Share Email Print
cover

Proceedings Paper

Three layers multi-granularity OCDM switching system based on learning-stateful PCE
Author(s): Yubao Wang; Yanfei Liu; Hao Sun
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

In the existing three layers multi-granularity OCDM switching system (TLMG-OCDMSS), F-LSP, L-LSP and OC-LSP can be bundled as switching granularity. For CPU-intensive network, the node not only needs to compute the path but also needs to bundle the switching granularity so that the load of single node is heavy. The node will paralyze when the traffic of the node is too heavy, which will impact the performance of the whole network seriously. The introduction of stateful PCE(S-PCE) will effectively solve these problems. PCE is composed of two parts, namely, the path computation element and the database (TED and LSPDB), and returns the result of path computation to PCC (path computation clients) after PCC sends the path computation request to it. In this way, the pressure of the distributed path computation in each node is reduced. In this paper, we propose the concept of Learning PCE (L-PCE), which uses the existing LSPDB as the data source of PCE’s learning. By this means, we can simplify the path computation and reduce the network delay, as a result, improving the performance of network.

Paper Details

Date Published: 24 October 2017
PDF: 6 pages
Proc. SPIE 10464, AOPC 2017: Fiber Optic Sensing and Optical Communications, 1046406 (24 October 2017); doi: 10.1117/12.2282832
Show Author Affiliations
Yubao Wang, Yanshan Univ. (China)
Yanfei Liu, Yanshan Univ. (China)
Hao Sun, Yanshan Univ. (China)


Published in SPIE Proceedings Vol. 10464:
AOPC 2017: Fiber Optic Sensing and Optical Communications
Zi-Sen Zhao; Leping Wei; Yanbiao Liao; Weixu Zhang; Desheng Jiang; Wei Wang; Kenneth T. V. Grattan, Editor(s)

© SPIE. Terms of Use
Back to Top