Method of hybrid message passing with shared memory

Patent metadata

Field	Value
Publication number	US-10069779-B2
Application number	US-201313849716-A
Country	US
Kind code	B2
Filing date	Mar 25, 2013
Priority date	Mar 25, 2013
Publication date	Sep 4, 2018
Grant date	Sep 4, 2018

Abstract

A method of communicating data between at least two coprocessors in a single computing environment wherein one of the coprocessors is a transmitting processor and the other of the coprocessors is a receiving processor and the single computing environment includes a memory accessible by the coprocessors includes, among other things, determining in the transmitting processor the size of data to be communicated, attaching the data to a message and transmitting the message to the receiving processor if the size of the data is below or equal to a predetermined threshold value, and transmitting to the receiving processor a pointer that points to a location of the data in the memory if the size of the data is above the predetermined threshold value.

First claim

What is claimed is:

1. A method of communicating data between coprocessors, comprising:
determining, at a first processor, whether data to be communicated exceeds a an adaptable size threshold, wherein the adaptable size threshold is based at least in part on a network latency;
in response to determining that the data is at least one of less than or equal to the predetermined size threshold, including the data in a message;
in response to determining that the data exceeds the predetermined size threshold, including a unique identifier associated with the data in the message;
determining whether there is metadata associated with the data;
in response to determining that there is metadata associated with the data, including the metadata in the message; and
transmitting the message to a second processor.

2. The method of claim 1 , further comprising moving the data to a memory location accessible by the first processor and the second processor, in response to determining that the data exceeds the threshold.

3. The method of claim 2 , wherein the unique identifier is prepended to the data in the memory location accessible by the first processor and the second processor.

4. The method of claim 1 , further comprising including a pointer in the message, in response to determining that the data exceeds the threshold, wherein the pointer includes an address of a memory location accessible by the first processor and the second processor.

5. The method of claim 4 , wherein the unique identifier includes a timestamp.

6. The method of claim 1 , wherein the network latency is determined from a timestamp included with the data.

7. The method of claim 1 , wherein the network latency is based at least in part on a transmit time between the first processor and the second processor.

8. The method of claim 1 , wherein the first processor transmits the message to the second processor and a third processor.

9. The method of claim 1 , wherein the first processor and the second processor are included in a single computing environment.

10. A method of efficiently communicating data between coprocessors in a single computing environment, comprising:
determining, at a first processor, whether data to be communicated exceeds an adaptable size threshold, wherein the adaptable size threshold is based at least in part on a network latency;
in response to determining that the data is at least one of less than or equal to the adaptable size threshold, including the data in a message;
in response to determining that the data exceeds the adaptable size threshold, determining if the data already exists in a memory location accessible by the first processor and the second processor;
in response to determining that the data does not exist in the memory location accessible by the first processor and the second processor, including a unique identifier associated with the data in the message, and moving the data and the unique identifier to the memory location accessible by the first processor and the second processor;
determining whether there is metadata associated with the data;
in response to determining that there is metadata associated with the data, including the metadata in the message; and
transmitting the message to a second processor.

11. The method of claim 10 , wherein the network latency is determined from a timestamp included with the data.

12. The method of claim 11 , wherein the unique identifier includes the timestamp included with the data.

13. The method of claim 12 , wherein the adaptable size threshold value is included in a configuration file.

Assignees

• Ge Aviation Systems Llc

Inventors

• Occhipinti Benjamin Thomas
• Kuczynski Konrad Robert
• Buehler Eric Daniel

Classifications

• G06F15/163 Primary

  Interprocessor communication · CPC title

• H04L51/18 Primary

  Commands or executable codes · CPC title

• H04L51/046

  Interoperability with other network applications or services · CPC title

• H04L51/08

  Annexed information, e.g. attachments · CPC title