Self-learning green application workloads
Techniques are described for determining the energy usage of a data center and invoking one or more actions to improve the energy usage of the data center. For example, a computing system may obtain energy usage data of a data center deploying an application. The computing system may also determine, based on a comparison of the energy usage data of the data center deploying the application to a percentage of energy provided by one or more renewable energy sources to the data center, a green quotient of the application that specifies a value that indicates whether the data center deploying the application is energy efficient. The computing system may further invoke, based on the green quotient of the application that specifies a value that indicates the data center deploying the application is not energy efficient, an action to improve energy usage of the data center deploying the application.
1 . A method comprising:
obtaining, by a computing system, energy usage data of a data center deploying an application;
computing, by a computing system and based on a comparison of the energy usage data of the data center deploying the application to a percentage of energy provided by one or more renewable energy sources to the data center, a green quotient of the application that specifies a value that indicates whether the data center deploying the application is energy efficient; and
invoking, by a computing system and based on the green quotient of the application that specifies a value that indicates the data center deploying the application is not energy efficient, an action to improve energy usage of the data center deploying the application, wherein the action comprises configuring one or more devices of the data center.
2 . The method of claim 1 , wherein computing the green quotient of the application comprises:
determining, based on energy usage data of the data center when the application is running, a first green quotient that indicates the energy efficiency of the data center when the application is running;
determining, based on energy usage data of the data center when the application is not running, a second green quotient that indicates the energy efficiency of the data center when the application is not running; and
computing a delta between the first green quotient and the second green quotient;
specifying, based on comparing the delta between the first green quotient and the second green quotient to the percentage of energy provided by one or more renewable energy sources to the data center, a value for the green quotient of the application that indicates whether the data center deploying the application is energy efficient.
3 . The method of claim 2 ,
wherein the energy usage data comprises a current energy usage of the data center and a total energy usage of the data center if the data center were to run at full capacity;
wherein determining the first green quotient comprises:
determining a first energy quotient, wherein the first energy quotient of the data center indicates a percentage of the total energy usage consumed by the current energy usage of the data center when the application is running, and
comparing the first energy quotient to a percentage of energy provided by one or more renewable energy sources to the data center; and
wherein determining the second green quotient comprises:
determining a second energy quotient, wherein the second energy quotient of the data center indicates a percentage of the total energy usage consumed by the current energy usage of the data center when the application is not running, and
comparing the second energy quotient to the percentage of energy provided by one or more renewable energy sources to the data center.
4 . The method of claim 2 , wherein the first green quotient comprises an average of a plurality of green quotients of data center when the application is running that were determined over a period of time.
5 . The method of claim 2 , wherein specifying, based on comparing the delta between the first green quotient and the second green quotient to a percentage of energy provided by one or more renewable energy sources to the data center, the value for the green quotient of the application that indicates whether the data center deploying the application is energy efficient comprises:
determining that the delta between the first green quotient and the second green quotient is less than or equal to the percentage of energy provided by one or more renewable energy sources to the data center; and
in response to determining that the delta between the first green quotient and the second green quotient is less than or equal to the percentage of energy provided by one or more renewable energy sources to the data center, specifying a value of the green quotient of the application that indicates the data center deploying the application is energy efficient.
6 . The method of claim 2 , wherein specifying, based on comparing the delta between the first green quotient and the second green quotient to a percentage of energy provided by one or more renewable energy sources to the data center, the value for the green quotient of the application that indicates whether the data center deploying the application is energy efficient comprises:
determining that the delta between the first green quotient and the second green quotient is greater than the percentage of energy provided by one or more renewable energy sources to the data center; and
in response to determining that the delta between the first green quotient and the second green quotient is greater than the percentage of energy provided by one or more renewable energy sources to the data center, specifying a value of the green quotient of the application that indicates the data center deploying the application is not energy efficient.
7 . The method of claim 1 , wherein configuring the one or more devices of the data center comprises one or more of:
automatically deploying the application to one or more select servers;
migrating the application from a first number of hosts to a second number of hosts that is fewer than the first number of hosts;
activating a cooling system of the data center deploying the application; or
disabling workload scaling.
8 . The method of claim 1 ,
wherein computing the green quotient of the application that specifies the value that indicates whether the data center deploying the application is energy efficient comprises determining, based on a pattern of the energy usage of the data center, that the data center deploying the application is not energy efficient, and
wherein invoking the action to improve energy usage of the data center comprises invoking, in response to determining that the data center deploying the application is not energy efficient, the action to improve energy usage of the data center before a next determining of whether the data center deploying the application is not energy efficient.
9 . The method of claim 1 , wherein invoking the action to improve energy usage of the data center comprises:
configuring a processing unit of a network interface card in a server of the data center to perform the action.
10 . The method of claim 1 , wherein invoking the action to improve energy usage of the data center comprises:
outputting an alert indicating to an administrator of the data center that the data center deploying the application is not energy efficient.
11 . A computing system comprising:
a memory; and
one or more processors in communication with the memory, the one or more processors configured to:
obtain energy usage data of a data center deploying an application;
compute, based on a comparison of the energy usage data of the data center deploying the application to a percentage of energy provided by one or more renewable energy sources to the data center, a green quotient of the application that specifies a value that indicates whether the data center deploying the application is energy efficient; and
invoke, based on the green quotient of the application that specifies a value that indicates the data center deploying the application is not energy efficient, an action to improve energy usage of the data center deploying the application, wherein the action comprises configuring one or more devices of the data center.
12 . The computing system of claim 11 , wherein to compute the green quotient of the application, the one or more processors are configured to:
determine, based on energy usage data of the data center when the application is running, a first green quotient that indicates the energy efficiency of the data center when the application is running;
determine, based on energy usage data of the data center when the application is not running, a second green quotient that indicates the energy efficiency of the data center when the application is not running; and
computing a delta between the first green quotient and the second green quotient;
specify, based on comparing the delta between the first green quotient and the second green quotient to the percentage of energy provided by one or more renewable energy sources to the data center, the value for the green quotient of the application that indicates whether the data center deploying the application is energy efficient.
13 . The computing system of claim 12 ,
wherein the energy usage data comprises a current energy usage of the data center and a total energy usage of the data center if the data center were to run at full capacity;
wherein to determine the first green quotient, the one or more processors are configured to:
determine a first energy quotient, wherein the first energy quotient of the data center indicates a percentage of the total energy usage consumed by the current energy usage of the data center when the application is running, and
compare the first energy quotient to a percentage of energy provided by one or more renewable energy sources to the data center; and
wherein to determine the second green quotient, the one or more processors are configured to:
determine a second energy quotient, wherein the second energy quotient of the data center indicates a percentage of the total energy usage consumed by the current energy usage of the data center when the application is not running, and
compare the second energy quotient to the percentage of energy provided by one or more renewable energy sources to the data center.
14 . The computing system of claim 12 , wherein the second green quotient comprises an average of a plurality of green quotients of data center when the application is running that were determined over a period of time.
15 . The computing system of claim 12 , wherein to specify, based on comparing the delta between the first green quotient and the second green quotient to a percentage of energy provided by one or more renewable energy sources to the data center, the value for the green quotient of the application that indicates whether the data center deploying the application is energy efficient, the one or more processors are configured to:
determine that the delta between the first green quotient and the second green quotient is less than or equal to the percentage of energy provided by one or more renewable energy sources to the data center; and
in response to determining that the delta between the first green quotient and the second green quotient is less than or equal to the percentage of energy provided by one or more renewable energy sources to the data center, specify a value of the green quotient of the application that indicates the data center deploying the application is energy efficient.
16 . The computing system of claim 12 , wherein to specify, based on comparing the delta between the first green quotient and the second green quotient to a percentage of energy provided by one or more renewable energy sources to the data center, the value for the green quotient of the application that indicates whether the data center deploying the application is energy efficient, the one or more processors are configured to:
determine that the delta between the first green quotient and the second green quotient is greater than the percentage of energy provided by one or more renewable energy sources to the data center; and
in response to determining that the delta between the first green quotient and the second green quotient is greater than the percentage of energy provided by one or more renewable energy sources to the data center, specify a value of the green quotient of the application that indicates the data center deploying the application is not energy efficient.
17 . The computing system of claim 11 , wherein to configure the one or more devices of the data center, the one or more processors are configured to perform one or more of:
automatically deploy the application to one or more select servers;
migrate the application from a first number of hosts to a second number of hosts that is fewer than the first number of hosts;
activate a cooling system of the data center deploying the application; or
disable workload scaling.
18 . The computing system of claim 11 ,
wherein to compute the green quotient of the application that specifies the value that indicates whether the data center deploying the application is energy efficient, the one or more processors are further configured to determine, based on a pattern of the energy usage of the data center, that the data center deploying the application is not energy efficient, and
wherein to invoke the action to improve energy usage of the data center, the one or more processors are further configured to invoke, in response to determining that the data center deploying the application is not energy efficient, the action to improve energy usage of the data center before a next determining of whether the data center deploying the application is not energy efficient.
19 . The computing system of claim 11 , wherein to invoke the action to improve energy usage of the data center, the one or more processors are configured to:
configure a processing unit of a network interface card in a server of the data center to perform the action.
20 . Non-transitory computer-readable media having instructions that, when executed, cause one or more processors to:
obtain energy usage data of a data center deploying an application;
compute, based on a comparison of the energy usage data of the data center deploying the application to a percentage of energy provided by one or more renewable energy sources to the data center, a green quotient of the application that specifies a value that indicates whether the data center deploying the application is energy efficient; and
invoke, based on the green quotient of the application that specifies a value that indicates the data center deploying the application is not energy efficient, an action to improve energy usage of the data center deploying the application, wherein the action comprises configuring one or more devices of the data center.