Quanta Computer Incorporated

Fa-Da Lin

Jen-Hsuen Huang

Pin-Hao Hung

Chih-Wei Yu

Yi-Ping Lin

Chih-Wei Yang

Pao-Ching Wang

Wei-Tsung Tang

Sheng-Hao Juang

Various examples of the present technology provide systems and methods for incorporating a switch card and adapter cards in a server system to provide flexible HDD and SSD supports. More specifically, a server system comprises a switch card having at least two different types of interfaces (e.g., a Serial Attached SCSI (SAS) interface, a serial ATA (SATA) interface, or a Peripheral Component Interconnect Express (PCIe) interface), and a controller that comprises a first Central Processing Unit (CPU) and a second CPU. The first CPU is connected to a first adapter card while the second CPU is connected to a second adapter card. The first adapter and the second adapter are coupled to the switch card of the server system.

Systems and methods for flexible HDD/SSD storage support

The present disclosure provides a system and method for providing a basic power to a system in an event that a standby power of power supply units (PSUs) of the system fails. The system comprises a plurality of active components, one or more PSUs, and a power switch. The power switch is connected to a standby power output and a main power output of the PSUs. The power switch can receive status information of the PSUs and determine whether the main power of the PSUs is within a predetermined range. In an event that the standby power of the PSUs fails and the main power of the PSUs is within the predetermined range, the power switch can switch an input power from the standby power to the main power of the PSUs and output a basic power for system use.

Operable server system when standby power of PSU fails

A system and method to dynamically balance power to a multi-node system is disclosed. A chassis management controller is operable to regulate the power from a power source to each of the nodes. The chassis management controller determines a setting power for each nodes and a real power consumed by each node. The chassis management controller determines a next setting power for each node based on the real power and unused total power and total additional balance power for the plurality of nodes. The chassis management controller commands each node to regulate the power consumption of the node up to the setting power value.

System and method for dynamic power capping and load balance management for a multi-node system

The present application discloses a server system. The system includes a chassis, a motherboard installed within the chassis, and a daughterboard installed within the chassis. The motherboard includes one or more central processing unit sockets and one or more memory chip sockets. The daughterboard is communicatively connected to the motherboard. The daughterboard includes one or more input/output (I/O) connectors configured for direct external connections to outside of the chassis. The motherboard has no (I/O) connector configured for direct external connection to outside of the chassis.

Motherboard with daughter input/output board

A computing device configured to detect proper cable assembly to improve assembly and problem diagnosis is provided. The computing device includes a motherboard, a function board, and a middle plane connecting the motherboard and function board. The motherboard includes a baseboard management controller (BMC). The BMC is connected to IC buses. The function board includes integrated circuits. The middle plane includes cable connections interconnecting the IC buses that are connected to the BMC and the integrated circuits. The integrated circuits have unique system addresses that are identifiable by the BMC.

Cable detection

A computing device can flexibly connect bidirectional processor interconnect ports (BPIPs). An exemplary computing device includes a motherboard structure, a first processor, a second processor, and a plurality of connectors disposed on the motherboard structure. The first processor and the second processor can each have at least three BPIPs. A first and second of BPIPs of the first processor can be connected to a first and second BPIPs of the second processor. A third BPIP of the first processor and a third BPIP of the second processor can be connected to a first one and a second one of the plurality of connectors. The plurality of connectors can be connected to a computing card. In some examples, the computing device includes a switching element to selectively couple the connectors to any other element in the computing device.

Flexible interconnect port connection

A server system includes an enclosure, a first section, a motherboard section, a fan section, and a midplane board. The first section is disposed at a first end of the enclosure and includes at least one heat generating component. The motherboard section is disposed at a second end of the enclosure and includes at least one motherboard. The fan section is disposed between the front section and the motherboard section and includes at least one fan for directing air in from the first end and out of the second end. The midplane board is vertically interposed between the fan section and the motherboard section, and is configured to electrically and communicatively couple the at least one motherboard to one or more components in the first section, and includes one or more airflow cutouts. The one or more airflow cutouts of the midplane board make up at least thirty percent of a cross-sectional surface area of the midplane board.

Server system

A riser card is disclosed. The riser card includes a circuit board that has two sides, two longitudinal edges extending along a length of the board, and two orthogonal edges extending along a width of the board. The card also includes a first bus connector along one of the longitudinal edges; a second bus connector on one side of the board extending substantially parallel to the longitudinal edges; and a third bus connector on the other side of the board extending substantially parallel to the orthogonal edges. The second bus connector is configured to connect an add-on unit to the card so that the add-on unit extends substantially perpendicular from the board. The third bus connector is configured to connect an add-on unit to the card so that the add-on unit extends substantially parallel to the board and the first edges.

Riser card

Various examples of the present technology provide systems and methods for monitoring an AC power to a computing system, generating a power failure signal when the AC power is determined abnormal, and cause one or more components of the computing system to be switched to a low power mode or shut down to reduce power consumption of the computing system. In particular, the various embodiments of the present technology provide a method to extend hold time of a power supply unit (PSU), thus extend power-on time of a server system during an input power interruption.

Method for extending hold time of power supply units

Systems, methods, and non-transitory computer-readable storage media for smart power clamping of a redundant power supply. A system configured according to this disclosure can measure, at a baseboard management controller, a system power consumption which indicates total power being delivered by a first power supply unit and a second power supply unit. The system can determine that the system power consumption exceeds a system power consumption capacity and, in response to the determination, communicate a power clamping signal to a processor, resulting in a reduced system power consumption. The system can further identify that the reduced system power consumption exceeds the system power consumption capacity and initiate a hardware throttling of at least one of the first power supply unit and the second power supply unit.

System and method for smart power clamping of a redundant power supply

A separated back plane design within a server rack is provided which allows for the hot-swapping of a control module board. The separated back plane design consists of a base plane board and a control module board which connects to the base plane board. The base plane board can draw power from a power supply and can connect a number of electrical components which are not significantly susceptible to damage or failure. The control module board can consist of a number of other electrical components, which are susceptible to damage or failure. The control module board connects to the base plane board, and can be removed during failure of the control module board. When the control module board is removed from the base plane board, the base plane board and its functions continue to operate, and power need not be shut off in order to replace the control module board.

Separated server back plane

An apparatus may include a chassis that can receive a sled and a locking mechanism. The locking mechanism can mechanically lock the sled to the chassis to prevent a sudden power loss that can be caused from an unexpected removal of the sled from the chassis. To avoid a sudden power loss, a voltage-sensing electrical switch lock can be implemented to the chassis to mechanically lock the sled to the chassis until the sled is ready to be removed. The sled may include one or more computing devices that need to be inactive before removing the sled. The apparatus includes a controller that may detect whether at least one of computing devices in the sled are in an active state or in an inactive state. Based on the determination of the state of the computing devices in the sled, the controller may activate the locking mechanism or de-active locking mechanism.

Chassis with lock mechanism

A server rack includes a rack management controller (RMC) configured to manage a first function and a backplane including a backplane controller (BPC). The BPC is configured to monitor the RMC, determine that the RMC is unavailable, and manage the first function, in response to determining that the RMC is unavailable.

Backup backplane management control in a server rack system

A system includes a top-of-rack, a first plurality of chassis, and a first backplane. The first plurality of chassis each includes at least one rack module and a chassis network switch that connects to each of the at least one rack module. The first backplane includes a first backplane network switch that connects to the top-of-rack and connects to each of the first plurality of chassis via the chassis network switch of each of the first plurality of chassis.

LAN port consolidation in rack architecture

In some implementations, a rack management controller can dynamically manage power supply units in a server rack system. For example, the rack management controller can determine how many power supply units should be powered on based on the power requirements of the server rack system. The management controller can configure a power management controller to turn on the determined number of power supply units. The rack management controller can receive status signals from the power supply units in the server rack. The rack management controller can detect when a power supply unit fails based on the status signals. The rack management controller can dynamically reconfigure the power management controller to replace the failed power supply unit by turning on a reserved power supply unit.

Growing community of inventors

Taoyuan, Taiwan

Fa-Da Lin