Research Article
Differential Grey Wolf Load-Balanced Stochastic Bellman Deep Reinforced Resource Allocation in Fog Environment
Algorithm 2
Stochastic Bellman Gradient Deep Reinforcement Learning-based Resource Allocation.
| Input: Dataset “,” task “,” fog nodes “” | | Output: Energy minimized optimal resource allocation | (1) | Begin | (2) | For each dataset “” with task “” and fog nodes “” | | //State space | (3) | Load balancer acquires the input from fog environment “” | (4) | Mathematically formulate data size as in equation (16) | (5) | Mathematically formulate waiting time as in equation (17) | (6) | Mathematically formulate queue length as in equation (18) | | //Action space | (7) | For each action “” with the consolidated state “” | (8) | If task “” generated by fog node “” is executed locally | (9) | Then “” | (10) | Else “” | (11) | End if | (12) | If Task “” generated by fog node “” is executed on the host node | (13) | Then “” | (14) | Else “” | (15) | End if | (16) | If Task “” generated by fog node “” is executed by neighbor | (17) | Then“” | (18) | Else “” | (19) | End if | | //Reward function | (20) | For each action “” with the consolidated state “” and task “” generated by fog node “” | (21) | Total all the obtained rewards as in equation (19) | (22) | Measure stochastic bellman gradient optimality function as in equation (20) | (23) | End for | (24) | End for | (25) | End |
|