Resource sharing among multiple RSVP sessions
AssigneeCisco Technology, Inc.
InventorsMichael V. DiBiasio, Bruce S. Davie
Life of this patent
Estimating the term…
Abstract
In one embodiment, a technique is provided to share resources among data flows. An apparatus receives a request for a new data flow where the new data flow is associated with a session group (ID). The apparatus determines that the session group ID of the new data flow matches a session group ID of one or more existing data flows, and that the one or more existing data flows are not already sharing resources with another data flow that has a session group ID different from the first session group ID. Provided this is true, the apparatus shares resources reserved for the one or more existing data flows with the new data flow.
Claims25
A device comprising:
- a traffic scheduler having one or more resources for use in forwarding network traffic received at the device at different rates;
- a classification engine configured to identify received network traffic based upon predefined criteria;
- and
- a resource reservation engine in communicating relationship with the traffic scheduler and the classification engine,
- wherein, in response to a request to reserve resources for a first data flow associated with a first session group identifier (ID) and belonging to a first session, the resource reservation engine is configured to,
- determine that the session group ID of the first data flow matches the session group ID of one or more second data flows,
- check if the one or more second data flows are already sharing resources with a third data flow having a session group ID that differs from the first session group ID,
- if so, direct the traffic scheduler to prevent resources reserved for the one or more second data flows from being shared with the first data flow, and
- if not, direct the traffic scheduler to share resources reserved for the one or more second data flows with the first data flow.
A device comprising:
- a traffic scheduler having one or more resources for use in forwarding network traffic;
- a resource reservation engine configured to;
- determine that a first session group ID of a new data flow matches a second session group ID of one or more existing data flows,
- check if the one or more existing data flows are already sharing resources with another data flow that has a third session group ID different from the first session group ID, and if so, direct the traffic scheduler to prevent resources reserved for one or more existing data flows from being shared with the new data flow, and if not, direct the traffic scheduler to share resources reserved for one or more existing data flows with the new data flow.
A method comprising:
- receiving a request for a new data flow, the new data flow associated with a first session group (ID);
- determining that the first session group ID of the new data flow matches a second session group ID of one or more existing data flows;
- checking if the one or more existing data flows are a1ready sharing resources with another data flow that has a third session group ID different from the first session group ID;
- if so, preventing resources reserved for the one or more existing data flows from being shared with the new data flow;
- and
- if not sharing resources reserved for the one or more existing data flows with the new data flow.
An apparatus comprising:
- means for receiving a request for a new data flow, the new data flow associated with a first session group (ID);
- means for determining that the first session group ID of the new data flow matches a second session group ID of one or more existing data flows;
- means for checking if the one or more existing data flows are already sharing resources with another data flow that has a third session group ID different from the first session group ID;
- means for preventing resources reserved for the one or more existing data flows from being shared with the new data flow if so;
- and
- means for sharing resources reserved for one or more existing data flows with the new data flow if not.
Dependencies are inferred from the claim text.