Browse Prior Art Database

Method to optimize optical connections for wattage consumption

IP.com Disclosure Number: IPCOM000250506D
Publication Date: 2017-Jul-26
Document File: 3 page(s) / 309K

Publishing Venue

The IP.com Prior Art Database

Abstract

This article explores a method to automatically tune and optimize optical transceiver power consumption used for connections within or between data center environments.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Absract This article explores a method to automatically tune and optimize optical transceiver power consumption used for connections within or between data center environments.

Data centers consume lots of power. Power cost is one major cost factor of running a data center. Reducing unnecessary optical power used in tens of thousands of optical links deployed between switches and servers can greatly help to reduce power cost. When deploying network systems in data centers, one of the major workload is to sort out hundreds to thousands of optical transceivers between short, medium, and long distance connections. It is time consuming and error-prone, and may damage the optical components if not handled correctly. Longer distance optical links require more power to operate than shorter distances, but excessive power may damage or shorten the lifespan of the optical receiver. It is important how to sustain the most suitable optical power for long distance communication, so as to avoid service disruption due to unsuitable optical power output (e.g. due to operating temperature variations). All of these above problems lead to the thinking of software defined network in the domain of optical power consumption, and how to leverage advances in data analytics to solve operational issues.

Disclosure is a system that enables software defined control of data center network optical transmission. The system collects sensor data from both the transmitting and the receiving ends of a optical transmission system, and feeds the collected operational data to a software-defined centralized controller (SCC) for data analytics and central processing, and finally distributes the power decisions across the data center to control optical transmission power.

The SCC within the system calculates best power decision based on sensor data from all

optical components. The SCC can be much more efficient and intelligent when a large number of optical components are involved, as it can levera...