Browse Prior Art Database

SERVICE AWARE TAIL DROPPING OF PACKETS AT THE EGRESS QUEUE

IP.com Disclosure Number: IPCOM000245535D
Publication Date: 2016-Mar-15
Document File: 9 page(s) / 121K

Publishing Venue

The IP.com Prior Art Database

Abstract

This disclosure provides service aware dropping of packets at the egress queue. This disclosure provides the capability to user to prioritize one service over the other within a particular queue according to his/her needs based on prioritization block and hence helps the user to give more importance to a particular service to get buffered and allow lower priority service to get dropped first during tail dropping.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 16% of the total text.

SERVICE AWARE TAIL DROPPING OF PACKETS AT THE EGRESS QUEUE

ABSTRACT

[0001]               This disclosure provides service aware dropping of packets at the egress queue.  This disclosure provides the capability to user to prioritize one service over the other within a particular queue according to his/her needs based on prioritization block and hence helps the user to give more importance to a particular service to get buffered and allow lower priority service to get dropped first during tail dropping.

BACKGROUND

[0002]               The traditional technique for managing egress queue lengths is to set a maximum length (in terms of packets) for each queue, accept packets for the queue until the maximum length is reached, then reject (drop) subsequent incoming packets until the queue decreases because a packet from the queue has been transmitted or aged out at the head of the queue and have been dropped. This technique is known as "tail drop," since the packet that arrived most recently (i.e., the one on the tail of the queue) is dropped when the queue is full.

FIG. 1: Packet queue

[0003]               The tail drop discipline allows queues to maintain a full (or almost full) status for long periods of time since tail drop signals congestion (via a packet drop) only when the queue has become full.  It is important to reduce the steady-state queue size, and this is perhaps queue management's most important goal.

[0004]               Now during tail dropping when the queue is full (or, almost full) some packet needs to be dropped and these dropped packets will be chosen on the Last-in-First-Drop basis by the tail dropping mechanism irrespective of the service the packet belongs to and user has no control/capability to give the precedence to one service over another service at the same egress queue.

[0005]               Presently tail dropping that occurs at the egress queue during queue full will drop the packets on the basis of Last-in-First-Drop at the egress queue tail irrespective of the service from where the traffic is coming. Currently, the user has no capability to give the precedence to one service over another service during tail dropping at the same egress queue; hence prioritization of different services landing on same egress queue is not possible.

[0006]               Due to this limitation, the user is unable to prioritize the traffic coming from different services to a single queue during tail dropping and hence traffic from different services will be dropped without giving precedence to service to which packet belong during congestion scenario.

Example:

Assume, we have packets coming from different services on the same egress queue at the following rate.

Service A = 2 Packets

Service B = 2 Packets

Service C = 2 Packets

Total        = 6 Packets Ingress which needs to be Egressed

[0007]               Let us consider that we have left with space of only 2 packets in the egress queue. In this scenario, out of these 6 packets, 2 packet needs to be buffered in the queue and rest 4 packet needs to be dropped. Now these 4 pa...