Method and system for automated migration of high performance computing application to serverless platform

An application migration system, comprising:

• a memory storing instructions;
• one or more communication interfaces;
• and
• one or more hardware processors coupled to the memory via the one or more communication interfaces, wherein the one or more hardware processors are configured by the instructions to:
• receive one or more applications that are to be migrated on a serverless platform and a configuration file associated with each application of the one or more applications, wherein the configuration file associated with each application of the one or more applications comprises a plurality of parameters;
• determine suitability of each application of the one or more applications for serverless platform migration by iteratively performing:
• determining whether a first application of the one or more applications is eligible for serverless platform migration based on the plurality of parameters of the first application using a machine learning based application qualification model;
• upon determining that the first application is eligible for serverless platform migration, computing an execution time of the first application when running on the serverless platform based on the plurality of parameters of the first application using a machine learning based runtime prediction model;
• determining whether the first application is executable on the serverless platform based, at least in part, on the computed execution time and one or more permissible limits, wherein the one or more permissible limits are accessed from the configuration file associated with the first application;
• upon determining that the first application is executable on the serverless platform, calculating a running cost of the first application on the serverless platform based, at least in part, on the computed execution time and a compute capacity of the serverless platform, wherein the compute capacity of the serverless platform is accessed from a serverless platform configuration file;
• storing the running cost of the first application in a system database;
• adding the first application in a suitable application queue;
• and
• identifying a second application in the one or more applications as the first application, until the suitability of all applications in the one or more applications is determined;
• and
• perform automatic migration of each application present in the suitable application queue to the serverless platform using an infrastructure automation engine, where each migrated application is configured to run on the serverless platform.

One or more non-transitory machine-readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors cause:

• receiving, by an application migration system, one or more applications that are to be migrated on a serverless platform and a configuration file associated with each application of the one or more applications, wherein the configuration file associated with each application of the one or more applications comprises a plurality of parameters;
• determining, by the application migration system, suitability of each application of the one or more applications for serverless platform migration by iteratively performing:
• determining, by the application migration system, whether a first application of the one or more applications is eligible for serverless platform migration based on the plurality of parameters of the first application using a machine learning based application qualification model;
• upon determining that the first application is eligible for serverless platform migration, computing, by the application migration system, an execution time of the first application when running on the serverless platform based on the plurality of parameters of the first application using a machine learning based runtime prediction model;
• determining, by the application migration system, whether the first application is executable on the serverless platform based, at least in part, on the computed execution time and one or more permissible limits, wherein the one or more permissible limits are accessed from the configuration file associated with the first application;
• upon determining that the first application is executable on the serverless platform, calculating, by the application migration system, a running cost of the first application on the serverless platform based, at least in part, on the computed execution time and a compute capacity of the serverless platform, wherein the compute capacity of the serverless platform is accessed from a serverless platform configuration file;
• storing, by the application migration system, the running cost of the first application in a system database;
• adding, by the application migration system, the first application in a suitable application queue;
• and
• identifying, by the application migration system, a second application in the one or more applications as the first application, until the suitability of all applications in the one or more applications is determined;
• and
• performing, by the application migration system, automatic migration of each application present in the suitable application queue to the serverless platform using an infrastructure automation engine, wherein each migrated application is configured to run on the serverless platform.