Electrical computers and digital processing systems: support – Computer power control – Power sequencing
Reexamination Certificate
2000-04-13
2003-07-15
Myers, Paul R. (Department: 2181)
Electrical computers and digital processing systems: support
Computer power control
Power sequencing
C713S340000, C714S022000
Reexamination Certificate
active
06594771
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to power management for electronic devices with hot-swappable components and more specifically to a system for providing power to components newly added to the device only if the device has sufficient power for all previously powered components as well as the newly added components.
REFERENCE TO MICROFICHE APPENDIX
This application is submitted with a microfiche appendix consisting of 1 microfiche transparency with a total of 19 frames.
BACKGROUND
Computers and related electronic devices have become widely used, and their continuous and error-free operation is vital in many applications. The increased use of complex electronic systems with multiple electronic components has resulted in a need to reduce the space occupied by the electronic components and to simplify their mounting assemblies. Electronic components are therefore often mounted in a rack or cabinet. The electronic components, such as printed circuit boards, may either be mounted directly in the rack, or may be placed in chassis which are mounted in the rack. The use of racks or cabinets for large complex electronic systems having multiple electronic components has the advantage of simplifying assembly and maintenance of the equipment, and reducing the space occupied by the systems.
Multiple linked racks are often used in large multiprocessor computer systems. For example, a typical system may have eight linked racks, each with eight microprocessor clusters, operating together as one computer system and running a single operating system. This type of computer system may be used in banks, Internet stores, flight managements systems, etc, where the constant availability of the computer system is critical. In these types of applications, a rack mounted computer system may be required to operate with 99.999% uptime, meaning that the system must be operative for all but about five minutes each year. If a single electronic component in a rack fails, causing the rack to fail, all the linked racks are likely to fail. Therefore, in such a rack mounted computer system with many linked electronic components, the power capacity and power requirements of the system must be carefully monitored and managed.
Data transfer between rack mounted electronic components is simplified by placing a backbone in the rack. The backplane is a wiring board containing electrical conductors such as a data bus, address bus, custom electrical signals as needed by the electronic components, and power lines. The electronic components are connected to the backplane as they are mounted in the rack. The electronic components can then exchange information through the electrical conductors on the backplane.
The use of a rack for large complex electronic systems can also simplify the electronic components, since power is typically supplied by the rack rather than by individual power supplies on each electronic component. This also simplifies cooling of the electronic components, since the main power supplies, which produce a great deal of heat, may be grouped in one location and isolated from the electronic components. The electronic components may then use simpler power supplies or regulators. However, this requires that the power supply capacity of the rack be carefully managed to ensure that each electronic component has enough power. If the rack is underpowered and all the electronic components attempt to draw power from the rack, none of the electronic components will operate correctly.
A conservative solution is to include power supplies that can provide as much power as the rack might ever need. To calculate the power needed in this solution, the maximum power requirements of the most power hungry electronic components are determined, multiplied by the number of slots for electronic component in the rack. However, some of the electronic component slots may remain unused in many applications, and electronic components vary greatly in their power requirements. This conservative solution thus will almost always have unused power capacity and is heavy and costly. On the other hand, newly developed electronic components for the rack may require even more power than the previous components, so the operator of the electronic system still needs to keep track of the power capacity and requirements for the system to add power capacity as needed.
Another typical solution is to include power supplies that can provide as much power as the rack might need under the most common configuration. This reduces the cost of the power system and minimizes unused power capacity. However, this increases the burden on the operator of the electronic system to keep track of the power capacity and requirements for the system. Each time the electronic components in the rack are changed, the operator will need to calculate the power requirements and change power supplies as needed.
Power management in a rack for large complex electronic systems is further complicated by the need to keep the electronic components powered and operating, even when a electronic component is removed or a new electronic component is added to the system. For example, if five electronic components are powered and operating in the rack, adding a sixth electronic component must not cause the five existing components to fail. These types of electronic components that may added to or removed from the system during operation are referred to as “hot-swappable” electronic components.
Consequently, a need exists for a power management system for electronic devices having multiple electronic components. A further need exists for a system to manage power for hot-swappable electronic components. A still further need exists for a system to manage power from hot-swappable power supply modules. A still further need exists for a power management system for an electronic device to automatically calculate power requirements in the device. A still further need exists for a system to provide power to newly added components in an electronic device only when preexisting components in the device will not be deprived of power. A still further need exists for a power management system for an electronic device to alert an operator of the system if the electronic device has insufficient electrical power to meet the needs of all the electronic components in the device.
SUMMARY
To assist in achieving the aforementioned needs, the inventors have devised a power management system for an electronic device having hot-swappable components and redundant hot-swappable power supplies. Each component reports its power requirement to a power monitor in the device. The power monitor reads the power capacity and status from the power supplies to determine the total power capacity for the device. If the electronic device has sufficient power capacity to supply the installed components, the power monitor allows the components to draw power from the device. If the electronic device does not have sufficient power capacity to supply the installed components, the power monitor alerts the operator of the device of the problem.
As hot-swappable components are added to the device, the power monitor receives the reports with their power requirements and calculates whether the device has sufficient power for them along with the already powered components. If the device has sufficient power, the power monitor signals the newly added components that they may draw power from the device. If the device does not have sufficient power, the power monitor does not signal the components to draw power, leaving them unpowered, and the power monitor alerts the operator of the device of the problem.
The invention may comprise a method of managing power in an electronic device having at least one connectable component. The method includes determining a total power requirement for the at least one connectable component. The available power level for a power supply connected to the electronic device is determined. The total power requirement is compared with the available power level. The at least one connectable component
Bouchier Paul Henry
Koerber Christine L.
Hewlett--Packard Development Company, L.P.
Myers Paul R.
Phan Raymond N
LandOfFree
Method and apparatus for managing power in an electronic device does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for managing power in an electronic device, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for managing power in an electronic device will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3055477