ACEA: A Queueing Model-Based Elastic Scaling Algorithm for Container Cluster

Algorithm 1

The algorithm of ACEA.

Input: achievement rate of historical tasks , task processing interval , number of containers in the current cluster .

Output: indicator information for container cluster, volume of autoscaling, mode of adaptive scaling.

1. global var , Fitness; //Fitness: real-time comprehensive resource utilization rate of cluster.

2. const var , , ;// : maximum waiting time of tasks.

3. define func getConfig(): ;

4. Input

5. par[Pnum]=

6.  PSOInit(Pnum): ParticleInit(Pnum), ParticleEvaluate();

7. Fitness = Wq(getUall(s), getUlimit());

8. PSORun():

9. For each par.

10.  ParticleUpdate();

11.  ParticleEvaluate();

12. End For;

13. For Iteration times do.

14. If ():

15.  Contract(service);

16. End if;

17. If(()|| ):

18.  Expand(service);

19. End if;

20. If():

21.  PSOshowresult();

22. End if;

23. End For;