Systems, methods, and media for online server workload management
View Patent ↗Methods, using a hardware processor, for online server workload management are provided, comprising: receiving information regarding client device requests; determining, using a hardware processor, a workload distribution for the requests based on electricity cost and carbon footprint of one or more data centers using Lyapunov optimization; sending the workload distribution to the one or more data centers; and causing servers in the one or more data center to be active or inactive based on the workload distribution. Systems are provided, comprising at least one hardware processor configured to: receive information regarding client device requests; determine a workload distribution for the requests based on electricity cost and carbon footprint of one or more data centers using Lyapunov optimization; send the workload distribution to the one or more data centers; and cause servers in the one or more data center to be active or inactive based on the workload distribution.
1. A method, using a hardware processor, for online server workload management, the method comprising:
receiving information regarding client device requests;
determining, using a hardware processor, a workload distribution for the requests based on electricity cost and carbon footprint of one or more data centers using Lyapunov optimization, wherein the workload distribution minimizes a combination of a cost for electricity for a plurality of servers, and a total excess carbon footprint for the plurality of servers beyond an average carbon footprint multiplied by a total carbon footprint of the plurality of servers and another plurality of servers;
sending the workload distribution to the one or more data centers; and
causing servers in the one or more data center to be active or inactive based on the workload distribution.
2. The method of claim 1 , wherein the information regarding the client device requests are received from at least one proxy server.
3. The method of claim 1 , wherein the workload distribution is sent to the one or more data centers via at least one proxy server.
4. A system for online server workload management comprising:
at least one hardware processor configured to:
receive information regarding client device requests;
determine a workload distribution for the requests based on electricity cost and carbon footprint of one or more data centers using Lyapunov optimization, wherein the workload distribution minimizes a combination of a cost for electricity for a plurality of servers, and a total excess carbon footprint for the plurality of servers beyond an average carbon footprint multiplied by a total carbon footprint of the plurality of servers and another plurality of servers;
send the workload distribution to the one or more data centers; and
cause servers in the one or more data center to be active or inactive based on the workload distribution.
5. The system of claim 4 , wherein the information regarding the client device requests are received from at least one proxy server.
6. The system of claim 4 , wherein the workload distribution is sent to the one or more data centers via at least one proxy server.
7. A non-transitory computer readable medium containing computer-executable instructions that, when executed by a processor, cause the processor to perform methods for online server workload management, the method comprising:
receiving information regarding client device requests;
determining a workload distribution for the requests based on electricity cost and carbon footprint of one or more data centers using Lyapunov optimization, wherein the workload distribution minimizes a combination of a cost for electricity for a plurality of servers, and a total excess carbon footprint for the plurality of servers beyond an average carbon footprint multiplied by a total carbon footprint of the plurality of servers and another plurality of servers;
sending the workload distribution to the one or more data centers; and
causing servers in the one or more data center to be active or inactive based on the workload distribution.
8. The non-transitory computer readable medium of claim 7 , wherein the information regarding the client device requests are received from at least one proxy server.
9. The non-transitory computer readable medium of claim 7 , wherein the workload distribution is sent to the one or more data centers via at least one proxy server.