System and method for throttling service requests having non-uniform workloads

Patent metadata

Field	Value
Publication number	US-10257288-B2
Application number	US-201414570900-A
Country	US
Kind code	B2
Filing date	Dec 15, 2014
Priority date	Dec 15, 2011
Publication date	Apr 9, 2019
Grant date	Apr 9, 2019

Abstract

A system that provides services to clients may receive and service requests, various ones of which may require different amounts of work. The system may determine whether it is operating in an overloaded or underloaded state based on a current work throughput rate, a target work throughput rate, a maximum request rate, or an actual request rate, and may dynamically adjust the maximum request rate in response. For example, if the maximum request rate is being exceeded, the maximum request rate may be raised or lowered, dependent on the current work throughput rate. If the target or committed work throughput rate is being exceeded, but the maximum request rate is not being exceeded, a lower maximum request rate may be proposed. Adjustments to the maximum request rate may be made using multiple incremental adjustments. Service request tokens may be added to a leaky token bucket at the maximum request rate.

First claim

What is claimed is:

1. A method, comprising:
performing by one or more computing devices:
receiving a plurality of service requests at a network-based service;
servicing a portion of the plurality of service requests at the network-based service based, at least in part, on a maximum allowable request rate;
determining respective numbers of units of work performed to service individual requests of the portion of the plurality of service requests at the network-based service, wherein the respective numbers of units of work performed to service the individual requests are non-uniform;
comparing, at the network-based service, an observed rate of the respective numbers of units of work performed to service the portion of the plurality of service requests with respect to a target rate determined based, at least in part, on a user specified throughput to service requests at the network-based service;
based, at least in part, on the comparison, adjusting the maximum allowable request rate; and
subsequent to said adjusting:
receiving one or more additional service requests at the network-based service; and
servicing a portion of the additional service requests at the network-based service based, at least in part, on the adjusted maximum allowable request rate.

2. The method of claim 1 ,
wherein the comparison of the observed rate with the target rate determines that the observed rate exceeds the target rate; and
wherein said adjusting comprises:
determining an amount by which to lower the maximum allowable request rate; and
lowering the maximum allowable request rate by the determined amount.

3. The method of claim 1 ,
wherein the comparison of the observed rate with the target rate determines that the observed rate is less than the target rate; and
wherein said adjusting comprises:
determining an amount by which to raise the maximum allowable request rate; and
raising the maximum request rate by the determined amount.

4. The method of claim 1 , wherein said adjusting comprises:
determining a rejection rate for those service requests of the plurality of service requests; and
dependent on the rejection rate, calculating an amount by which the maximum allowable request rate is adjusted is determined.

5. The method of claim 4 , wherein said determining the rejection rate comprises calculating a moving average for service request rejections over a plurality of sample periods.

6. The method of claim 1 ,
wherein the plurality of service requests and the additional service requests are access requests directed toward a partition of multiple partitions of a database table maintained on behalf of one or more clients at the network-based service;
wherein the method further comprises:
prior to receiving the plurality of service requests, initializing the maximum allowable request rate, wherein the maximum allowable request rate is initialized according to the user specified throughput level, and wherein the observed rate at which work was performed is a data transfer rate required to satisfy the portion of the service requests.

7. The method of claim 1 , wherein the plurality of service requests and the additional service requests are one type of service request, wherein the maximum allowable request rate is specific to accept service requests of the one type of service request, wherein the receiving, the servicing, the determining, the comparing, the adjusting, and the subsequent receiving and servicing are performed for another plurality of service requests of a different type of service request than the one type such that another maximum allowable request rate specific to the different type of service request is adjusted to service requests of the different type of service request.

8. A system, comprising:
one or more processors; and
a memory coupled to the one or more processors and storing program instructions that when executed by the one or more processors cause the one or more processors to:
receive a plurality of service requests at a network-based service;
service a portion of the plurality of service requests at the network-based service based, at least in part, on a maximum allowable request rate to service requests at the network-based service;
determine respective numbers of units of work performed to service individual requests of the portion of the plurality of service requests at the network-based service, wherein the respective numbers of units of work performed to service the individual requests are non-uniform;
compare an observed rate of the respective number of units work performed to service the portion of the service requests with respect to a target rate determined, based, at least in part, on a user specified throughput to service requests at the network-based service;
based, at least in part, on the comparison, adjust the maximum allowable request rate; and
subsequent to the adjust:
receive one or more additional service requests at the network-based service; and
service a portion of the additional service requests at the network-based service based, at least in part, on the adjusted maximum allowable request rate.

9. The system of claim 8 ,
wherein the comparison of the observed rate with the target rate determines that the observed rate exceeds the target rate for performing work; and
wherein to adjust the maximum allowable request rate, the program instructions cause the one or more processors to:
determine an amount by which to lower the maximum allowable request rate; and
lower the maximum allowable request rate by the determined amount.

10. The system of claim 8 ,
wherein the comparison of the observed rate with the target rate determines that the observed rate is less than the target rate for performing work; and
wherein to adjust the maximum allowable request rate, the program instructions cause the one or more processors to:
determine an amount by which to raise the maximum allowable request rate; and
raise the maximum allowable request rate by the determined amount.

11. The system of claim 8 , wherein to adjust the maximum allowable request rate, the program instructions cause the one or more processors to:
determine a rejection rate for those service requests of the plurality of service requests; and
dependent on the rejection rate, calculate an amount by which the maximum allowable request rate is adjusted is determined.

12. The system of claim 11 wherein to determine the rejection rate, the program instructions cause the one or more processors to calculating a moving average for service request rejections over a plurality of sample periods.

13. The system of claim 8 ,
wherein the plurality of service requests and the additional service requests are access requests directed toward a partition of multiple partitions of a database table maintained on behalf of one or more clients at the network-based service;
wherein the program instructions further cause the one or more processors to:
prior to the receipt of the plurality of service requests, initialize the maximum allowable request rate, wherein the maximum allowable request rate is initialized according to the user specified throughput level, and wherein the observed rate at which work was performed is a data transfer rate required to satisfy the portion of the service requests.

14. The system of claim 8 , wherein the plurality of service requests and the additional service requests are one type of service request, wherein the maximum allowable request rate is specific to accept service requests of the one type of service request, wherein the receive, the service, the determination, the co

Assignees

• Amazon Tech Inc

Inventors

• Xiao Wei
• Lutz David Alan
• Rath Timothy Andrew
• Maccanti Maximiliano
• Filipe Miguel Mascarenhas
• Yanacek David Craig

Classifications

• H04L41/0896

  Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities (flow or congestion control using dynamic resource allocation, e.g. in-call renegotiation, H04L47/76) · CPC title

• G06F11/3452

  Performance evaluation by statistical analysis · CPC title

• G06F2201/81

  Threshold · CPC title

• G06F2209/504

  Resource capping · CPC title

• G06F11/3485

  for I/O devices · CPC title