| 摘要 |
Disclosed are systems, methods and non-transitory computer-readable storage media for managing resources between a web server and an on-demand compute environment. A system configured to practice the method receives, at the on-demand compute environment, an indication that web traffic directed to a web server should at least be partially served via the on-demand compute environment, provisions compute resources within the on-demand compute environment to respond to web traffic for the web server based on the indication to yield a provisioned on-demand compute environment. The system serves web traffic at the provisioned on-demand compute environment based on an established routing of the web traffic from the web server to the provisioned on-demand compute environment such that the use of the provisioned on-demand compute environment for the web traffic is transparent. After serving the web traffic via the provisioned resources, the system can release the provisioned compute resources for other uses. |
| 主权项 |
1. A method of redirecting web traffic to an on-demand compute environment, the method comprising:
determining that an increase of projected web traffic has exceeded a threshold and should be at least partially served via the on-demand compute environment, wherein the on-demand compute environment has a separate workload manager from a workload manager that manages a web server; assigning an agent to manage redirection of a portion of the projected web traffic to be served via the on-demand compute environment, wherein the agent does not process the projected web traffic as would the web server; determining whether content is able to be mirrored in the on-demand compute environment to yield a determination, wherein:
when the determination indicates that the content is able to be mirrored, provisioning compute resources within the on-demand compute environment by mirroring the content from the web server to enable the on-demand compute environment to respond to the portion of the projected web traffic to yield provisioned compute resources;when the determination indicates that the content is not able to be mirrored, identifying, selecting and provisioning substitute content via negotiation between the on-demand workload manager and the separate workload manager to yield substitute provisioned compute resources; intercepting, via the agent, the portion of the projected web traffic comprising multiple web access requests to the web server; redirecting, via the agent, the portion of the projected web traffic to be served by the provisioned compute resources and the substitute provisioned compute resources such that the redirecting is transparent to users initiating the portion of the projected web traffic; and in response to determining the projected web traffic has decreased below a certain threshold, automatically transferring web requests to the web server and un-provisioning one of the provisioned compute resources and the substitute provisioned compute resources in the on-demand compute environment. |
