Optimized network traffic patterns for co-located heterogeneous network attached accelerators

1. A method for optimizing a use of accelerators, the method comprising:

receiving, by one or more computer processors, a job request from an application, wherein the job request is associated with a machine learning request;

identifying, by the one or more computer processors, one or more accelerators on a network;

creating, by the one or more computer processors, a network mapping of the one or more accelerators based on a utilization component, wherein the one or more accelerators comprise a database accelerator, a web accelerator and a machine learning accelerator and the utilization component comprises user generated relationship rules based on pre-defined parameters associated with the one or more accelerators;

sending, by the one or more computer processors, a data request to a database server based on the job request by leveraging the database accelerator, wherein leveraging the database accelerator comprises:

submitting one or more data requests to the database accelerator, wherein the database accelerator is attached to the database server; and

receiving, by the one or more computer processors, data from the database accelerator and forwarding the data to the machine learning accelerator, wherein the machine learning accelerator completes the job request by processing the received data.

2. The method of claim 1 , further comprising:

identifying, by the one or more computer processors, the one or more accelerators based on broadcasting a discoverable protocol; and

receiving, by the one or more computer processors, an acknowledgement response from the one or more accelerators.

3. The method of claim 1 , wherein the relationship rules are dynamically adjusted by the user.

4. The method of claim 1 , wherein the utilization component contains a calculating module associated with creating the network mapping.

5. The method of claim 1 , wherein the network mapping further comprises an IP address, an SQL query instruction, and API code.

6. A computer program product for optimizing the use of accelerators, the computer program product comprising:

one or more computer readable storage devices and program instructions stored on the one or more computer readable storage devices, the stored program instructions comprising:

program instructions to receive a job request from an application, wherein the job request is associated with a machine learning request;

program instructions to identify one or more accelerators on a network;

program instructions to create a network mapping of the one or more accelerators based on a utilization component, wherein the one or more accelerators comprise a database accelerator, a web accelerator and a machine learning accelerator and the utilization component comprises user generated relationship rules based on pre-defined parameters associated with the one or more accelerators;

program instructions to send a data request to a database server based on the job request by leveraging the database accelerator, wherein leveraging the database accelerator comprises:

submitting one or more data requests to the database accelerator, wherein the database accelerator is attached to the database server; and

program instructions to receive data from the database accelerator and forward the data to the machine learning accelerator, wherein the machine learning accelerator completes the job request by processing the received data.

7. The computer program product of claim 6 , the stored program instructions further comprising:

program instructions to identify the one or more accelerators based on broadcasting a discoverable protocol; and

program instructions to receive an acknowledgement response from the one or more accelerators.

8. The computer program product of claim 6 , wherein the relationship rules are dynamically adjusted by the user.

9. The computer program product of claim 6 , wherein the utilization component contains a calculating module associated with creating the network mapping.

10. The computer program product of claim 6 , wherein the network mapping further comprises an IP address, an SQL query instruction, and API code.

11. A computer system for optimizing the use of accelerators, the computer system comprising:

one or more computer processors;

one or more computer readable storage devices; and

program instructions stored on the one or more computer readable storage devices for execution by at least one of the one or more computer processors, the stored program instructions comprising:

program instructions to receive a job request from an application, wherein the job request is associated with a machine learning request;

program instructions to identify one or more accelerators on a network;

program instructions to create a network mapping of the one or more accelerators based on a utilization component, wherein the one or more accelerators comprise a database accelerator, a web accelerator and a machine learning accelerator and the utilization component comprises user generated relationship rules based on pre-defined parameters associated with the one or more accelerators;

program instructions to send a data request to a database server based on the job request by leveraging the database accelerator, wherein leveraging the database accelerator comprises:

submitting one or more data requests to the database accelerator, wherein the database accelerator is attached to the database server; and

program instructions to receive data from the database accelerator and forward the data to the machine learning accelerator, wherein the machine learning accelerator completes the job request by processing the received data.

12. The computer system of claim 11 , the stored program instructions further comprising:

program instructions to identify the one or more accelerators based on broadcasting a discoverable protocol; and

program instructions to receive an acknowledgement response from the one or more accelerators.

13. The computer system of claim 11 , wherein the relationship rules are dynamically adjusted by the user.

14. The computer system of claim 11 , wherein the utilization component contains a calculating module associated with creating the network mapping.

15. The computer system of claim 11 , wherein the network mapping further comprises an IP address, an SQL query instruction, and API code.

16. The computer system of claim 11 , wherein leveraging the database accelerator comprises:

program instructions to submit one or more data request to the database accelerator, wherein the database accelerator is attached to the database server.