Closed-loop control is a central component in the delivery of most modern dynamical systems, represented by aircraft, automobile engines, spacecraft, and large distributed power grids. Engineers begin the design of these complex systems by modeling the governing differential equations, referred to as the "characteristic equations." They then apply control systems theory to attain optimal performance using real-time measurements. The application of a feedback controller to a dynamical system is called "closing the loop."

Autonomy is enabled via a Control System

Packet-switched routing networks are dynamical systems that to date have not benefited from closed-loop control. Unfortunately, for many decades the derivation of the underlying closed-form solution for the characteristic equations remained elusive. In the absence of preferred closed-loop control, very basic, heuristic protocols like BGP and OSPF were developed to control these networks.

Mode HALO supports higher throughput and preserves optimal latencies
Today's heuristic routing is static and leads to network underutilization

The National Science Foundation sponsored research at Cornell University to end this decades-long quest. These researchers revealed the characteristic equations that define any packet-switched network. They then implemented their breakthrough as HALO, the world's first distributed real-time control system for packet switched networks (read the original paper).

Mode HALO parallelizes each node's routing decisions, which are made every 150ms
The proportionalities behind Mode HALO

Mode was founded by these same researchers, who designed a commercial version, Mode HALO, and an autonomous global virtual network implementation, Mode SD-CORE.

Mode HALO was evaluated in a series of experiments including the National Science Foundation GENI Test, the AT&T SDN Network Design Challenge, and others, described below.

Mode Router

The logical next step was to build a global network controlled by Mode HALO, leveraging its inherent efficiencies to offer SLA-backed reliability at actual business-internet pricing. Unfortunately, legacy routers are not designed for dynamic control algorithms. In the face of this constraint, the Mode team accelerated the development of virtual routers. The result of this design effort was a virtual, carrier-grade router that could be dynamically provisioned, modified, and controlled: the Mode Router.

Mode is everywhere business is done.
Mode Routers in the Mode SD-CORE overlay measure, calculate, and control undleray traffic in under 150ms – in parallel

Each Mode Router is carved out of the resources available in a standard blade server. Each router consumes two CPU cores — one for the containerized control plane, and one for the Open Virtual Switch (OVS)-based data plane.

Currently, a single virtual router using a small fraction of the available server CPU capacity delivers over 6 Gbps of throughput. Extrapolating to the typical multi-core server used in deployments, Mode achieves line rate throughput from the NICs (typically 40 Gbps).

Every Mode Router is able to support approximately 100K flows/sec, while storing 100K flow rules in the flow table — sufficient for most enterprise private networks. For more demanding enterprise users, Mode has leveraged architectural innovations that guarantee high availability and multi-tenancy with Mode Router, and developed techniques to combine these individual virtual routers into larger routers which can scale to match any demand.