy integration with IT tools, and more efficient IT resource utilization.