Disk power manager for network servers

• US 5,666,538 A
• Filed: 06/07/1995
• Issued: 09/09/1997
• Est. Priority Date: 06/07/1995
• Status: Expired due to Fees

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: AST Research, Inc. (Samsung Electronics Co. Ltd.)
• Inventors: DeNicola, Richard M.
• Primary Examiner(s): Harvey, Jack B.
• Assistant Examiner(s): PANCHOLI, JIGAR K
• Application Number:
  US08/475,567
• Time in Patent Office: 825 Days
• Field of Search: 395/750, 364/707, 365/226-229, 324/103 R
• US Class Current: 713/320
• CPC Class Codes: G06F 1/3221 of disk drive devices
• CPC Class Codes: G06F 1/3268 Power saving in hard disk drive
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3447 Performance evaluation by modeling
• CPC Class Codes: G06F 11/3476 Data logging G06F11/14, G06F11/2205 take precedence
• CPC Class Codes: G06F 11/3485 for I/O devices
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A computer network includes a network server in communication with a plurality of user terminals. The network server includes disk drives which store data accessible via the user terminals. To increase power efficiency of the network server, the number of accesses to the disk drives is monitored and a histogram is generated to display the distribution of disk accesses over time. A network administrator subsequently selects time intervals to spin down one or more of the disk drives during periods of disk inactivity as depicted in the histogram. In a preferred embodiment, the network administrator uses the histogram for failure analysis to determine which disks are more likely to fail over long time periods.

Claims

1. A computer network which provides for increased power efficiency during selected time intervals, said network comprising:

a plurality of user terminals;

a network server in communication with said plurality of user terminals to transmit data to said user terminals and to receive data from said user terminals, said network server comprising;

a processor which executes commands contained in a network operating system and an application module; and

a plurality of data storage disk drives in communication with said processor, one or more of said disk drives being independently spun-down to conserve power for predetermined intervals in response to commands issued by said application module in response to commands issued by said network operating system executed by said processor, said predetermined intervals being chosen to correspond to time intervals when said one or more of said disk drives have a reduced probability of being accessed, thereby minimizing the impact on the operation of said network server.

2. A computer network as defined in claim 1, wherein said application module comprises a device driver.

3. A computer network as defined in claim 1, wherein said network server further includes a user input module which receives instructions from a user to determine time intervals over which said disk drives are to be spun-down.

4. A computer network as defined in claim 3, wherein said network server further comprises a monitoring module, executable by said processor, which monitors accesses to said disk drives to derive a statistical record therefrom, and a display terminal which displays data relating to said statistical record.

5. A network server which provides for increased power efficiency during selected time intervals, said network server comprising:

a processor which executes commands contained in a network operating system and an application module; and

a plurality of data storage disk drives in communication with said processor, one or more of said disk drives being independently spun-down to conserve power for predetermined intervals in response to commands issued by said application module in response to commands issued by said network operating system executed by said processor, said predetermined intervals being chosen to correspond to time intervals when said one or more of said disk drives have a reduced probability of being accessed, thereby minimizing the impact on the operation of said network server.

6. A method of managing disk drive power within a network server having a processor which executes commands contained in a plurality of application modules, said network server having a plurality of disk drives responsive to said commands executed by said processor, said method comprising the steps of:

loading a device driver application module to be executed by said processor;

loading a monitoring application module to be executed by said processor;

loading a console application module to be executed by said processor;

monitoring accesses to each individual disk drive by said monitoring application module to generate a statistical record of said accesses for each individual disk drive;

inputting data relating to spin-up and spin-down of said disk drives via said console application module based upon said statistical record of said access to said disk drives; and

controlling spin-ups and spin-downs of said disk drives to reduce power consumption of said disk drives via commands executed by said processor to said device driver module based upon said input data, said commands based upon the time of day and day of the week.

7. A method as defined in claim 6, wherein said inputting step, comprises inputting data relating to spin-up and spin-down of individual disk drives of said disk drive subsystem.

8. A method as defined in claim 6, wherein said controlling step, comprises controlling spin-up and spin-down of individual disk drives of said disk drive subsystem.

9. A method of managing disk drive power within a network server having a processor which executes commands, said network server having a plurality of disk drives responsive to said commands executed by said processor, said disk drives being accessible from a plurality of user terminals, said method comprising the steps of:

monitoring accesses to said disk drives;

tabulating a number of accesses to said disk drives within a selected time interval to establish a statistical record; and

controlling power supplied to said disk drives in response to commands executed by said processor based upon said statistical record such that said disk drives are independently spun-up and spun-down without substantially affecting an availability of said disk drives having a statistically significant probability of being accessed by said network server at a point in time.

10. A method as defined in claim 9, wherein said monitoring step, monitors accesses to individual disk drives of said disk drive subsystem.

11. A method as defined in claim 9, wherein said controlling step comprises controlling power supplied to individual disk drives of said disk drive subsystem.

12. A method as defined in claim 9, wherein said controlling step, further comprises the steps of:

displaying information relating to said statistical record to a user;

inputting control parameters for determining time intervals over which said disk drives are to be spun-down based upon said displayed information; and

spinning-down said disk drives at said determined time intervals in response to said commands executed by said processor.

13. A method as defined in claim 9, wherein said control parameters are selected to spin-down said disk drives during periods where said disk drives are statistically less likely to be accessed.