Research Article
Latency-Aware Computation Offloading for 5G Networks in Edge Computing
Algorithm 3
Shortest Time-to-Completion First with multiusers under low resource p-MEFP.
| | Inputs: E, U, Gi = (Ti, DPi), M, ri,s, fi, Li, | | Output: Optimal offloading strategy ofmin for multiusers | | (1) | List all optional offloading strategies OF = {of1, of2, …, ofN} | | (2) | for ofi in OF do | | (3) | for aj in ofido | | (4) | for ek in E do | | (5) | Create a set Uk for each user with aj = ek | | (6) | for ti in Ukdo | | (7) | Update the transfer rate ri,s of each tasks | | (8) | Call EFP algorithm | | (9) | if Decision(i, j) ! = 0 | | (10) | Compute migration probability possibility(j) | | (11) | random = Random(0, 1) | | (12) | if random ≥ possibility(j) and < Li | | (13) | Execute ti,j locally | | (14) | else | | (15) | Offload ti,j to edge server | | (16) | end if | | (17) | end if | | (18) | end for | | (19) | end for | | (20) | end for | | (21) | Compute the total execution delay Tifinish for computing offloading strategy ofi | | (22) | end for | | (23) | Select the offloading strategy ofmin with minimal execution delay |
|
