发明名称 Delaycast queue prioritization
摘要 Systems and methods are described for optimizing resource utilization in a communications network while also optimizing subscriber engagement with media content over the communications network. Requested content objects can be identified as delayable objects that can be queued for opportunistically delayed communication to both requesting and non-requesting subscribers. Queued delayed content objects are scored with an eye toward optimizing both subscriber engagement and utilization of opportunistically available communications link resources. For example, a storage manager calculates a likelihood that each subscriber will engage with the content if it is opportunistically delivered, and a scheduler calculates a priority order in which to queue each requested delayable content object. Content objects can then be multicast to the subscribers in priority order and with associated information that can be used by the subscribers to determine whether to locally store the content objects as they are opportunistically received.
申请公布号 US9154436(B2) 申请公布日期 2015.10.06
申请号 US201313830671 申请日期 2013.03.14
申请人 VIASAT INC. 发明人 Dankberg Mark D.;Newman Daniel M.;Hart Steven R.
分类号 H04L12/875;H04L12/18;H04L29/08 主分类号 H04L12/875
代理机构 Marsh Fischmann & Breyfogle LLP 代理人 Marsh Fischmann & Breyfogle LLP ;Sherwinter Daniel J.
主权项 1. A method for prioritizing communications of content objects in a communications infrastructure that provides sharing of a communications link over which a provider-side system is in communication with a plurality of subscriber-side systems, the method comprising: determining, by a provider-side system, that a content object received at the provider-side system is a delayable content object requested by at least one of a plurality of subscriber-side systems, the subscriber-side systems being associated with a plurality of subscriber clusters; calculating, by the provider-side system, a scheduler score for the delayable content object according to a set of link-level scalars; calculating, by the provider-side system, a set of storage scores for the delayable content object according to a set of storage-level scalars as applied to at least some of the subscriber clusters, wherein the provider-side system comprises a storage manager instance associated with each of the subscriber clusters, and each storage manager instance calculates the set of storage scores for each content object according to the set of storage-level scalars as applied to its associated subscriber cluster; assigning the delayable content object to a provider-side queue of content objects in a priority order according to the scheduler score of the delayable content object; and multicasting the content objects from the queue to the subscriber-side systems along with associated control data corresponding to the respective storage scores of the content objects and usable by the subscriber-side systems in determining whether to locally store the content objects.
地址 Carlsbad CA US