Method and a system for predicting a combination of optimal and stable instances

1 . A method for predicting a combination of optimal and stable instances, the method being implemented by at least one processor, the method comprising:

receiving, by the at least one processor via a communication interface, a configuration information of an application for execution by the combination of optimal and stable instances, wherein the combination of optimal and stable instances is hosted by a host platform, wherein “optimal” refers to instances selected based on performance and cost factors, and “stable” refers to instances selected based on historical reliability and availability factors;

identifying, by the at least one processor, at least one parameter related to the application; identifying, by the at least one processor, at least one optimal instance based on the identified at least one parameter, wherein the at least one optimal instance comprises at least one optimal spot instance; fetching, by the at least one processor, historical performance data of at least one historical spot instance from the host platform; predicting, by the at least one processor, a stability score for the at least one optimal spot instance based on at least the historical performance data of the at least one historical spot instance;

predicting, by the at least one processor, an intermediate set of at least one optimal and stable spot instance from the at least one optimal spot instance based on the stability score of the at least one optimal spot instance;

predicting, by the at least one processor, the combination of optimal and stable instances, wherein the combination of optimal and stable instances is based on at least on the cost factors, and based on at least one of the intermediate set of at least one optimal and stable spot instance and a set of optimal on-demand instances, wherein the optimal on-demand instances, different from the optimal and stable instances, which are provisioned at fixed cost independent of bidding or historical reliability;

presenting the combination of optimal and stable instances to a user via the communication interface for selection; and

executing the application by the selected combination of optimal and stable instances.

2 . The method according to claim 1 , wherein the at least one parameter includes at least a size, a computation complexity, and a memory.

3 . The method according to claim 1 , wherein each of the at least one parameter is assigned a weightage.

4 . The method according to claim 3 , wherein the combination of optimal and stable instances is predicted based on the weightage.

5 . The method according to claim 1 , wherein the historical performance data of the at least one spot instance comprises a spot instance interruption frequency of the at least one spot instance for a pre-defined period of time.

6 . The method according to claim 1 , wherein the at least one processor implements a machine learning model, wherein the machine learning model is trained by the at least one processor.

7 . The method according to claim 6 , wherein the training of the machine learning model by the at least one processor, comprises:

obtaining, by the at least one processor via the communication interface, a user feedback information of the combination of optimal and stable instances from at least one user; and

updating, by the at least one processor, the machine learning model implemented by the at least one processor based on the user feedback information and the combination of optimal and stable instances.

8 . A computing device for predicting a combination of optimal and stable instances, the computing device comprising:

a processor;

a communication interface;

a memory unit connected to the processor and the communication interface,

wherein the processor is configured to:

receive, via the communication interface, a configuration information of an application for execution by the combination of optimal and stable instances, wherein the combination of optimal and stable instances is hosted by a host platform, wherein “optimal” refers to instances selected based on performance and cost factors, and “stable” refers to instances selected based on historical reliability and availability factors;

identify at least one parameter related to the application;

identify at least one optimal instance based on the identified at least one parameter, wherein the at least one optimal instance comprises at least one optimal spot instance;

fetch historical performance data of at least one spot instance from the host platform;

predict a stability score for the at least one optimal spot instance based on at least the historical performance data of the at least one spot instance;

predict an intermediate set of at least one optimal and stable spot instance from the at least one optimal spot instance based on the stability score of the at least one optimal spot instance;

predict the combination of optimal and stable instances, wherein the combination of optimal and stable instances is based on at least on the cost factors, and is based on at least one of the intermediate set of at least one optimal and stable spot instance and a set of optimal on-demand instances, wherein the optimal on-demand instances, different from the optimal and stable instances, which are provisioned at fixed cost independent of bidding or historical reliability;

present the combination of optimal and stable instances to a user via the communication interface for selection; and

execute the application by the selected combination of optimal and stable instances.

9 . The computing device according to claim 8 , wherein the at least one parameter includes at least a size, a computation complexity, and a memory.

10 . The computing device according to claim 8 , wherein each of the at least one parameter is assigned a weightage.

11 . The computing device according to claim 10 , wherein the combination of optimal and stable instances is predicted based on the weightage.

12 . The computing device according to claim 8 , wherein the historical performance data of the at least one spot instance comprises a spot instance interruption frequency of the at least one spot instance for a pre-defined period of time.

13 . The computing device according to claim 8 , wherein the processor implements a machine learning model, wherein the machine learning model is trained by the processor.

14 . The computing device according to claim 13 , wherein the processor for training of the machine learning model, is further configured to:

obtain, via the communication interface, a user feedback information of the combination of optimal and stable instances from at least one user; and

update the machine learning model implemented by the processor based on the user feedback information and the combination of optimal and stable instances.

15 . A non-transitory computer readable storage medium storing instructions for predicting a combination of optimal and stable instances, the storage medium comprising executable code which, when executed by a processor, causes the processor to:

receive via a communication interface, a configuration information of an application for execution by the combination of optimal and stable instances, wherein the combination of optimal and stable instances is hosted by a host platform, wherein “optimal” refers to instances selected based on performance and cost factors, and “stable” refers to instances selected based on historical reliability and availability factors;

identify at least one parameter related to the application;

identify at least one optimal instance based on the identified at least one parameter, wherein the at least one optimal instance comprises at least one optimal spot instance;

fetch historical performance data of at least one spot instance from the host platform;

predict a stability score for the at least one optimal spot instance based on at least the historical performance data of the at least one spot instance;

predict an intermediate set of at least one optimal and stable spot instance from the at least one optimal spot instance based on the stability score of the at least one optimal spot instance;

predict the combination of optimal and stable instances, wherein the combination of optimal and stable instances is based the cost factors, and is based on at least one of the intermediate set of at least one optimal and stable spot instance and a set of optimal on-demand instances, wherein the optimal on-demand instances, different from the optimal and stable instances, which are provisioned at fixed cost independent of bidding or historical reliability;

present the combination of optimal and stable instances to a user via the communication interface for selection; and

execute the application by the selected combination of optimal and stable instances.

16 . The non-transitory computer readable storage medium according to claim 15 , wherein the at least one parameter includes at least a size, a computation complexity, and a memory, and each of the at least one parameter is assigned a weightage.

17 . The non-transitory computer readable storage medium according to claim 16 , wherein the combination of optimal and stable instances is predicted based on the weightage.

18 . The non-transitory computer readable storage medium according to claim 15 , wherein the historical performance data of the at least one spot instance comprises a spot instance interruption frequency of the at least one spot instance for a pre-defined period of time.

19 . The non-transitory computer readable storage medium according to claim 15 , wherein when executed by the processor, the executable code further causes the processor to implement a machine learning model, wherein the machine learning model is trained by the processor.

20 . The non-transitory computer readable storage medium according to claim 19 , wherein when executed by the processor, the executable code further causes the processor to train the machine learning model which is further configured to:

obtain, via the communication interface, a user feedback information of the combination of optimal and stable instances from at least one user; and

update, the machine learning model implemented by the processor based on the user feedback information and the combination of optimal and stable instances.