# US 2023/0273889 A1 · Compute sled providing high speed storage access through a pci express fabric between compute nodes and a storage server

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0273889 A1
- Document ID: US20230273889A1
- Status: Published application
- Kind: A1
- Application number: 18/314,693
- Filed: 2023-05-09
- Published: 2023-08-31
- Earliest claimed: 2020-02-14
- Assignees: Sony Interactive Entertainment Inc.
- Applicants: Sony Interactive Entertainment Inc.
- Inventors: Roelof Roderick Colenbrander
- Web page: https://patentgrep.com/patent/US20230273889A1

## CPC

- G06F 13/382: {using universal interface adapter}
- H04L 49/15: Interconnection of switching modules
- G06F 1/181: {Enclosures (for portable computers G06F1/1613)}
- G06F 1/187: Mounting of fixed or removable disk drives
- G06F 3/061: {Improving I/O performance}
- G06F 3/0659: {Command handling arrangements, e.g. command buffers, queues, command scheduling}
- G06F 3/067: {Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]}
- G06F 13/1668: {Details of memory controller}
- G06F 13/4022: {using switching circuits, e.g. switching matrix, connection or expansion network (G06F13/4009 takes precedence)}
- G06F 13/4221: {being an input/output bus, e.g. ISA bus, EISA bus, PCI bus, SCSI bus}
- G06F 13/4282: {on a serial bus, e.g. I2C bus, SPI bus (on daisy chain buses G06F13/4247)}
- H04L 49/351: for local area network [LAN], e.g. Ethernet switches
- H04L 67/1097: for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
- A63F 13/355: Performing operations on behalf of clients with restricted processing capabilities, e.g. servers transform changing game scene into an encoded video stream for transmitting to a mobile phone or a thin client
- G06F 1/183: Internal mounting support structures, e.g. for supporting printed circuit boards
- G06F 2213/0026: PCI express
- G06F 2213/3808: Network interface controller
- H05K 7/1485: {Servers; Data center rooms, e.g. 19-inch computer racks}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- A63F: CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- H05K: PRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS

## 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

### Claim 1

A compute sled comprising:
a plurality of compute nodes, wherein each of the plurality of compute nodes includes one or more Peripheral Component Interconnect Express (PCIe) lanes and one or more universal asynchronous receiver transmitter (UART) connections configured for transmitting and receiving serial data configured;
a sled level PCIe switch electrically coupled to the plurality of compute nodes, the sled level PCIe switch including a plurality of incoming lanes and a plurality of outgoing lanes,
wherein the first plurality of incoming lanes is greater than the second plurality of outgoing lanes,
wherein the one or more PCIe lanes of the each of the plurality of compute nodes are electrically coupled to the plurality of incoming lanes,
wherein the plurality of outgoing lanes is electrically coupled to network storage via an array-level PCIe switch; and
a board management controller (BMC) electrically coupled to the one or more UART connections of the each of the plurality of compute nodes, the BMC configured for management of one or more components of each of the plurality of compute nodes using a plurality of UART signals delivered over the one or more UART connections to the each of the plurality of compute nodes.

### Claim 2 (depends on 1)

The compute sled of claim 1, further comprising:
an Ethernet network interface configured to provide network interfacing to an Ethernet fabric,
wherein the Ethernet fabric communicatively couples the plurality of compute nodes to the network storage via a network switch.

### Claim 3 (depends on 2)

The compute sled of claim 2,
wherein the Ethernet network interface is electrically coupled to the one or more of the plurality of outgoing lanes from the sled level PCIe switch to allow each of the plurality of compute nodes to communicate with the sled level PCIe switch.

### Claim 4 (depends on 1)

The compute sled of claim 1,
wherein the BMC is electrically coupled to the sled level PCIe switch.

### Claim 5 (depends on 4)

The compute sled of claim 4,
wherein one of the plurality of outgoing lanes of the sled level PCIe switch is dedicated to electrically coupling the BMC to the sled-level PCIe switch.

### Claim 6 (depends on 1)

The compute sled of claim 1, further 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 PCIe fabric of the streaming array configured to provide direct access to the network storage from the one or more compute nodes of the each compute sled of the plurality of compute sleds of the streaming array, the PCIe fabric including one or more array-level PCIe switches directly coupling each of the one or more compute nodes of the each compute sled of the plurality of compute sleds of the streaming array to the network storage via a corresponding sled level PCIe switch.

### Claim 7 (depends on 1)

The compute sled of claim 1, further comprising:
a power interposer electrically coupled to plurality of compute nodes and configured for providing a plurality of power signals,
wherein the power interposer is electrically coupled to a power bus bar providing power.

### Claim 8 (depends on 7)

The compute sled of claim 7, further comprising:
a general purpose input/output connection between the BMC to the power interposer,
wherein the BMC is configured to a power control signal to the power interposer for controlling power delivered to a compute node.

### Claim 9 (depends on 1)

The compute sled of claim 1, wherein the sled level PCIe switch includes:
a plurality of PCIe connectors electrically coupled to the plurality of compute nodes via corresponding connection interfaces;
wherein at least one of the plurality of outgoing lanes of the sled-level PCIe switch is dedicated for electrically coupling the plurality of compute nodes of the compute sled to an array management server.

### Claim 10 (depends on 9)

The compute sled of claim 9, wherein the sled-level PCIe switch includes:
a plurality of UART connectors electrically coupled to a plurality of UAT interfaces of the plurality of compute nodes;
a UART/GPIO controller electrically coupled to the plurality of UART connectors for communicating management control signals from the array management server to the plurality of compute nodes,
wherein the UART/GPIOP controller is electrically coupled to a general purpose input/output connector for communicating power signals to a power interposer.

### Claim 11 (depends on 1)

The compute sled of claim 1, wherein each compute node is configured for executing one or more instances of a plurality of gaming applications

### Claim 12 (depends on 1)

The compute sled 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 compute nodes of the plurality of compute sleds.
