摘要:
A cooling system for a high-end server includes four hot-pluggable fans plugged into a fan control board. The fans are arranged in two groups, with each group having two fans, one behind the other. One of the groups of fans is used to cool the processor boards and the other group is used to cool the system I/O board slots. Under normal operation, only one fan from each group is active, while the other fan freewheels, providing redundancy. A fan control board delivers power to each of the fans and further provides a signal, responsive to temperature sensors, to each of the fans to control their speeds. Each of the fans provides a fan fault signal and a fan not present signal to the fan control board. The temperature sensors are placed proximate the processors and I/O components to monitor the operating temperatures thereof, and communicate the respective temperatures back to the fan control board, via I.sup.2 C bus. An operating system is utilized to drive the fan controller but can be overridden by the fan controller during critical conditions. An air ramp is positioned between the processors to help direct the flow of air generated by the fans onto some of the processors. Additionally, each of the fans are configured with a quick installment device to facilitate single-hand insertion and removal of the fans from the computer system.
摘要:
A computer system or computer server system having redundant devices that are periodically checked in order to determine whether the redundant devices would be operational if the primary devices failed. The primary and redundant devices are in electrical communication with a controller circuit which, in turn, is responsive to a microprocessor and/or the computer's operating system. The primary and redundant devices can be hot-pluggable so that the computer system does not need to be powered down when and if the primary or redundant devices need to be replaced.
摘要:
Through a menu driven selection procedure, a user is given options for how a server should respond to an actuation of a power switch. In one option, the power switch is disabled to prevent accidental shut down of the power supply in response to power switch actuation. In another option, the shut down of the power supply follows soon after the expiration of a count down timer which is triggered by power switch actuation. A subsequent actuation of the power switch in this option aborts the count down. In yet another option, the shut down of the power supply follows soon after the completion of a graceful shut down of the server operating system which is triggered by power switch actuation. A subsequent actuation of the power switch in this option causes an immediate shut down of the power supply.
摘要:
Apparatus, and an associated method, for determining the level of power supply redundancy in a modular computer system. Determination of the level of power supply redundancy is made dynamically, during on-line operation of the computer system. Reconfiguration of the computer system, such as to increase the load which must be powered by modular power supply components, or removal or addition of power supply components to form portions of the computer system cause initiation of a new determination of the level of power supply redundancy. Indications of inadequate levels of power supply redundancy are provided to a user of the computer system so that corrective action can be taken.
摘要:
A flexible arrangement allows a single arrangement of Ethernet channel adapter (ECA) hardware functions to appear as needed to conform to various operating system deployment models. A PCI interface presents a logical model of virtual devices appropriate to the relevant operating system. Mapping parameters and values are associated with the packet streams to allow the packet streams to be properly processed according to the presented logical model and needed operations. Mapping occurs at both the host side and at the network side to allow the multiple operations of the ECA to be performed while still allowing proper delivery at each interface.
摘要:
A flexible arrangement allows a single arrangement of Ethernet channel adapter (ECA) hardware functions to appear as needed to conform to various operating system deployment models. A PCI interface presents a logical model of virtual devices appropriate to the relevant operating system. Mapping parameters and values are associated with the packet streams to allow the packet streams to be properly processed according to the presented logical model and needed operations. Mapping occurs at both the host side and at the network side to allow the multiple operations of the ECA to be performed while still allowing proper delivery at each interface.
摘要:
A flexible arrangement allows a single arrangement of Ethernet channel adapter (ECA) hardware functions to appear as needed to conform to various operating system deployment models. A PCI interface presents a logical model of virtual devices appropriate to the relevant operating system. Mapping parameters and values are associated with the packet streams to allow the packet streams to be properly processed according to the presented logical model and needed operations. Mapping occurs at both the host side and at the network side to allow the multiple operations of the ECA to be performed while still allowing proper delivery at each interface.
摘要:
A flexible arrangement allows a single arrangement of Ethernet channel adapter (ECA) hardware functions to appear as needed to conform to various operating system deployment models. A PCI interface presents a logical model of virtual devices appropriate to the relevant operating system. Mapping parameters and values are associated with the packet streams to allow the packet streams to be properly processed according to the presented logical model and needed operations. Mapping occurs at both the host side and at the network side to allow the multiple operations of the ECA to be performed while still allowing proper delivery at each interface.
摘要:
Improved techniques for facilitating identification of a computer system's hardware features to software executing on the computer system are disclosed. In one embodiment, the software is system management software that serves to monitor and control the computer system to reduce failures and/or improve performance. In a general sense, the invention provides a system feature table within a computer system to describe each of the hardware features of the computer system, including feature data and appropriate access mechanisms for additional information. In addition, a user can be notified when the software does not support all the hardware features of the computer system. The invention also is able to identify locations of failing hardware features for easy replacement or examination.
摘要:
A computer system in accordance with the present invention provides a method of communicating asynchronous events from a mini-port driver of a computer system to a software component without polling the mini-port driver. The computer system provides a monolithic driver as a dispatch between a mini-port driver and a software component. Until an asynchronous event is detected, an event thread in the monolithic driver to a software component is blocked. When an asynchronous event is detected by the mini-port driver, the mini-port driver calls the monolithic driver with a monolithic driver entry point which queues the event and unblocks the event thread. The monolithic driver provides the asynchronous event to the software component through the unblocked event thread. The event thread is then re-blocked and remains blocked until another asynchronous event is detected. Asynchronous events are thereby communicated from a mini-port driver to a software component without polling the mini-port driver.