Distributed computing using communities

• US 9,268,615 B2
• Filed: 05/28/2010
• Issued: 02/23/2016
• Est. Priority Date: 05/28/2010
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Calsyn, Martin, Brandle, Alexander, Lyutsarev, Vassily, Heil, Andreas
• Primary Examiner(s): BAYARD, DJENANE M
• Application Number:
  US12/790,105
• Publication Number:
  US 20110295949A1
• Time in Patent Office: 2,097 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 9/5072 Grid computing

Abstract

Distributed computing using communities is described. In an embodiment computations in a distributed computing system are driven and controlled by a document storing a distributed computing graph, a graph layout view of that graph and visualization elements. For example, the document is replicated and synchronized at each of a plurality of entities in the distributed computing system. In examples a community may be drawn as a rectangle or other shape in the graph layout view and represents one or more computing resources in the distributed computing system. For example by placing graphical elements representing currently executing processes into the community on the graph layout view a user is able to ensure that those processes execute using the computing resources of the community. In examples communities may be nested and may have parameters specifying conditions which are to be met by the computing resources they represent.

Claims

1. A method at an entity in a distributed computing system comprising:

providing an input arranged to connect to the distributed computing system and access a distributed computing graph associated with an activity to be carried out in the distributed computing system, the distributed computing graph representing currently executing processes in the distributed computing system;

arranging a processor to provide a dynamic mapping between the distributed computing graph and a data structure at the entity holding a graph layout view of the distributed computing graph such that changes in the graph layout view are reflected in the distributed computing graph;

receiving user input adding a first community in the graph layout view the first community comprising computing resources in the distributed computing system;

receiving user input placing a first graphical element in the first community, the first graphical element representing at least a first one of the currently executing processes;

using the processor to update the distributed computing graph such that the first currently executing process runs on computing resources of the first community;

receiving user input placing a second graphical element outside all communities displayed on the graph layout view, the second graphical element representing at least a second one of the currently executing processes; and

using the processor to update the distributed computing graph such that the second currently executing process represented by the second graphical element runs on computing resources of a local machine.

2. A method as claimed in claim 1 at an entity remote of the computing resources of the first community.

3. A method as claimed in claim 1 which comprises storing metadata for the first community.

4. A method as claimed in claim 1 which comprises using the processor to update the distributed computing graph to automatically incorporate data formatting processes such that data is able to flow between the first community and other processes in the distributed computing graph.

5. A method as claimed in claim 1 which comprises receiving user input adding a second community to the graph layout view inside the first community.

6. A method as claimed in claim 1 which comprises storing parameters for a community the parameters specifying conditions to be met by computing resources in the community.

7. A method as claimed in claim 1 which comprises storing a community control document listing members of the community and actions which those members are able to carry out with respect to the community;

and making the community control document available to all members of the community.

8. A method as claimed in claim 7 which comprises monitoring a number of community members which are able to write to the community control document versus a number of community members which are able to read from the community control document and, on the basis of the monitored numbers, facilitating changes to the storage of the community control document that are transparent to the users in the affected community.

9. A method as claimed in claim 1 which comprises discovering a community by any of:

receiving an invitation message, participating in a peer to peer distributed directory system and checking a web directory service.

10. A method as claimed in claim 1 which comprises forming a community by creating and storing a community control document and announcing or inviting members to join.

11. A method as claimed in claim 1 which comprises receiving user input adding an extension to the first community that extension being code which is available to all members of the first community.

12. A method as claimed in claim 11 which comprises monitoring use of the extension and assigning reputation points to the extension on the basis of the monitored use.

13. A method as claimed in claim 1 which comprises storing the graph layout view, the distributed computing graph and visualization elements associated with the distributed computing graph as a document;

and replicating and synchronizing that document at a plurality of entities in the distributed computing system.

14. A method as claimed in claim 13 which comprises storing the document using a semi-structured property store.

15. A method as claimed in claim 13 which comprises storing the document using a hierarchy of rows in a relational database table.

16. A method as claimed in claim 15 which comprises synchronizing the document by updating rows of the relational database stored at different entities in the distributed computing system.

17. A method as claimed in claim 13 which comprises:

using a tree structure to store the document where nodes of the tree represent components of the document and have associated properties; and

storing each node of the tree and each property as a row in a relational database.

18. A graph engine at an entity in a distributed computing system, the graph engine comprising:

an input arranged to connect to the distributed computing system and access a distributed computing graph associated with an activity to be carried out in the distributed computing system, the distributed computing graph representing currently executing processes in the distributed computing system;

a processor arranged to provide a dynamic mapping between the distributed computing graph and a data structure at the entity holding a graph layout view of the distributed computing graph such that changes in the graph layout view are reflected in the distributed computing graph;

a graphical user interface arranged to receive user input adding a first community in the graph layout view the first community comprising computing resources in the distributed computing system;

the graphical user interface also being arranged to receive user input placing a first graphical element in the first community, the first graphical element representing at least a first one of the currently executing processes;

the processor being arranged to update the distributed computing graph such that the first currently executing process runs on computing resources of the first community;

the graphical user interface also being arranged to receive user input placing a second graphical element outside all communities displayed in the graph layout view, the second graphical element representing at least a second one of the currently executing processes; and

the processor being arranged to update the distributed computing graph such that the second currently executing process runs on computing resources of a local machine.

19. A graph engine as claimed in claim 18 having a memory arranged to store parameters specifying conditions that computing resources in the community are to meet.

20. A distributed computing system comprising:

a plurality of computing devices connected using a communications network, each computing device having a copy of a controlling document for controlling computations currently executing at the computing devices and which are part of the same activity, wherein a graphical element, representing a computation, being positioned outside each community represented in a distributed computing graph associated with the controlling document indicates that the computation is executed on a local machine;

a replication mechanism arranged to replicate and synchronize the controlling document at each computing device; and

a graph engine provided at each of the computing devices and arranged to manage any computations of the activity which are currently executing at that computing device using the controlling document.