Huawei Releases Low Latency White Paper Based on PGM2 Concept
[Shenzhen, China, October 11, 2016] Huawei has released a White Paper on its MS-OTN Low-Latency Network Solution. The White Paper introduces detailed requirements for emerging network services like financial lines, the IOT, cloud computing and on-line entertainment, and provides a comprehensive analysis of network latency and its optimization methods.
The White Paper introduces Huawei’s proposal of the PGM2 concept for low-latency networks. The paper also discusses three potential business models for future low-latency networks.
Low Latency: Guaranteed Good Service, Experience, and Competitiveness
With information saturation becoming an accepted norm, more stringent requirements have been placed on network latency in interactive entertainment, financial transaction, and other real-time services. How to define low latency has become a key factor affecting current business success and safety, as well as an important network performance quantifier for evaluating telecom network quality.
Different services have different latency requirements and Huawei has classified low latency services into four types: financial, the most sensitive at <1ms; IoT which includes telemedicine between 1-100ms; Cloud at between 5-50ms and Interactive Entertainment (7-200ms)
Factors Affecting Network Latency and Critical Optimization Technologies
According to the white paper, factors affecting network latency can be classified in three main categories: Network mode, Equipment and components, and Optical fiber paths.
Network latency is mainly caused by device components and fiber paths. On an Optical Transport Network (OTN), devices account for a small part of the overall latency, while fiber paths account for the majority of the overall latency. However, optical fibers are limited by deployment conditions and leave limited space for optimization. Thus the main latency reduction measures are as follows:
- Networking mode: Deploy the OTN that has the lowest latency.
- Device component:
o Flexibly configure FEC based on application scenarios to reduce latency penalties.
o Flexibly configure FEC based on application scenarios to reduce latency penalties.
o Optimize FEC performance to extend the transmission distance and reduce regeneration sites.
o Use the advanced OTN optical-layer technologies to achieve direct optical connection and switching, as well as to avoid unnecessary electrical-layer processing. o Implement coherent communications with Raman amplifiers deployed.
- Fiber path:
o Optimize fiber routes and paths.
o Use new low-latency optical fibers in the future to achieve a transmission speed close to the speed of light in a vacuum.
Low-Latency Network Requirements —PGM2
Currently, the popularity of network services is at an all-time high. A low latency network does not just describe the low latency feature, it also describes a complete system solution covering finance, entertainment, and home scenarios. In addition, future low latency networks must comply with the following PGM2 requirements: Promisable, Guaranteeable, Manageable, and Monetizable.
- Promisable: The technology with the lowest latency must be used and link latency must be precisely quantified, so that network latency can be accurately promised.
- Guaranteeable: Under the premise of accurate network latency, low latency performance must be highly stable and be guaranteed regardless of the traffic load or any unforeseen accident.
- Manageable: Network latency resources can be intelligently managed, monitored, optimized, and scheduled.
- Monetizable: The ultimate goal is to sell low latency networks to win customer recognition and better support network services.
Future Business Modes of Low-Latency Networks
Based on low latency, Huawei proposes three business modes for the future:
- Leased line renting where enterprise customers can select leased lines based on the network latency requirement.
- Latency on demand where low latency resources are quickly scheduled, such as online shopping rushes and AR/VR event live broadcasts.
- Latency customization based on service requirements, where resources with the lowest latency are reserved for the users who have higher requirements
Huawei has recognized the importance of low-latency in future network communications and social life and has accumulated in-depth technical experience and market understanding.
Based on low-latency OTN technologies, Huawei fully promotes simplified network architecture and one-hop transmission to further reduce network latency and improve latency stability. Based on physical hard pipes of OTN, Huawei devices can accurately promise (achieving "Promisable") and guarantee E2E low latency at all-time regardless of accidents or workload (achieving "Guaranteeable").
Huawei WDM products support the standard-defined test and OTDR method for latency measurement to ensure high-precision E2E online latency monitoring (accurate to 1 μs). Based on these feedback systems, the T-SDN controller is able to implement real-time and intelligent management, monitoring, and optimization on the latency of the entire network (achieving "Manageable").
The comprehensive OTN+T-SDN solution from Huawei enables sale of secure and reliable network pipes with the lowest latency (achieving "Monetizable"). Based on T-SDN, this solution provides accurate management and monitoring capabilities, and has been applied in many low-latency leased line projects and verified and well accepted by the market.
For details on the White Paper, please visit the following link to download: White Paper on Huawei MS-OTN Low-Latency Network Solution.http://www-file.huawei.com/~/media/CORPORATE/PDF/white-paper/huawei_optical_network_low_Latency_solution_white_paper_en.pdf