Program, method, and apparatus for dynamically allocating servers to target system

• US 8,661,130 B2
• Filed: 03/10/2009
• Issued: 02/25/2014
• Est. Priority Date: 09/29/2006
• Status: Expired due to Fees

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Sonoda, Masataka, Tsuchiya, Satoshi, Koike, Kunimasa, Sekiguchi, Atsuji
• Primary Examiner(s): Daftuar, Saket K
• Application Number:
  US12/401,335
• Publication Number:
  US 20090172168A1
• Time in Patent Office: 1,813 Days
• Field of Search: 709/217, 709219-221, 709223-229, 709/200, 709/201, 709/202, 709/203, 709/204, 709/205, 709/206, 709/207, 709/218, 709/222, 714/3, 714/4, 714/4.1, 714/4.11, 714/4.2, 714/4.3, 714/4.4, 714/6.3, 714/13, 714/23, 714/24, 714/25, 714/37, 714/38.1, 714/38.12, 714/38.14, 714/39, 714/43, 714/44, 714/50, 714/51, 714/56, 714/38.13
• US Class Current: 709/226
• CPC Class Codes: G06F 11/004 Error avoidance G06F11/07 and subgroups take precedence
• CPC Class Codes: G06F 11/008 Reliability or availability analysis
• CPC Class Codes: G06F 11/2025 using centralised failover control functionality
• CPC Class Codes: G06F 11/2028 eliminating a faulty processor or activating a spare
• CPC Class Codes: G06F 11/2041 with more than one idle spare processing component
• CPC Class Codes: G06F 11/3055 Monitoring arrangements for monitoring the status of the computing system or of the computing system component, e.g. monitoring if the computing system is on, off, available, not available error or fault processing without redundancy G06F11/0703; error detection or correction by redundancy in data representation G06F11/08; error detection or correction by redundancy in operation G06F11/14; error detection or correction by redundancy in hardware G06F11/16
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3442 for planning or managing the needed capacity
• CPC Class Codes: G06F 2201/875 Monitoring of systems including the internet
• CPC Class Codes: G06F 2209/5019 Workload prediction
• CPC Class Codes: G06F 9/5061 Partitioning or combining of resources
• CPC Class Codes: G06F 9/5083 Techniques for rebalancing the load in a distributed system
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers

Abstract

Server management data describes observed operating condition of a pool of spare servers. Based on a demand forecast of a specific target system, a dynamic allocation period is determined as a period during which the target system needs additional server resources to handle an expected demand. Based on the dynamic allocation period and server management data, a set of allocation candidates are nominated from the spare server pool, by eliminating therefrom spare servers which are likely to fail during the dynamic allocation period. An appropriate allocation candidate is then selected for allocation to the target system, such that the selected candidate will satisfy a specified requirement during its allocation period.

Claims

1. A non-transitory computer-readable medium encoded with a server allocation program that performs, when executed on a computer, a process of allocating servers to a target system including a working system and a backup system, the process comprising:

monitoring operating condition of spare servers that are deallocated from the target system and belong to a spare server pool;

storing and managing server management data which describes performance and reliability of the spare servers and the operating condition of the spare servers, including operating time of each spare server which indicates continuing run time of that spare server since startup thereof;

determining, based on demand forecast data or previously specified data about the target system, a dynamic allocation period during which the target system is expected to need allocation of the spare servers;

calculating a remaining operating time of each individual spare server from the operating time and the reliability of the spare servers described in the stored server management data;

determining whether a predicted failure time of each of the spare servers comes during the dynamic allocation period or after the dynamic allocation period, by comparing the remaining operating time of each spare server with the dynamic allocation period;

nominating a set of allocation candidates from the spare server pool, by including therein the spare servers whose predicted failure times are determined to come after the dynamic allocation period, while excluding therefrom the spare servers whose predicted failure times are determined to come during the dynamic allocation period;

selecting, based on the stored server management data, which allocation candidate to allocate to the target system, such that the selected allocation candidate enables the target system to satisfy a specified performance requirement during the dynamic allocation period;

allocating the selected allocation candidate to the working system being a group of servers currently working to provide services to clients connected to the target system;

calculating reliability that the target system as a whole, including the allocated allocation candidate in the working system, is expected to attain during the dynamic allocation period; and

further allocating one of the other allocation candidates to the backup system being another group of servers to replace some or all of the servers in the working system in case of failure thereof, when the calculated reliability of the target system falls short of a reliability requirement that the target system as a whole, including both the working and backup systems, has to satisfy.

2. The non-transitory computer-readable medium according to claim 1, wherein said nominating eliminates a spare server from the set of allocation candidates if the remaining operating time of that spare server will be exhausted before the dynamic allocation period expires.

3. The non-transitory computer-readable medium according to claim 1, wherein said nominating eliminates a spare server from the set of allocation candidates if that spare server fails to satisfy a previously specified allocation condition.

4. The non-transitory computer-readable medium according to claim 3, wherein:

the allocation condition includes a cost requirement; and

said nominating eliminates a spare server from the set of allocation candidates if that spare server fails to satisfy the cost requirement.

5. The non-transitory computer-readable medium according to claim 1, wherein:

the target system comprises multiple layers of servers; and

said selecting is performed for each of the layers.

6. The non-transitory computer-readable medium according to claim 1, wherein:

said selecting selects an allocation candidate that compensates for shortage of processing capacity of the target system during the dynamic allocation period.

7. The non-transitory computer-readable medium according to claim 6, wherein said selecting considers another allocation condition other than the performance requirement when selecting an allocation candidate.

8. The non-transitory computer-readable medium according to claim 6, wherein:

the target system comprises multiple layers of servers; and

said selecting comprises;

identifying a layer exhibiting a lowest reliability, among the layers that fall short of processing capacity,

allocating an allocation candidate with a highest reliability to the identified layer,

assessing performance attained by the target system as a result of said allocating, and

repeating a reliability-based server allocation by using the other allocation candidates if the attained performance fails to satisfy the performance requirement.

9. The non-transitory computer-readable medium according to claim 1, wherein:

the target system comprises multiple layers of servers; and

said allocating to the backup system comprises;

identifying a layer exhibiting a lowest reliability,

allocating an allocation candidate with a highest reliability to the identified layer,

assessing reliability attained by the target system as a result of said allocating, and

repeating a reliability-based server allocation by using the other allocation candidates if the attained reliability fails to satisfy the reliability requirement.

10. The non-transitory computer-readable medium according to claim 9, wherein said allocating to the backup system comprises:

selecting an allocation candidate whose performance is equal to or greater than a server belonging to the working system of the target system.

11. The non-transitory computer-readable medium according to claim 1, the process further comprising:

storing previously an alternative to either or both of the specified performance and reliability requirements; and

determining which allocation candidate to allocate to the target system by selecting an allocation candidate that satisfies the alternative, when none of the allocation candidates can satisfy the specified performance and reliability requirements.

12. The non-transitory computer-readable medium according to claim 11, wherein:

the alternative relaxes either or both of the specified performance and reliability requirements; and

said selecting uses the relaxed requirement according to the alternative.

13. The non-transitory computer-readable medium according to claim 1, the process further comprising:

producing an alternative condition to replace either or both of the specified performance and reliability requirements if said selecting is unable to select an allocation candidate that satisfies the specified performance and reliability requirements; and

presenting the alternative condition to a user of the target system.

14. The non-transitory computer-readable medium according to claim 12, the process further comprising:

producing a plurality of different alternatives to replace either or both of the specified performance and reliability requirements.

19. The non-transitory computer-readable medium according to claim 1, wherein:

the server management data includes mean time between failure (MTBF) or mean time to failure (MTTF) to describe the reliability of each of the spare servers; and

the calculating of the remaining operating time is based on the operating time and the MTBF or MTTF included in the stored server management data.

15. A method for dynamically allocating servers to a target system including a working system and a backup system, the method comprising:

monitoring operating condition of spare servers that are deallocated from the target system and belong to a spare server pool;

storing and managing server management data which describes performance and reliability of the spare servers and the operating condition of the spare servers, including operating time of each spare server which indicates continuing run time of that spare server since startup thereof;

determining, based on demand forecast data or previously specified data about the target system, a dynamic allocation period during which the target system is expected to need allocation of the spare servers;

calculating a remaining operating time of each individual spare server from the operating time and the reliability of the spare servers described in the stored server management data;

determining whether a predicted failure time of each of the spare servers comes during the dynamic allocation period or after the dynamic allocation period, by comparing the remaining operating time of each spare server with the dynamic allocation period;

nominating a set of allocation candidates from the spare server pool, by including therein the spare servers whose predicted failure times are determined to come after the dynamic allocation period, while excluding therefrom the spare servers whose predicted failure times are determined to come during the dynamic allocation period; and

selecting, based on the stored server management data, which allocation candidate to allocate to the target system, such that the selected allocation candidate enables the target system to satisfy a specified performance requirement during the dynamic allocation period;

allocating the selected allocation candidate to the working system being a group of servers currently working to provide services to clients connected to the target system;

calculating reliability that the target system as a whole, including the allocated allocation candidate in the working system, is expected to attain during the dynamic allocation period; and

further allocating one of the other allocation candidates to the backup system being another group of servers to replace some or all of the servers in the working system in case of failure thereof, when the calculated reliability of the target system falls short of a reliability requirement that the target system as a whole, including both the working and backup systems, has to satisfy.

16. An apparatus for dynamically allocating servers to a target system, the apparatus comprising:

a processing device configured to execute a procedure comprising;

monitoring operating condition of spare servers that are deallocated from the target system and belong to a spare server pool;

storing and managing server management data which describes performance and reliability of the spare servers and the operating condition of the spare servers, including operating time of each spare server which indicates continuing run time of that spare server since startup thereof;

determining, based on demand forecast data or previously specified data about the target system, a dynamic allocation period during which the target system is expected to need allocation of the spare servers;

calculating a remaining operating time of each individual spare server from the operating time and the reliability of the spare servers described in the stored server management data;

determining whether a predicted failure time of each of the spare servers comes during the dynamic allocation period or after the dynamic allocation period, by comparing the remaining operating time of each spare server with the dynamic allocation period;

nominating a set of allocation candidates from the spare server pool, by including therein the spare servers whose predicted failure times are determined to come after the dynamic allocation period, while excluding therefrom the spare servers whose predicted failure times are determined to come during the dynamic allocation period;

selecting, based on the stored server management data, which allocation candidate to allocate to the target system, such that the selected allocation candidate enables the target system to satisfy a specified performance requirement during the dynamic allocation period;

allocating the selected allocation candidate to the working system being a arum of servers currently working to provide services to clients connected to the target system;

calculating reliability that the target system as a whole, including the allocated allocation candidate in the working system, is expected to attain during the dynamic allocation period; and

further allocating one of the other allocation candidates to the backup system being another group of servers to replace some or all of the servers in the working system in case of failure thereof, when the calculated reliability of the target system falls short of a reliability requirement that the target system as a whole, including both the working and backup systems, has to satisfy.

17. A non-transitory computer-readable medium storing a program that causes a computer to execute a procedure, the procedure comprising:

calculating a remaining operating time of a server of a server pool, based on reliability data of the server;

judging whether a predicted failure occurrence timing of the server is beyond a period or within the period in which a server group is required to satisfy a certain performance, by comparing the period with the remaining operating time of the server; and

causing one or more servers of the server pool to be included in the server group based on the judging, wherein a server of the server pool is excluded from the server group when a predicted failure occurrence timing of the server is within the period and a server of the server pool is allowed to be included in the server group when a predicted failure occurrence timing of the server is beyond the period;

wherein the server group includes working servers currently working to provide services to clients connected to the server group and backup servers to replace some or all of the working servers in case of failure thereof, and said causing includes;

including one or more servers of the server pool as the working servers in the server group;

calculating reliability that the server group as a whole is expected to attain during the period; and

further including one or more servers of the server pool as the backup servers, in the server group when the calculated reliability of the server group falls short of a reliability requirement that the server group as a whole, including both the working and backup servers, has to satisfy.

18. An apparatus comprising:

a processing device configured to execute a procedure comprising;

calculating a remaining operating time of a server of a server pool, based on reliability data of the server;

judging whether a predicted failure occurrence timing of the server is beyond a period or within the period in which a server group is required to satisfy a certain performance, by comparing the period with the remaining operating time of the server; and

causing one or more servers of the server pool to be included in the server group based on the judging, wherein a server of the server pool is excluded from the server group when a predicted failure occurrence timing of the server is within the period and a server of the server pool is allowed to be included in the server group when a predicted failure occurrence timing of the server is beyond the period;

wherein the server group includes working servers currently working to provide services to clients connected to the server group and backup servers to replace some or all of the working servers in case of failure thereof, and said causing includes;

including one or more servers of the server pool as the working servers in the server group;

calculating reliability that the server group as a whole is expected to attain during the period; and

further including one or more servers of the server pool as the backup servers, in the server group when the calculated reliability of the server group falls short of a reliability requirement that the server group as a whole, including both the working and backup servers, has to satisfy.