Image Processing Reference
In-Depth Information
Other research approaches related to the performance of dynamic resource
allocation policies had led to the development of a computing framework [
10
],
which considers the countable and measureable parameters that will affect task
allocation. The idea of dynamically switching between local resources and remote
resources for serving availability and direct access to the requested resources (often
referred as multi-foraging behavior) has shed light on many research works [
11
-
13
]. Authors in [
11
] address this problem, by using the CloneCloud approach [
14
]
of a smart and efficient architecture for the seamless use of ambient computation to
augment mobile device applications, off-loading the right portion of their execution
onto device clones operating in a computational cloud. Authors in [
14
] statically
partition service tasks and resources between client and server portions, whereas in
a later stage the service is reassembled on the mobile device. The spine of the
proposal in [
14
] is based on a cloud-augmented execution, using a cloned VM
image as a powerful virtual device. This approach has many vulnerabilities as it has
to take into consideration the resources of each cloud rack, depending on the
expected workload and execution conditions (CPU speed, network performance).
Similar work, conducted on suspend-migrate-resume mechanisms [
15
] for
application-specific task migration, takes into account processing diversities of
the application but not the thread processing characteristics, thus presenting fore-
casting obstacles in safely determining the overall application execution and ser-
vice time. Authors in [
16
] address the resource poverty issue showing the obstacles
for many applications that typically require processing resources; whereas authors
address at the same time hardware capabilities for introducing resource hungry
applications, which are in need of ample computing resources such as face recog-
nition, speech recognition, and language translation on the move. In particular, the
application task and offloading process consist of other aspects, in terms of perfor-
mance and quality. In case where the resources on data centers or location-oriented
cloud racks are not enough to serve a certain resource sharing procedure resulting in
failures, there are schemes that aim to face the associated resource failures in the
context of data centers and are well addressed in recent literature [
17
,
18
]. However,
these schemes do not consider the cloud-to-cloud (in the context of data center to
data center) and the cloud-to-device resource migration, which plays an important
role in the resource manipulation and offloading procedure, in case where cloud
resources and cloud service requirements are not met. Authors in [
17
] provide a
manageable solution, according to the failure rates of servers in a large-scales
datacenter, whereas at the same time authors attempt to classify them, using a
range of criteria. However, these criteria along with the criteria set in [
19
-
21
]do
not include servers
communications diversities in the communication process with
'
mobile users
claims, as well as the utilization of memory and capacity of each
cloud terminal in the rack.
Towards considering the cloud context-aware requirements that presented
above, another related challenge concerns the real-time 3D video rendering in
mobile devices, where the challenge here is the need for large amount of network
and computing requirements (i.e., bandwidth, CPU). To address this issue, authors
in [
22
] propose a framework that supports remote rendering to 3D video streams, in
'
