Information Technology Reference
In-Depth Information
Table 1.
(
continued
)
Market-
Oriented-
Hierarchical
Scheduling
Minimize the
total execution
cost
Virtual
clusters
Total cost,
CPU time
Cloud Envi-
ronment
Cloud-
Sim
Concurrent
Level based
Workflow
Schedul-
ing(CLWS)
Decrease the
time pieces and
optimize the
total executing
cost.
Cost with
user desig-
nated
Dependency
mode
Cloud Envi-
ronment
Other
Adaptive
Hybrid Heu-
ristic schedul-
ing algorithm
Budget,
deadline
and execu-
tion cost
Identifie dy-
namic schedul-
ing approaches
Dependency
mode
Cloud Envi-
ronment
Other
As Table 1 shows, we make a comprehensive comparison with these algorithms,
and make analysis based scheduling model, scheduling parameter, scheduling result,
the tools and the environment they used. We can find that all algorithms used above
are cloud-based environment. Most of algorithms are batch-based and dependency-
based, and the simulation based CloudSim or other simulators. The batch-based is
clustering some relatively separate instances to dispatch; and the dependency-based is
that the single instance composed of multi-activity instances; and what the dispatch of
Virtual clusters is that using integrated tasks and virtual machines to define the local
task and virtual machine list which are the input of scheduling algorithm. Most of
algorithms focus on the execution cost and time which are the two hot research areas.
In addition, all the scheduling results of algorithms have better performance than the
non-optimized algorithms.
4
Typical Tools for Cloud Workflow Scheduling Research
CloudSim
1
4.1
Cloud computing can satisfy the different service requests with different configura-
tion, deployment condition and service resources of various user at different time
point. With the influence of multidimensional factors, it is unreality to test with dif-
ferent parameters in actual cloud computing center. So CloudSim, WorkflowSim and
so on have been used for simulating with cloud workflow scheduling.
CloudSim is a toolkit (library) for simulation of cloud computing scenarios. It pro-
vides basic classes for describing data centers, virtual machines, applications, users,
computational resources, and policies for management of diverse parts of the system
(e.g., scheduling and provisioning) [18].
1
http://www.cloudbus.org/cloudsim
