In order to ensure the I/O efficiency of the cloud platform, according to the data of the primary node and the data storage information from the node, the mapping module allocates the data transfer task to the core node of the communication ability.
This is depicted as a fundamental challenge by [127], which states that "in most studies, data transfer between tasks is not directly considered, data uploading and downloading are assumed as part of task execution".
In addition, data transfer between tasks is not directly considered in most existing studies, thus being assumed as part of task execution.
Accordingly, if request M n is accepted, then at least one light-path l is assigned to allow data transfer between tasks t and t ′ assigned respectively to MEC-DC locations u and u ′. sumlimits_{n in N}sumlimits_{t in T_{n}} x_{u}^{t} leq M times y_{u}, ,,, u in H_{m}.
In such workflows, when scheduling tasks to distributed resources, the data transfers between tasks are not a negligible factor and may comprise a significant portion of the total execution time and cost.
Dependencies between tasks imply that there can be heavy communication induced by data transfers between tasks executed on separate nodes.
Similar to default offloading, the exclusive mode is more useful if the data transfer and kernel tasks can be aggregated to be launched from the minimum set of host CPU cores.
We consider that tasks are computationally intensive, that is, the input data transfer for each task is negligible.
Moreover, there is no workflow characterization in terms of data transfers and task costs.
