Router with class of service mapping
US6463068B1 · kind B1 · utility
Assignee
Inventors
Key dates
| Filing date | Dec 31, 1997 |
| Grant date | Oct 8, 2002 |
| Priority date | — |
| Expiry date | Dec 31, 2017 |
Classification
- Technology area (CPC H)Electricity
- CPC primaryH04L47/32
- WIPO fieldDigital communication
- WIPO sectorElectrical engineering
Abstract
A router includes a classifier that classifies packets assigned to 2n−m classes of service into 2n classes of service that are supported by the router. The classifier then sets the loss priorities of the respective packets to one of m levels. The router uses a modified weighted random early detection scheme that is based on probabilities of discard associated with the 2n+m classes of service to determine whether to retain or discard particular packets. The router uses a single buffer to store packets directed to all of the various output ports. The available storage locations in the buffer are linked to a free queue and a weighted average depth of the free queue, is used to determined whether or not to retain a given packet. The router compares the weighted average depth of the free queue to maximum and minimum thresholds associated with the particular 2n+m class of service to which the packet is assigned. If the weighted average is above the maximum threshold, the packet is retained. If the weighted average is below the minimum threshold, the packet is discarded. If the weighted average is between the two thresholds, a probability of discard that is based on the 2n+m classes…
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.