Moving a portion of a streaming application to a public cloud based on sensitive data

Abstract

A streams manager determines which portions of a streaming application process sensitive data, and when performance of the streaming application needs to be increased, selects based on the sensitive data which portion(s) of the streaming application can be moved to a public cloud. The streams manager then interacts with the public cloud manager to move the selected portion(s) of the streaming application to the public cloud. By taking sensitive data into account, the streams manager protects the integrity of sensitive data while still taking advantage of the additional resources available in a public cloud.

1 - 8 . (canceled) 
     
     
         9 . A computer-implemented method executed by at least one processor for executing a streaming application, the method comprising:
 executing the streaming application comprising a flow graph that includes a plurality of processing elements that process a plurality of data tuples;   monitoring performance of the plurality of processing elements in the flow graph;   identifying which of the plurality of processing elements in the flow graph process sensitive data;   defining at least one sensitive data split criterion that specifies how to split the flow graph based on sensitive data so at least a portion of the flow graph can be moved to a public cloud in a manner that assures the sensitive data is not in the public cloud; and   selecting based on the at least one sensitive data split criterion at least one portion of the flow graph to move to the public cloud when the monitored performance indicates performance of the streaming application needs to be improved, and in response, moving the selected at least one portion of the flow graph to the public cloud.   
     
     
         10 . The method of  claim 9  wherein the at least one sensitive data split criterion specifies to move to the public cloud connected processing elements in the flow graph that do not process sensitive data. 
     
     
         11 . The method of  claim 9  wherein the at least one sensitive data split criterion specifies to move to the public cloud individual processing elements in the flow graph that do not process sensitive data. 
     
     
         12 . The method of  claim 9  wherein the at least one sensitive data split criterion specifies to move to the public cloud connected processing elements in the flow graph that process sensitive data that can have a data input encrypted and a data output decrypted. 
     
     
         13 . The method of  claim 9  wherein the at least one sensitive data split criterion specifies to unfuse one of the plurality of processing elements into a plurality of operators so at least one of the plurality of operators that does not process sensitive data is moved to the public cloud while at least one of the plurality of operators that process sensitive data is not moved to the public cloud. 
     
     
         14 . The method of  claim 9  wherein moving the selected at least one portion of the flow graph to the public cloud is performed by requesting a public cloud manager to provision at least one virtual machine with logic to implement the selected at least one portion of the flow graph, and when the public cloud manager provisions the at least one virtual machine, the streams manager modifies the flow graph to include the at least one virtual machine in the flow graph of the streaming application. 
     
     
         15 . The method of  claim 9  wherein determining when the monitored performance indicates performance of the streaming application needs to be improved is performed by comparing current performance of the streaming application to at least one performance threshold. 
     
     
         16 . The method of  claim 9  wherein determining when the performance of the streaming application needs to be improved is performed by comparing current performance of the streaming application to historical performance of the streaming application. 
     
     
         17 . A computer-implemented method executed by at least one processor for executing a streaming application, the method comprising:
 executing the streaming application comprising a flow graph that includes a plurality of processing elements that process a plurality of data tuples;   monitoring performance of the plurality of processing elements in the flow graph;   identifying which of the plurality of processing elements in the flow graph process sensitive data;   defining a plurality of sensitive data split criteria that specify how to split the flow graph based on sensitive data so at least a portion of the flow graph can be moved to a public cloud in a manner that assures the sensitive data is not in the public cloud, the sensitive split criteria including:
 specifying to move to the public cloud connected processing elements in the flow graph that do not process sensitive data; 
 specifying to move to the public cloud individual processing elements in the flow graph that do not process sensitive data; 
 specifying to move to the public cloud connected processing elements in the flow graph that process sensitive data that can have a data input encrypted and a data output decrypted; and 
 specifying to unfuse one of the plurality of processing elements into a plurality of operators so at least one of the plurality of operators that does not process sensitive data is moved to the public cloud while at least one of the plurality of operators that process sensitive data is not moved to the public cloud; 
   selecting based on the sensitive data split criteria at least one portion of the flow graph to move to the public cloud when the monitored performance indicates performance of the streaming application needs to be improved by comparing current performance of the streaming application to at least one performance threshold and by comparing current performance of the streaming application to historical performance of the streaming application; and   moving the selected at least one portion of the flow graph to the public cloud by requesting a public cloud manager to provision at least one virtual machine with logic to implement the selected at least one portion of the flow graph, and when the public cloud manager provisions the at least one virtual machine, the streams manager modifies the flow graph to include the at least one virtual machine in the flow graph of the streaming application.