Determining a deployment schedule for operations performed on devices using device dependencies and predicted workloads
An apparatus comprises a processing device configured to generate a model characterizing relationships between devices in an information technology infrastructure, to build a device dependency chain for the devices based at least in part on the generated model, and to predict workload for each of the devices in each of two or more time slots of a given time period. The processing device is also configured to determine a deployment schedule for the devices based at least in part on the device dependency chain and the predicted workload for each of the devices in each of the two or more time slots of the given time period, to identify an operation to be performed on the devices, and to utilize the deployment schedule to select ones of the devices on which to perform the identified operation in each of the two or more time slots of the given time period.
1. An apparatus comprising:
at least one processing device comprising a processor coupled to a memory;
the at least one processing device being configured to perform steps of:
generating a model of a plurality of devices in an information technology infrastructure, the generated model characterizing relationships between the plurality of devices;
building a device dependency chain for the plurality of devices based at least in part on the generated model characterizing the relationships between the plurality of devices;
predicting workload for each of the plurality of devices in each of two or more time slots of a given time period;
determining a deployment schedule for the plurality of devices based at least in part on the device dependency chain and the predicted workload for each of the plurality of devices in each of the two or more time slots of the given time period;
identifying an operation to be performed on the plurality of devices; and
utilizing the deployment schedule to select ones of the plurality of devices on which to perform the identified operation in each of the two or more time slots of the given time period.
2. The apparatus of claim 1 wherein the plurality of devices comprise servers in a data center.
3. The apparatus of claim 1 wherein the plurality of devices comprise Internet of Things devices.
4. The apparatus of claim 1 wherein the model of the plurality of devices comprises a directed weighted graph with nodes representing the plurality of devices and edges between the nodes representing call relationships between the plurality of devices.
5. The apparatus of claim 4 wherein a given edge weight associated with a given edge between a first node and a second node in the directed weighted graph is based at least in part on a number of requests between the first node and the second node and an amount of data exchanged between the first node and the second node.
6. The apparatus of claim 5 wherein the given edge weight comprise a normalized edge weight.
7. The apparatus of claim 1 wherein building the device dependency chain for the plurality of devices comprises computing an overall dependency score for each of the plurality of devices, the overall dependency score for a given device comprising a sum of pairwise dependency scores characterizing activity between the given device and each of at least a subset of other ones of the plurality of devices that have call relationships with the given device.
8. The apparatus of claim 1 wherein the given time period comprises a day, and the two or more time slots comprise hours in the day.
9. The apparatus of claim 1 wherein predicting the workload for each of the plurality of devices in each of the two or more time slots of the given time period comprises utilizing a machine learning regression analysis.
10. The apparatus of claim 9 wherein the machine learning regression analysis comprises a multivariate linear regression analysis.
11. The apparatus of claim 10 wherein the multivariate linear regression analysis is based at least in part on two or more of processor utilization of the plurality of devices, memory utilization of the plurality of devices, availability of the plurality of devices, response time of the plurality of devices, and disk utilization of the plurality of devices.
12. The apparatus of claim 1 wherein determining the deployment schedule comprises, for a given one of the two or more time slots of the given time period:
identifying a subset of the devices whose lowest predicted workload is in the given time slot; and
selecting an order in which the identified subset of the devices whose lowest predicted workload is in the given time slot in decreasing order of dependency scores obtained from the device dependency chain.
13. The apparatus of claim 1 wherein the plurality of devices host an application and wherein the identified operation to be performed on the plurality of devices comprises updating the application.
14. The apparatus of claim 1 wherein the deployment schedule specifies a sequence in which the identified operation is to be performed on the plurality of devices.
15. A computer program product comprising a non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device to perform steps of:
generating a model of a plurality of devices in an information technology infrastructure, the generated model characterizing relationships between the plurality of devices;
building a device dependency chain for the plurality of devices based at least in part on the generated model characterizing the relationships between the plurality of devices;
predicting workload for each of the plurality of devices in each of two or more time slots of a given time period;
determining a deployment schedule for the plurality of devices based at least in part on the device dependency chain and the predicted workload for each of the plurality of devices in each of the two or more time slots of the given time period;
identifying an operation to be performed on the plurality of devices; and
utilizing the deployment schedule to select ones of the plurality of devices on which to perform the identified operation in each of the two or more time slots of the given time period.
16. The computer program product of claim 15 wherein determining the deployment schedule comprises, for a given one of the two or more time slots of the given time period:
identifying a subset of the devices whose lowest predicted workload is in the given time slot; and
selecting an order in which the identified subset of the devices whose lowest predicted workload is in the given time slot in decreasing order of dependency scores obtained from the device dependency chain.
17. The computer program product of claim 15 wherein the plurality of devices host an application and wherein the identified operation to be performed on the plurality of devices comprises updating the application.
18. A method comprising:
generating a model of a plurality of devices in an information technology infrastructure, the generated model characterizing relationships between the plurality of devices;
building a device dependency chain for the plurality of devices based at least in part on the generated model characterizing the relationships between the plurality of devices;
predicting workload for each of the plurality of devices in each of two or more time slots of a given time period;
determining a deployment schedule for the plurality of devices based at least in part on the device dependency chain and the predicted workload for each of the plurality of devices in each of the two or more time slots of the given time period;
identifying an operation to be performed on the plurality of devices; and
utilizing the deployment schedule to select ones of the plurality of devices on which to perform the identified operation in each of the two or more time slots of the given time period;
wherein the method is performed by at least one processing device comprising a processor coupled to a memory.
19. The method of claim 18 wherein determining the deployment schedule comprises, for a given one of the two or more time slots of the given time period:
identifying a subset of the devices whose lowest predicted workload is in the given time slot; and
selecting an order in which the identified subset of the devices whose lowest predicted workload is in the given time slot in decreasing order of dependency scores obtained from the device dependency chain.
20. The method of claim 18 wherein the plurality of devices host an application and wherein the identified operation to be performed on the plurality of devices comprises updating the application.