1. A method for adjusting power consumption during server operation, the method comprising:
initializing the server to an initial system power performance state (SPP-state) that comprises a plurality of constituent power settings contained in a table;
modifying at least one of the plurality of constituent power settings stored in the table; and
comparing a maximum power consumption threshold to a power consumption measurement at the initial SPP-state and reducing at least one of the plurality of constituent power settings associated with the initial SPP-state if the maximum power consumption threshold is less than the power consumption measurement at the initial SPP-state.
2. The method of claim 1, wherein the plurality of constituent power settings includes a p-state and clock modulation.
3. The method of claim 1, wherein the initial SPP-state is based on a predetermined maximum power rating of the server.
4. The method of claim 1, further comprising adjusting the maximum power consumption threshold of the server when modifying at least one of the plurality of constituent power settings.
5. The method of claim 1, wherein the amount that the at least one of the plurality of constituent power settings is reduced by is less than a predetermined threshold.
6. The method of claim 1, further comprising reallocating power reductions to at least one other server.
7. The method of claim 1, wherein the plurality of constituent power settings includes a p-state and power reduction of one or more server subsystems other than the CPU.
8. The method of claim 1, wherein the method is executed by a management processor.
9. The method of claim 1, wherein at least some of the method is executed by a zone manager.
10. A server comprising:
a central processing unit (CPU);
a management processor coupled to the CPU via a bus, wherein the management processor adjusts power consumption according to executable code, the code causes the management processor to:
initialize the server to an initial system power performance state (SPP-state) that comprises a plurality of constituent power settings contained in a table;
during a boot process, negotiate power consumption for the server with a management device external to the server; and
modify at least one of the plurality of constituent power settings stored in the table based on the negotiated power consumption for the server.
11. The server of claim 10, wherein the bus is a PCI bus.
12. The server of claim 10, wherein the initial SPP-state is based on a stored maximum power consumption measurement of the server and wherein the code further causes the management processor to adjust the stored maximum power consumption measurement of the server when at least one of the plurality of constituent power settings is increased.
13. The server of claim 12, wherein the code further further causes the management processor to compare the stored maximum power consumption measurement to the power consumption of the server at the current SPP-state and to change at least one power setting in the table for the current SPP-state if the stored maximum power consumption measurement is greater than the power consumption at the current SPP-state.
14. The server of claim 13, wherein changes to the power settings are by less than a predetermined threshold.
15. The server of claim 14, wherein the code further cause the management processor to reallocate power initially allocated to the server to at least one other server that is coupled to the management processor.
16. The server of claim 15, wherein at least some of the code is executed by a zone manager.
17. The server of claim 10, wherein the management processor further couples to a second management processor that controls the power consumption for at least one other server.
18. A server comprising:
a central processing unit (CPU);
a means for managing power within a server coupled to the CPU, the means for managing power performs a method comprising:
initializing the server to an initial system power performance state (SPP-state) that comprises CPU frequency and clock modulation settings contained in a table; and
comparing a maximum power consumption threshold to a real-time power consumption measurement for the initial SPP-state and reducing at least one of the CPU frequency and the clock modulation settings associated with the initial SPP-state if the maximum power consumption threshold is less than the real-time power consumption measurement for the initial SPP-state.
19. The server of claim 18, wherein the means for managing power further couples to an additional means for managing power that controls the power consumption for at least one other server.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
1. A video information recording method that records video data comprising a plurality of video units each of which includes an intra coded I-picture, a predictive coded P-picture including a group of blocks predicted from one picture and a bi-directionally-predictive coded B-picture including a group of blocks predicted from two pictures, said method comprising steps of:
recording at least an access point P-picture including a group of blocks each of which is coded by motion compensation prediction using an I-picture located at the beginning of said video unit including said access point P-picture or a selected one of preceding P-pictures;
recording a P-picture and a B-picture coded by motion compensation prediction using said access point P-picture preceding said P-picture and said B-picture, wherein any P-pictures, other than access point P-pictures, and any B-pictures following said access point P-picture in the same video unit are coded without referring to any picture preceding said access point P-picture;
recording an attribute information preceding said I-picture including picture type information specifying a type of each picture; and
recording said video units and said attribute information in a video data recording area.
2. A playback apparatus for reproducing video data recorded on a medium having a video data recording area, said video data comprising a plurality of video units each of which includes an intra coded I-picture, a predictive coded P-picture including a group of blocks predicted from one picture and a bi-directionally-predictive coded B-picture including a group of blocks predicted from two pictures,
at least one of said video units including an access point P-picture coded by motion compensation prediction using an I-picture located at the beginning of said video unit or a selected one of preceding P-pictures, wherein any P-pictures, other than access point P-pictures, and any B-pictures following said access point P-picture in the same video unit are coded without referring to any picture preceding said access point P-picture,
said medium containing an attribute information in said video data recording area preceding said I-picture including picture type information specifying a type of each picture;
said apparatus comprising,
a video unit identifying unit for identifying a video unit which includes a start picture to display, in the case of starting playback from said start picture;
an attribute information reading unit for reading said attribute information to identify an I-picture or a selected preceding P-picture;
a picture accessing unit for accessing said I-picture or said P-picture used for coding said access point P-picture according to said picture type information of said attribute information;
a first decoder for decoding said I-picture or said P-picture used for coding said access point P-picture;
a second decoder for decoding said access point P-picture using decoded I-picture or P-picture; and
a reproducing unit for reproducing said start picture using decoded access point P-picture, wherein said start picture is decoded without referring to any picture preceding said access point P-picture.