Software Application Striping

• US 20090198792A1
• Filed: 12/18/2008
• Published: 08/06/2009
• Est. Priority Date: 02/05/2008
• Status: Active Grant

• Current Assignee: Nant Holdings IP, LLC (NantWorks, LLC)
• Original Assignee: Raptor Networks Technology Inc. (Fonon Corp.)
• Inventors: Wittenschlaeger, Thomas
• Granted Patent: US 7,603,428 B2
• Time in Patent Office: Days
• US Class Current: 709/215
• CPC Class Codes: G06F 15/16 Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs coordinating program control therefor G06F9/52; in regulating and control system G05B
• CPC Class Codes: G06F 9/54 Interprogram communication

Abstract

A distributed computing system comprising networking infrastructure and methods of executing an application on the distributed computing system is presented. Interconnected networking nodes offering available computing resources form a network fabric. The computing resources can be allocated from the networking nodes, including available processing cores or memory elements located on the networking nodes. A software application can be stored in a system memory comprising memory elements allocated from the nodes. The software application can be disaggregated into a plurality of executable portions that are striped across the allocated processing cores by assigning each core a portion to execute. When the cores are authenticated with respect to their portions, the cores are allowed to execute the portions by accessing the system memory over the fabric. While executing the software application, the networking nodes having the allocated cores concurrently forward packets through the fabric.

Claims

1. A method of forming a distributed computing platform, the method comprising:

providing a network fabric of interconnected networking nodes where each node comprises a memory element and a multi-core processor;

allocating a first and a second processing core from different networking nodes for use by a software application;

allocating a system memory from at least some of the nodes'"'"' memory elements where the system memory is addressable by the first processing core and the second processing cores over the fabric;

storing the software application in the system memory;

striping a plurality of executable portions of the software application across the first and the second processing cores by assigning one or more of the portions to each core according to a secured function;

allowing the first and the second cores to each execute their respective portions by accessing the portions from the system memory over the fabric while, concurrently, allowing the networking nodes having the cores to forward packets from a first device through the fabric to a second device.

2. The method of claim 1, further comprising the first processor core storing instructions of one of the plurality of executable portions in a local cache.

3. The method of claim 1, wherein the first processing core and the second processing core are geographically separated by at least one kilometer.

4. The method of claim 1, wherein the networking nodes comprise layer two switches.

5. The method of claim 1, further comprising establishing a memory map of the system memory and disseminating the memory map to the networking nodes having the first and the second processing cores.

6. The method of claim 1, wherein the step of storing the software application includes storing the software application in a first and a second element of the system memory, respectively, where the first and second elements are located on different networking nodes.

7. The method of claim 1, wherein at least one of the plurality of portions comprises machine code executable by the first processing core.

8. The method of claim 1, wherein at least one of the network nodes executes at least two different executable portions of the software application.

9. The method of claim 8, wherein the at least one of the network nodes executes at least three different portions of the software application.

10. The method of claim 1, wherein at least two of the executable portions are mirrored portions.

11. The method of claim 10, further comprising recovering from a loss of connectivity with the second processing core by activating the first processing core.

12. The method of claim 1, wherein at least two of the executable portions are different and comprise overlapping code segments.

13. The method of claim 1, further comprising the first and the second processing cores storing results of execution of the portions in the system memory over the fabric.

14. The method of claim 13, further comprising aggregating the results from the system memory and presenting the results locally to an individual.

15. The method of claim 1, further comprising encrypting the plurality of portions.

16. The method of claim 1, wherein the step of allowing the cores to execute includes authenticating the first and the second processing cores with respect to accessing their respective portions.

17. The method of claim 1, further comprising partitioning the system memory into a first and a second segment that are for exclusive use by the first and the second processing core, respectively.