Loading patent…
AssigneeInternational Business Machines Corporation
InventorsIsrael Cidon, Inder S. Gopal, Roch A. Guerin
Estimating the term…
A fast bandwidth reservation method and system used in packet-switched networks that reserve bandwidth for a data transfer over links in a path between a source node, at least one intermediate node, and a destination node with minimal latency where the source node desires to send data to the destination node along the specific path using at least a minimum bandwidth and up to a maximum bandwidth. The goal of the method and system is to reserve bandwidth for a data transfer in the minimal amount of time. No link on the path between the source node and destination node has more bandwidth allocated for the source node than the minimum allocated for the previous forward links on the path. Links on the path between the source node and destination node have their bandwidth allocation for the data transfer adjusted based on the final minimum bandwidth allocated for the data transfer.
A method of requesting resources in a communications network with minimal latency, said method comprising the steps of:
A method of requesting resources in a communications network with minimal latency, said method comprising the steps of:
A method of requesting resources in a communications network with minimal latency, said method comprising the steps of:
A method of requesting resources in a communications network with minimal latency, said method comprising the steps of:
A method of requesting resources in a communications network with minimal latency, said method comprising the steps of:
A method of requesting resources in a communications network with minimal latency, said method comprising the steps of:
A system for requesting resources in a communications network with minimal latency having a path from a source node through at least one intermediate node towards a destination node, said nodes coupled by respective links and said request packet containing a request for resources to complete a requested transmission along the path to the destination node, said system comprising:
A system for requesting resources in a communications network with minimal latency having a path from a source node through at least one intermediate node towards a destination node, said nodes coupled by respective links and said request packet containing a minimum request for resources and a desired request for resources to complete a requested transmission along the path to the destination node where the desired request for resources is greater than or equal to the minimum request for resources, said system comprising:
A method of requesting resources in a communications network with minimal latency, said method comprising the steps of:
Dependencies are inferred from the claim text.
Drawings are temporarily unavailable. Open the PDF at USPTO
Transfer mode dependent, e.g. ATM: bandwidth control in ATM Networks, e.g. leaky bucket
Integrated services digital network, i.e. systems for transmission of different types of digitised signals, e.g. speech, data, telecentral,…:…
Resource management and allocation: bandwidth allocation
Bandwidth allocation: in-call negotiation
Term ended 16 Dec 2012 (estimated)
USPTO file history isn’t available for this patent, so only indexed dates are shown.
The present invention is embodied in packet-switched networks which reserve bandwidth, specifically, in packet-switched networks that have the ability to reserve bandwidth along a path between a source node and a destination node.
A fast reservation mechanism which has the ability to allocate bandwidth within a short period of time is desirable in fast packet-switched networks with high-speed bursty connections. In these networks, the high-speed bursty connections alternate between periods requiring a bandwidth allocation at peak and no bandwidth. In networks employing a static bandwidth allocation technique, as the alternating periods increase, the static bandwidth allocation technique becomes less efficient because the high-speed bursty connections generally require a bandwidth allocation close to their peak rate irrespective of how often they actually need it.
To resolve the bandwidth allocation problems associated with high-speed bursty connections, bandwidth reservation protocols have been presented. One bandwidth reservation protocol is described in U.S. Pat. No. 5,081,620 by Girard et al which is hereby incorporated by reference. This bandwidth reservation protocol sends a call packet containing a desired bit rate (DEB) from a source node to the destination node where the source node desires to send data at the desired bit rate to the destination node.
FIG. 1 is a block diagram of an exemplary data path using Automatic Network Routing (ANR).
FIGS. 2A-2C are block diagrams of Node 1 shown in FIG. 1 which depict two different exemplary configurations of the bandwidth allocation devices.
FIGS. 3A to 3I are listings of exemplary packet information used in the exemplary data path shown in FIG. 1.
An exemplary data path using Automatic Network Routing (ANR) is shown in FIG. 1. A brief overview of the invention is presented by reference to this exemplary data path. In this example, Node 0 desires to send data to Node 3 at a minimum data rate requiring a minimum bandwidth allocation, B.sub.-- min, and optimally at a desired maximum data rate, requiring a maximum bandwidth allocation, B.sub.-- max. Note that B.sub.-- min may equal B.sub.-- max. In this example, the data path from Node 0, the source node, to Node 3, the destination node, is through two intermediate nodes, namely Nodes 1 and 2. The present invention is directed to a path having at least one intermediate node. In the exemplary embodiment of the invention, a reservation request packet is generated by the source node, Node 0, that indicates the minimum bandwidth allocation, B.sub.-- min, and the desired maximum bandwidth allocation, B.sub.-- max. The reservation request packet is processed by bandwidth allocation devices in the path. A bandwidth allocation device uses the current bandwidth allocation and request for resources in the request packet to determine whether to reserve bandwidth for this request for resources.
Two exemplary configurations of bandwidth allocation devices are shown in FIG. 2. FIGS. 2A to 2B illustrate Node 1 shown in FIG. 1 with exemplary bandwidth allocation device configurations. FIG. 2A illustrates an exemplary configuration where the node has a single bandwidth allocation device which processes request packets which desire to reserve bandwidth on either of the outgoing links attached to Node 1, ANR link 12 and ANR link 21.
FIG. 2B illustrates an exemplary configuration where each outgoing link has a separate bandwidth allocation device (BAD). Outgoing ANR link 21 and ANR link 12 each has a BAD which processes request packets which desire to reserve bandwidth on the respective outgoing link. In addition, the BADs receive and process reply packets from the incoming ANR link 5 and ANR link 2. The BADs use the information in the reply packets to update the current bandwidth allocation for the outgoing links as appropriate.