Staggered software synchronization

Patent metadata

Field	Value
Publication number	US-11704167-B2
Application number	US-202217888082-A
Country	US
Kind code	B2
Filing date	Aug 15, 2022
Priority date	Jan 31, 2020
Publication date	Jul 18, 2023
Grant date	Jul 18, 2023

Abstract

Approaches in accordance with various embodiments can reduce scheduling delays due to concurrent processing requests, as may involve VSyncs in multi-streaming systems. The software synchronization signals can be staggered relative to each other by offsetting an initial synchronization signal. These software synchronization signals can be readjusted over time such that each synchronization signal maintains the same relative offset, as may be with respect to other applications or containers.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
assigning a first VSync offset for a first stream of a multi-streaming system, the multi-streaming system corresponding to a plurality of individual streams having an associated synchronization period;
selecting a next stream of the multi-streaming system;
determining a largest sub-interval between already assigned VSync offsets within the synchronization period;
assigning a second VSync offset corresponding to a midpoint of the determined largest sub-interval; and
storing VSync offset assignments for use in performing subsequent operations for the individual streams.

2. The computer-implemented method of claim 1 , further comprising:
scheduling a time to perform a VSync operation using the first VSync offset; and
transmitting a signal to perform the VSync operation when a duration of time equivalent to the time to perform the VSync operation has elapsed.

3. The computer-implemented method of claim 1 , wherein the first VSync offset is assigned by staggering the first VSync offset for the stream based on a number of individual streams in the multi-streaming system.

4. The computer-implemented method of claim 1 , wherein each stream of at least two streams of the multi-streaming system corresponds to a separate container corresponding to a separate instance of an operating environment.

5. The computer-implemented method of claim 1 , wherein assigning the first VSync offset for the first stream comprises applying a randomizer using a seed initialized based on a value unique to the first stream.

6. The computer-implemented method of claim 1 , wherein assigning the first VSync offset for the first stream comprises:
dividing each processing period by a number of individual streams corresponding to the multi-streaming system to determine a quotient;
assigning a different stream identifier to each individual stream; and
assigning a different coefficient to each stream identifier, wherein an initial VSync offset for any individual stream of the multi-streaming system is the product of the quotient and the coefficient of the stream identifier corresponding to the individual stream.

7. The computer-implemented method of claim 6 , wherein the number of individual streams corresponding to the multi-streaming system is equal to a number of streams the multi-streaming system is able to support.

8. The computer-implemented method of claim 1 , further comprising:
calculating the second VSync offset for the next stream; and
maintaining a staggered VSync schedule by dynamically adjusting the second VSync offset of the next stream of the multi-streaming system.

9. A system, comprising:
a network interface to communicate data over a network with a plurality of client devices over a plurality of steams; and
one or more processing units configured to:
assign a first VSync offset for a first stream of a multi-streaming system, the multi-streaming system corresponding to a plurality of individual streams having an associated synchronization period;
select a next stream of the multi-streaming system;
determine a largest sub-interval between already assigned VSync offsets within the synchronization period;
assign a second VSync offset corresponding to a midpoint of the determined largest sub-interval; and
store VSync offset assignments for use in performing subsequent operations for the individual streams.

10. The system of claim 9 , wherein at least one processing unit of the one or more processing units is further configured to:
schedule a time to perform a VSync operation using the first VSync offset; and
transmit a signal to perform the VSync operation when a duration of time equivalent to the time to perform the VSync operation has elapsed.

11. The system of claim 9 , wherein the first VSync offset is assigned by staggering the first VSync offset for the stream based on a number of individual streams in the multi-streaming system.

12. The system of claim 9 , wherein each stream of at least two streams of the multi-streaming system corresponds to a separate container corresponding to a separate instance of an operating environment.

13. The system of claim 9 , wherein assigning the first VSync offset for the first stream comprises applying a randomizer using a seed initialized based on a value unique to the first stream.

14. The system of claim 9 , wherein assigning the first VSync offset for the first stream comprises:
dividing each processing period by a number of individual streams corresponding to the multi-streaming system to determine a quotient;
assigning a different stream identifier to each individual stream; and
assigning a different coefficient to each stream identifier, wherein an initial VSync offset for any individual stream of the multi-streaming system is the product of the quotient and the coefficient of the stream identifier corresponding to the individual stream.

15. The system of claim 14 , wherein the number of individual streams corresponding to the multi-streaming system is equal to a number of streams the multi-streaming system is able to support.

16. The system of claim 9 , wherein at least one processing unit of the one or more processing units is further configured to:
calculate the second VSync offset for the next stream; and
maintain a staggered VSync schedule by dynamically adjusting the second VSync offset of the next stream of the multi-streaming system.

17. A processor, comprising:
one or more processing units to:
assign a first VSync offset for a first stream of a multi-streaming system, the multi-streaming system corresponding to a plurality of individual streams having an associated synchronization period;
select a next stream of the multi-streaming system;
determine a largest sub-interval between already assigned VSync offsets within the synchronization period;
assign a second VSync offset corresponding to a midpoint of the determined largest sub-interval; and
store VSync offset assignments for use in performing subsequent operations for the individual streams.

18. The processor of claim 17 , wherein at least one processing unit of the one or more processing units is further configured to:
schedule a time to perform a VSync operation using the first VSync offset; and
transmit a signal to perform the VSync operation when a duration of time equivalent to the time to perform the VSync operation has elapsed.

19. The processor of claim 17 , wherein the first VSync offset is assigned by staggering the first VSync offset for the stream based on a number of individual streams in the multi-streaming system.

20. The processor of claim 17 , wherein assigning the first VSync offset for the first stream comprises:
dividing each processing period by a number of individual streams corresponding to the multi-streaming system to determine a quotient;
assigning a different stream identifier to each individual stream; and
assigning a different coefficient to each stream identifier, wherein an initial VSync offset for any individual stream of the multi-streaming system is the product of the quotient and the coefficient of the stream identifier corresponding to the individual stream.

Assignees

• Nvidia Corp

Inventors

• Poddar Bimal
• Ryu Donghan
• Gold Michael
• Koser Samuel Reed
• Zhao Zhang Xiao Bo

Classifications

• G06F9/52 Primary

  Program synchronisation; Mutual exclusion, e.g. by means of semaphores · CPC title

• G06F9/505 Primary

  considering the load · CPC title

• G06F9/544

  Buffers; Shared memory; Pipes · CPC title

• G06F9/4843

  by program, e.g. task dispatcher, supervisor, operating system · CPC title

• G06N3/08

  Learning methods · CPC title