IP Library Granted Patent US 11,645,218
Granted Patent B2
US 11,645,218 · App. 17/175,636 · Granted May 9, 2023

Network architecture providing high speed storage access through a PCI express fabric between a compute node and a storage server within an array of compute nodes

Inventor: Roelof Roderick Colenbrander (Costa Mesa, CA)
Assignee: Sony Interactive Entertainment Inc.
G06F13/382G06F13/1668G06F13/4022G06F13/4221G06F13/4282H04L49/351H04L67/1097G06F2213/0026G06F2213/3808
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,645,218
App. No.
17/175,636
Granted
May 9, 2023
Kind
B2
Abstract

A network architecture including a streaming array that includes a plurality of compute sleds, wherein each compute sled includes one or more compute nodes. The network architecture including a network storage of the streaming array. The network architecture including a PCIe fabric of the streaming array configured to provide direct access to the network storage from a plurality of compute nodes of the streaming array. The PCIe fabric including one or more array-level PCIe switches, wherein each array-level PCIe switch is communicatively coupled to corresponding compute nodes of corresponding compute sleds and communicatively coupled to the network storage. The network storage is shared by the plurality of compute nodes of the streaming array.

Claims (70)

1. A network architecture, comprising:

a streaming array including a plurality of compute sleds, wherein each compute sled of the plurality of compute sleds includes one or more compute nodes;

a network storage of the streaming array; and

a first Peripheral Component Interconnect Express (PCIe) fabric of the streaming array configured to provide direct access to the network storage from a plurality of compute nodes of the streaming array, the first PCIe fabric including one or more array-level PCIe switches directly coupling each of the plurality of compute nodes in the streaming array to the network storage;

an array management server (AMS) configured for management of the plurality of compute sleds; and

a second PCIe fabric of the streaming array configured to provide control communication between the AMS and the plurality of compute sleds of the streaming array,

wherein the network storage is shared by the plurality of compute nodes of the streaming array,

wherein the plurality of compute sleds provides a first plurality of lanes to the one or more array level PCIe switches,

wherein the one or more array-level PCIe switches provide a second plurality of lanes to the network storage,

wherein the second plurality of lanes is less than the first plurality of lanes.

2. The network architecture of claim 1 ,

wherein each of the one or more array-level PCIe switches provides communications from corresponding compute nodes of corresponding compute sleds for purposes of streaming data and management information for the plurality of compute sleds and corresponding compute nodes.

3. The network architecture of claim 1 , further comprising:

an Ethernet fabric;

a network switch configured for providing communications from the plurality of compute nodes to a cluster switch over the Ethernet fabric,

wherein the Ethernet fabric communicatively couples the network switch to the network storage and to the plurality of compute nodes.

4. The network architecture of claim 3 ,

wherein the cluster switch is communicatively coupled to another network switch of another streaming array, and

wherein the cluster switch is communicatively coupled to a communications network to provide network communications that are external to the network architecture,

wherein the cluster switch is communicatively coupled to a distributed storage that is configured for storing a plurality of gaming applications.

5. The network architecture of claim 3 , further comprising:

a rack assembly including one or more streaming arrays, each of the one or more streaming arrays being independently operable and including a corresponding network storage, a corresponding PCIe fabric configured to provide direct access to the corresponding network storage from compute nodes of the each of the one or more streaming arrays, a corresponding Ethernet fabric, and corresponding network switch.

6. The network architecture of claim 1 , wherein the network storage includes:

at least one random access memory (RAM) drive; and

at least one non-volatile memory express (NVMe) drive.

7. The network architecture of claim 1 ,

wherein the each of the plurality of compute nodes is configured for executing one or more instances of a plurality of gaming applications.

8. The network architecture of claim 1 , wherein the each compute sled of the plurality of compute sleds includes:

a sled-level PCIe switch communicatively coupled to corresponding compute nodes and a corresponding array-level PCIe switch,

wherein the sled-level PCIe switch is configured to provide communications between the corresponding compute nodes and the network storage through the first PCIe fabric via the corresponding array-level PCIe switch.

9. The network architecture of claim 1 , wherein the each of the plurality of compute sleds includes:

a board management controller (BMC) configured for controlling one or more components on a corresponding sled.

10. The network architecture of claim 1 ,

wherein the network storage stores read-only game content of a gaming application, such that the read-only game content may be shared between compute instances executing the gaming application on one or more of the plurality of compute nodes of the streaming array.

11. A network architecture, comprising:

a cluster switch;

a plurality of streaming arrays, wherein a corresponding streaming array in the plurality of streaming arrays includes:

a plurality of compute sleds, wherein each compute sled of the plurality of compute sleds includes one or more compute nodes;

a network storage; and

a first Peripheral Component Interconnect Express (PCIe) fabric configured to provide direct access to the network storage from a plurality of compute nodes of the corresponding streaming array, the first PCIe fabric including at least one array-level PCIe switch directly coupling each of the plurality of compute nodes in the corresponding streaming array to the network storage, wherein the network storage is shared by the plurality of compute nodes of the corresponding streaming array;

an array management server (AMS) configured for management of the plurality of compute sleds; and

a second PCIe fabric of the streaming array configured to provide control communication between the AMS and the plurality of compute sleds of the streaming array,

wherein the plurality of compute sleds provides a first plurality of lanes to the at least one array level PCIe switch,

wherein the at least one array-level PCIe switch provides a second plurality of lanes to the network storage,

wherein the second plurality of lanes is less than the first plurality of lanes; and

a plurality of network switches coupled to compute nodes of the plurality of streaming arrays, wherein each of the plurality of network switches is configured to provide communications from the plurality of compute nodes of the corresponding streaming array to the cluster switch.

12. The network architecture of claim 11 ,

wherein each of the at least one array-level PCIe switches of the corresponding streaming array provides communications from corresponding compute nodes of corresponding compute sleds for purposes of streaming data and management information for the corresponding compute sleds and the corresponding compute nodes.

13. The network architecture of claim 11 , wherein the corresponding streaming array includes:

an Ethernet fabric; and

a network switch configured for providing communications from the plurality of compute nodes to a cluster switch over the Ethernet fabric,

wherein the Ethernet fabric communicatively couples the network switch to the network storage and to the plurality of compute nodes,

wherein the cluster switch is communicatively coupled to another network switch of another streaming array.

14. The network architecture of claim 13 , further comprising:

a rack assembly including one or more streaming arrays, each of the one or more streaming arrays being independently operable and including a corresponding network storage, a corresponding PCIe fabric configured to provide direct access to the corresponding network storage from compute nodes of the each of the one or more streaming arrays, a corresponding Ethernet fabric, and corresponding network switch.

15. The network architecture of claim 13 ,

wherein the cluster switch is communicatively coupled to a communications network to provide network communications that are external to the network architecture,

wherein the cluster switch is communicatively coupled to a distributed storage that is configured for storing a plurality of gaming applications.

16. The network architecture of claim 11 , wherein the network storage includes:

at least one random access memory (RAM) drive; and

at least one non-volatile memory express (NVMe) drive.

17. The network architecture of claim 11 ,

wherein the each of the plurality of compute nodes is configured for executing one or more instances of a plurality of gaming applications.

18. The network architecture of claim 11 , wherein the each compute sled of the plurality of compute sleds of the corresponding streaming array includes:

a sled-level PCIe switch communicatively coupled to corresponding compute nodes and a corresponding array-level PCIe switch,

wherein the sled-level PCIe switch is configured to provide communications between the corresponding compute nodes and the network storage through the first PCIe fabric via the corresponding array-level PCIe switch.

19. The network architecture of claim 11 , wherein the each compute sled of the plurality of compute sleds of the corresponding streaming array includes:

a board management controller (BMC) configured for controlling one or more components on a corresponding sled.

20. The network architecture of claim 11 ,

wherein the network storage of the corresponding streaming array stores read-only game content of a gaming application, such that the read-only game content may be shared between compute instances executing the gaming application on one or more of the plurality of compute nodes of the plurality of the corresponding streaming array.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 1, 2021
From: COLENBRANDER, ROELOF RODERICK
To: SONY INTERACTIVE ENTERTAINMENT INC.
Reel/Frame 055801/0732 →
Continuity (2)
Provisional Application 62977138 · Feb 14, 2020
Related Publication 20210255979A1 · Aug 19, 2021