Image Processing Reference
In-Depth Information
order to provide over the mobile devices the rendered content. In another point of
view, authors in [
23
] specified the difficulties of large loads in the channels, the
limitations in computation, and the battery life on mobile phones about free
viewpoint video (FVV)/free viewpoint TV (FTV). To avoid these drawbacks, a
cloud-based FVV/FTV rendering framework is proposed towards obtaining the
maximum QoE for users. A prototype resource allocation scheme is described
that correlates both render allocation based on end user QoE and rate allocation,
as a matter of channel rate, of texture, and depth quality. Results of the proposed
framework indicate the optimization of the video quality in standard mode of
mobile devices. Moreover, the research work in [
24
] considers a novel cloud
framework that allows scalable real-time 3D applications. This infrastructure
exploits a scalable-pipelined process that comprised of a virtualization server
network for running 3D virtual appliances, a graphics rendering network for
processing graphics workload with load balancing and a media streaming network
for transcoding rendered frames into H.264/MPEG-4 media.
In this context, this chapter is making progress beyond the current state-of-the-
art, by proposing a dynamic resource migration mechanism based on multi-
migrate-resume mechanisms, for resource/task migration of mobile users, and
efficient 3D content manipulation through the cloud-based network due to
processing limitations of the mobile devices. It uses a similar—but optimized—
methodology with the migration in [
14
], whereas it allows native resources that are
claimed by a mobile device to be offloaded both on a datacenter rack and to a
mobile device as a file/resource sharing datacenter. This chapter also elaborates on
the design and development of a modular scenario, where the proposed migration
scheduling policy algorithm for establishing partitionable resource/process execu-
tion, enables task execution reliability and guarantees the efficiency in the execu-
tion of mobile users
tasks/applications. The proposed scheme avoids any failures
when the resources, for running a task on a mobile device, are not sufficient.
'
7.3 Dynamic Cloud resource Migration Based
on Temporal and Capacity-Aware Policy
Computing-as-service is becoming the dominant paradigm in today
s computing
environments. Considering the delays caused by WAN processing mechanisms and
the potential lack of executing resources onto mobile devices, there is a great need
to encompass in the resource/task execution processes, the cloud resource migration
services. This service can encounter resource failures for reliable execution and can
take into consideration network-and-service oriented metrics such as bandwidth
and ping delays as well as migration/processing delays and manageability metrics
of the resources claimed by cloud. Although bandwidth and access will continue to
improve over time, the service and “transfer to device-back” latency is unlikely to
get improved. Therefore, the main need in such a case is the deployment of a
'
