Isolating Resources and Performance in a Database Management System

• US 20140207740A1
• Filed: 01/23/2013
• Published: 07/24/2014
• Est. Priority Date: 01/23/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Narasayya, Vivek R., Syamala, Manoj A., Chaudhuri, Surajit, Chandramouli, Badrish, Li, Feng, Das, Sudipto, Park, Hyunjung
• Granted Patent: US 9,477,710 B2
• Time in Patent Office: Days
• US Class Current: 707/688
• CPC Class Codes: G06F 11/302 where the computing system component is a software system
• 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 11/3485 for I/O devices
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 16/21 Design, administration or maintenance of databases
• CPC Class Codes: G06F 16/217 Database tuning G06F16/2282 takes precedence; database performance monitoring G06F11/3409
• CPC Class Codes: G06F 16/2379 Updates performed during online database operations; commit processing
• CPC Class Codes: G06F 16/24561 Intermediate data storage techniques for performance improvement
• CPC Class Codes: G06F 2201/80 Database-specific techniques
• CPC Class Codes: G06F 2201/815 Virtual middleware or OS functionality using virtual machines to implement generic software techniques for error detection or fault masking G06F11/1484

Abstract

Techniques for tenant performance isolation in a multiple-tenant database management system are described. These techniques may include providing a reservation of server resources. The server resources reservation may include a reservation of a central processing unit (CPU), a reservation of Input/Output throughput, and/or a reservation of buffer pool memory or working memory. The techniques may also include a metering mechanism that determines whether the resource reservation is satisfied. The metering mechanism may be independent of an actual resource allocation mechanism associated with the server resource reservation.

Claims

1. A computer-implemented method comprising:

providing, in a database server process, a reservation of one or more server resources for a tenant, the one or more server resources including at least one of a central processing unit (CPU), memory, and an Input/Output (I/O) throughput associated with a server; and

metering the reservation to determine whether the reservation for the tenant is met.

2. The computer-implemented method of claim 1, wherein the database server process hosts multiple databases of multiple tenants, and the reservation includes at least an amount of the one or more server resources.

3. The computer-implemented method of claim 1, wherein the reservation of the one or more server resources comprises a reservation of a CPU utilization with respect to a metering interval.

4. The computer-implemented method of claim 3, wherein the CPU utilization specifies a percentage of CPU time for which a task of the tenant runs on a core of a CPU of the server.

5. The computer-implemented method of claim 1, wherein the reservation of the one or more server resources comprises a reservation of a CPU time associated with the server, and wherein the metering the reservation comprises:

determining that the tenant fails to utilize the CPU time in response to determining that the tenant has a running task or a runnable task on a core of the server; and

establishing an accountability of the server for the tenant.

6. The computer-implemented method of claim 1, wherein the reservation of the one or more server resources comprises a reservation of an I/O throughput (IOPS) amount of an IOPS capacity of the server.

7. The computer-implemented method of claim 1, wherein the reservation of the one or more server resources comprises a reservation of at least one of a buffer pool and a working memory of the server.

8. The computer-implemented method of claim 1, wherein the metering is independent from the providing the reservation for the tenant.

9. The computer-implemented method of claim 1, wherein the metering the reservation comprises metering the reservation based on a metering algorithm specific to an individual resource of the one or more server resources.

10. The computer-implemented method of claim 1, further comprising:

enforcing a maximum limit on resource usages of the one or more server resources for the tenant, the resource usages being associated with a predetermined relative priority that is assigned to the tenant.

11. One or more computer-readable media storing computer-executable instructions that, when executed on one or more processors, causes the one or more processors to perform acts comprising:

promising an allocation including at least an amount of one or more server resources for a tenant in a database server process; and

metering the allocation to determine whether the tenant is allocated the amount of the one or more server resources.

12. The one or more computer-readable media of claim 11, wherein the one or more server resources include at least one of a central processing unit (CPU), memory and or an Input/Output (I/O) throughput at a server associated with the database server process.

13. The one or more computer-readable media of claim 11, wherein the amount of the one or more server resources comprises an amount of I/O throughput associated with a server during a metering interval, and the metering the allocation comprising:

determining that an I/O throughput associated with a pending I/O request of the tenant is less than the amount of the I/O throughput, and

establishing an accountability of the server for the tenant.

14. The one or more computer-readable media of claim 11, wherein the amount of the one or more server resources comprises an amount of CPU time for which a task of the tenant runs on a core of a CPU of a server associated with the database server process, and the acts further comprise:

determining whether a workload of the tenant has a sufficient demand to consume the amount of CPU time.

15. The one or more computer-readable media of claim 14, wherein the acts further comprise:

determining whether the server fails to allocate the amount of CPU time; and

in response to determining that the server fails to allocate the amount of the CPU time and that the workload of the tenant has the sufficient demand to consume the amount of CPU time, establishing an accountability of the server for the tenant.

16. The one or more computer-readable media of claim 11, wherein the amount of the one or more server resources comprises an amount of a CPU time associated with the server, and the metering the allocation comprising:

determining that the tenant fails to utilize the CPU time;

determining that the tenant has a running task or a runnable task on a core of the server; and

establishing an accountability of the server for the tenant.

17. A computing device comprising:

one or more processors; and

memory to maintain a plurality of components executable by the one or more processors, the plurality of components comprising;

a reservation module configured to promise a reservation including at least an amount of a server resource associated with a server to a tenant in a database server process that hosts multiple databases of multiple tenant including the tenant, and

a metering module configured to;

audit the reservation, and

establish an accountability of the server for the tenant when the tenant fails to receive the amount of the server resource.

18. The computing device of claim 17, wherein the server resource comprises at least one of a central processing unit (CPU), memory and an Input/Output (I/O) throughput of the server.

19. The computing device of claim 17, wherein the amount of the server resource comprises a percentage of CPU time for which a task of the tenant runs on a core of a CPU of the server, and the metering module is further configured to:

determine whether a workload of the tenant has a sufficient demand to consume the percentage of CPU time.

20. The computing device of claim 17, wherein the scheduling module is further configured to enable a maximum limit on resource usages of the server resource for the tenant, the maximum limit being determined based at least in part on a predetermined relative priority associated with the tenant.