Utilizing reinforcement learning for serverless function tuning
One or more metrics associated with performance of a serverless function chain are received. The one or more metrics are used by a reinforcement learning (RL) model to tune the serverless function chain.
1. A method, comprising:
receiving one or more metrics associated with a computing performance of a serverless function chain; and
using a reinforcement learning (RL) model to tune the serverless function chain based on the one or more metrics and based in part on relationships between serverless functions of the serverless function chain, wherein to tune the serverless function chain comprises to improve the computing performance of the serverless function chain.
2. The method of claim 1 , wherein the serverless function chain comprises a first serverless function having an output that is an input of a second serverless function.
3. The method of claim 1 , wherein the RL model generates a tuning factor to tune the serverless function chain, the method further comprising:
receiving one or more subsequent metrics associated with the computing performance of the serverless function chain;
determining an impact the tuning factor has on the computing performance of the serverless function chain in view of a goal;
determining a new tuning factor in view of the impact; and
using the new tuning factor to tune the serverless function chain.
4. The method of claim 1 , wherein the serverless function chain comprises a plurality of containers.
5. The method of claim 1 , wherein the serverless function chain comprises a plurality of virtual machines.
6. The method of claim 1 , wherein to tune the serverless function chain comprises scaling a number of serverless functions of the serverless function chain that are active.
7. The method of claim 1 , wherein to tune the serverless function chain comprises adjusting an amount of computing resources allocated to serverless functions of the serverless function chain.
8. A system comprising:
a memory; and
a processing device, operatively coupled to the memory, to:
receive one or more metrics associated with a computing performance of a serverless function chain; and
use a reinforcement learning (RL) model to tune the serverless function chain based on the one or more metrics and based in part on relationships between serverless functions of the serverless function chain, wherein to tune the serverless function chain comprises to improve the computing performance of the serverless function chain.
9. The system of claim 8 , wherein the serverless function chain comprises a first serverless function having an output that is an input of a second serverless function.
10. The system of claim 8 , wherein the RL model generates a tuning factor to tune the serverless function chain, the processing device further to:
receive one or more subsequent metrics associated with the computing performance of the serverless function chain;
determine an impact the tuning factor has on the computing performance of the serverless function chain in view of a goal;
determine a new tuning factor in view of the impact; and
use the new tuning factor to tune the serverless function chain.
11. The system of claim 8 , wherein the serverless function chain comprises a plurality of containers.
12. The system of claim 8 , wherein the serverless function chain comprises a plurality of virtual machines.
13. The system of claim 8 , wherein to tune the serverless function chain comprises to scale a number of serverless functions of the serverless function chain that are active.
14. The system of claim 8 , wherein to tune the serverless function chain comprises to adjust an amount of computing resources allocated to serverless functions of the serverless function chain.
15. A non-transitory computer-readable storage medium including instructions that, when executed by a processing device, cause the processing device to:
receive one or more metrics associated with a computing performance of a serverless function chain; and
use a reinforcement learning (RL) model to tune the serverless function chain based on the one or more metrics and based in part on relationships between serverless functions of the serverless function chain, wherein to tune the serverless function chain comprises to improve the computing performance of the serverless function chain.
16. The non-transitory computer-readable storage medium of claim 15 , wherein the serverless function chain comprises a first serverless function having an output that is an input of a second serverless function.
17. The non-transitory computer-readable storage medium of claim 15 , wherein the RL model generates a tuning factor to tune the serverless function chain, the non-transitory computer-readable storage medium further comprising instructions that cause the processing device to:
receive one or more subsequent metrics associated with the computing performance of the serverless function chain;
determine an impact the tuning factor has on the computing performance of the serverless function chain in view of a goal;
determine a new tuning factor in view of the impact; and
use the new tuning factor to tune the serverless function chain.
18. The non-transitory computer-readable storage medium of claim 15 , wherein the serverless function chain comprises a plurality of containers.
19. The non-transitory computer-readable storage medium of claim 15 , wherein to tune the serverless function chain comprises to scale a number of serverless functions of the serverless function chain that are active.
20. The non-transitory computer-readable storage medium of claim 15 , wherein to tune the serverless function chain comprises to adjust an amount of computing resources allocated to serverless functions of the serverless function chain.