In order to minimize the number of server and prepare computational resources Ajiro [4] implemented a load-balancing, least loaded algorithm and compared it with classical FFD (First Fit Decreasing) problem.
The production and demand profiles mismatch limits the network hosting capacity, which still can be enhanced by implementing a load management.
Due to the vastly different processing times of PSC operations when using different methods, we allow the master process: (a) to make a choice between sequential versus parallel job processing and (b) implement a load balancing scheme for the slaves.
In the future, we plan to implement a load-dependent mapper by post-processing the output of the gain-based mapper.
Dhillon et al. [47] suggest to implement a load-dependent access scheme wherein uncoordinated strategy is for light load and coordinated strategy for heavy load.
To overcome this challenge, we implemented a scalable load balancing strategy suitable for multi-stepped execution based on the hierarchical load balancing framework (Zheng 2005; Mei et al. 2011) in the Charm++ runtime.
By having the mapper list randomly ordered, we implement a simple load balancing mechanism.
The compressive state implemented a 500 N vertical load; the flexion state implemented a 500 N vertical load and a 10 Nm moment of forward sagittal direction; the lateral bending state implemented a 500 N vertical load and a 10 Nm moment of right lateral direction.
Because departure can also depend on fuel load and fuel deposition rate (see e.g. Weber et al. 1998; Schaub et al. 2008), we also implemented a minimum departure fuel load of 15% LBM at each stopover.
