Method and apparatus for network bandwidth allocation
Methods and apparatus for assigning bandwidth to a network based on two or more different service levels. In an exemplary embodiment, video on-demand (VOD) session requests within a service group of an HFC cable network are assigned bandwidth such that maximum opportunity is given to grant HD VOD request among SD VOD requests, while meeting the goal of minimizing service disruption by spreading the sessions over multiple QAM channels, and limiting the maximum number of active VOD sessions of any kind.
1. Network apparatus configured to selectively allocate bandwidth within a plurality of radio frequency (RF) channels configured to exist on a network, said RF channels comprising one or more service groups, said apparatus comprising:
a processor;
a storage device in operative communication with said processor;
a plurality of units configured to selectively modulate data onto said RF channels; and
a session manager program configured to run on said processor and to receive a plurality of on-demand (OD) session requests from other entities connected to said network, said plurality of requests comprising a first plurality of requests for high definition (HD) data, and a second plurality of requests for standard definition (SD) data, said session manager further being configured to:
identify each request as being one of said first or second plurality;
limit a number of HD sessions currently in process on said channels via a failure to service requests for HD sessions in excess of a predetermined HD session limit;
limit a number of SD sessions currently in process on said channels via a failure to service requests for SD sessions in excess of a predetermined SD session limit; and
for said HD and SD sessions which do not exceed said respective HD and SD session limits:
compare an HD bandwidth utilized by at least one of said channels against a predetermined bandwidth threshold for HD sessions;
when said HD bandwidth is less than said predetermined bandwidth threshold for HD sessions, allocate said request via said at least one of said channels in accordance with a Least-Loaded algorithm; and
when said HD bandwidth exceeds said predetermined bandwidth threshold for HD sessions, allocate said request via said at least one of said channels in accordance with a Most-Loaded algorithm.
2. A method of allocating channel bandwidth between standard definition (SD) and high definition (HD) sessions in a content distribution network configured for on-demand (OD) content delivery, said channel bandwidth comprising a plurality of substantially similar channels formed within a service group, said method comprising:
receiving a plurality of OD session requests from other entities connected to said network, said plurality of requests comprising a first plurality of requests for HD data, and a second plurality of requests for SD data;
evaluating existing bandwidth allocation on said channels of said service group in relation to a first predetermined allocation variable, said variable comprising information relating to a maximum number of SD session permitted and a maximum number of HD sessions permitted for that service group; and
when said evaluation indicates said maximum number of SD and said maximum number of HD sessions are not violated, identifying one or more of said channels as having a lowest amount of used bandwidth, allocating at least a portion of said requests to one or more of said identified channels based at least in part on a pre-determined threshold, when said existing bandwidth exceeds said threshold, identifying a second one or more of said channels as having a highest amount of used bandwidth yet being able to accommodate said portion of said requests, and allocating at least said portion of said requests to one or more of said identified channels;
wherein said act of allocating comprises assigning bandwidth such that maximum opportunity is given to grant HD OD requests among SD OD requests.
3. The method of claim 2 , further comprising providing a service manager configured to:
minimize service disruption by spreading said sessions associated with said requests over multiple quadrature amplitude modulated (QAM) channels; and
limit a maximum number of active OD sessions associated with both said SD and HD sessions.
4. The method of claim 3 , wherein said act of spreading said sessions associated with said requests over multiple QAM channels comprises assigning bandwidth so as to spread said sessions associated with at least said SD requests over multiple QAM channels while simultaneously minimizing stranded or unused bandwidth within said QAM channels so as to minimize an impact of QAM failures on said sessions.
5. The method of claim 2 , further comprising predicting an estimated average hold time of an OD session request based at least in part on a length of content requested.
6. The method of claim 2 , further comprising predicting an estimated average hold time of a OD session request based at least in part on whether said OD session request comprises an HD session or SD session.
7. The method of claim 2 , further comprising predicting whether a next session request is an HD session request or an SD session request based at least in part on HD digital set top box penetration rates, demographics, and content availability.
8. The network apparatus of claim 1 , wherein said session manager is further configured to predict an estimated average hold time of a session request based at least in part on a length of content requested.
9. The network apparatus of claim 1 , wherein said session manager is further configured to predict an estimated average hold time of a session request based at least in part on an average of a prior moving time period.
10. A computer-readable storage apparatus comprising a non-transitory storage medium with at least one program stored thereon, said at least one program comprising a plurality of instructions which are configured to, when executed, cause a network apparatus connected to a network to:
receive a plurality of on-demand (OD) session requests from other entities connected to said network, said plurality of requests comprising a first plurality of requests for high definition (HD) data, and a second plurality of requests for standard definition (SD) data;
identify each request as being one of said first or second plurality;
limit a number of HD sessions currently in process on said channels by via a failure to service requests for HD sessions in excess of a predetermined HD session limit;
limit a number of SD sessions currently in process on said channels by via a failure to service requests for SD sessions in excess of a predetermined SD session limit; and
for said HD and SD sessions which do not exceed said respective HD and SD session limits:
compare a HD bandwidth utilized by at least one of said channels against a predetermined bandwidth threshold for HD sessions;
when said HD bandwidth is less than said predetermined bandwidth threshold for HD sessions, allocate said request via said at least one of said channels in accordance with a Least-Loaded algorithm; and
when said HD bandwidth exceeds said predetermined bandwidth threshold for HD sessions, allocate said request via said at least one of said channels in accordance with a Most-Loaded algorithm.
11. The computer-readable storage apparatus of claim 10 , wherein said plurality of instructions are further configured to, when executed, predict an estimated average hold time of a session request based at least in part on a length of content requested.
12. The computer-readable storage apparatus of claim 10 , wherein said plurality of instructions are further configured to, when executed, predict an estimated average hold time of a session request based at least in part on an average of a prior moving time period.
13. The computer-readable storage apparatus of claim 10 , wherein said plurality of instructions are further configured to, when executed:
minimize service disruption via an allocation of said HD and SD sessions associated with said plurality of requests over multiple quadrature amplitude modulated (QAM) channels; and
limit a maximum number of active OD sessions associated with both said SD and HD sessions.
14. The computer-readable storage apparatus of claim 13 , wherein said HD and SD sessions associated with said plurality of requests are spread over multiple QAM channels via an assignment of bandwidth so as to spread said sessions associated with at least said SD requests over multiple QAM channels, and a simultaneously minimization of stranded or unused bandwidth within said QAM as to minimize an impact of QAM failures on said sessions.
15. The computer-readable storage apparatus of claim 10 , wherein said plurality of instructions are further configured to, when executed, when it is determined that said allocation shall not be performed, send a notification message to a device associated with said request.
16. The computer-readable storage apparatus of claim 10 , wherein said Least-Loaded algorithm comprises a grant of bandwidth on a channel with a most unallocated bandwidth available, and said Most-Loaded algorithm comprises a grant of bandwidth on a channel that has a least unallocated bandwidth available yet still able to accommodate said request.