Cluser-aware networking fabric update system
A networking fabric update system includes a cluster coordination subsystem coupled to servers and switches. The cluster coordination subsystem identifies a networking topology provided by connections between the servers and switches, and forecasts workload operations for the servers based on, for example, current workload operations and historical workload operations of the servers. Based on the networking topology and the workload operations forecast for the servers, the cluster coordination subsystem updates a first subset of the switches and, following the updating of the first subset of the switches, updates a second subset of the switches. The updating of the first subset of the switches may include determining that current workload operations for the servers during a current time period are above a workload threshold for the first subset of the switches, and scheduling and updating the first subset of the switches during a subsequent time period.
1. A networking fabric update system, comprising:
a plurality of server computing devices;
a plurality of switch computing devices; and
a cluster coordination subsystem that is coupled to each of the plurality of server computing devices and each of the plurality of switch computing devices, wherein the cluster coordination subsystem includes one or more processors that are configured to:
identify a networking topology provided by connections between the plurality of server computing devices and the plurality of switch computing devices;
forecast workload operations for the plurality of server computing devices;
update, based on the networking topology and the workload operations forecast for the plurality of server computing devices, a first subset of the plurality of switch computing devices; and
update, following the updating of the first subset of the plurality of switch computing devices, a second subset of the plurality of switch computing devices.
2. The system of claim 1 , wherein the networking topology provided by connections between the plurality of server computing devices and the plurality of switch computing devices includes one of: a fully-converged networking topology or a non-converged networking topology.
3. The system of claim 1 , wherein the forecasting the workload operations for the plurality of server computing devices includes:
monitoring at least one Network Interface Controller (NIC) device in each server computing device; and
forecasting, based on network traffic provided via the at least one Network Interface Controller (NIC) device in each server computing device, the workload operations for the plurality of server computing devices.
4. The system of claim 1 , wherein the forecasting the workload operations for the plurality of server computing devices includes:
accessing a historical workload operation database that includes historical workload operations for the plurality of server computing devices; and
forecasting, based on the historical workload operations for the plurality of server computing devices, the workload operations for the plurality of server computing devices.
5. The system of claim 1 , wherein the updating the first subset of the plurality of switch computing devices based on the networking topology and the workload operations forecast for the plurality of server computing devices includes:
determining that current workload operations for the plurality of server computing devices during a current time period are above a workload threshold for the first subset of the plurality of switch computing devices;
scheduling a subsequent time period for performing the updating the first subset of the plurality of switch computing devices; and
updating the first subset of the plurality of switch computing devices during the subsequent time period.
6. The system of claim 5 , wherein the cluster coordination subsystem is configured to:
provide switch computing device update information on each of the first subset of the plurality of switch computing devices prior to the subsequent time period.
7. An Information Handling System (IHS), comprising:
a processing system that includes one or more processors; and
a memory system that is coupled to the processing system and that includes instructions that, when executed by the one or more processors, cause the processing system to provide a networking fabric update engine that is configured to:
identify a networking topology provided by connections between a plurality of server computing devices and a plurality of switch computing devices;
forecast workload operations for the plurality of server computing devices;
update, based on the networking topology and the workload operations forecast for the plurality of server computing devices, a first subset of the plurality of switch computing devices; and
update, following the updating of the first subset of the plurality of switch computing devices, a second subset of the plurality of switch computing devices.
8. The IHS of claim 7 , wherein the networking topology provided by connections between the plurality of server computing devices and the plurality of switch computing devices includes one of: a fully-converged networking topology or a non-converged networking topology.
9. The IHS of claim 7 , wherein the forecasting the workload operations for the plurality of server computing devices includes:
monitoring at least one Network Interface Controller (NIC) device in each server computing device; and
forecasting, based on network traffic provided via the at least one Network Interface Controller (NIC) device in each server computing device, the workload operations for the plurality of server computing devices.
10. The IHS of claim 7 , wherein the forecasting the workload operations for the plurality of server computing devices includes:
accessing a historical workload operation database that includes historical workload operations for the plurality of server computing devices; and
forecasting, based on the historical workload operations for the plurality of server computing devices, the workload operations for the plurality of server computing devices.
11. The IHS of claim 7 , wherein the updating the first subset of the plurality of switch computing devices based on the networking topology and the workload operations forecast for the plurality of server computing devices includes:
determining that current workload operations for the plurality of server computing devices during a current time period are above a workload threshold for the first subset of the plurality of switch computing devices;
scheduling a subsequent time period for performing the updating the first subset of the plurality of switch computing devices; and
updating the first subset of the plurality of switch computing devices during the subsequent time period.
12. The IHS of claim 11 , wherein the cluster coordination subsystem is configured to:
provide switch computing device update information on each of the first subset of the plurality of switch computing devices prior to the subsequent time period.
13. The IHS of claim 7 , wherein the first subset of the plurality switch computing devices include a first management networking switch device providing access to a management network and a first storage networking switch device providing access to a storage network, and wherein the second subset of the plurality switch computing devices include a second management networking switch device providing access to the management network and a second storage networking switch device providing access to the storage network.
14. A method for updating a networking fabric, comprising:
identifying, by one or more processors included in a cluster coordination subsystem, a networking topology provided by connections between a plurality of server computing devices and a plurality of switch computing devices;
forecasting, by the one or more processors included in the cluster coordination subsystem, workload operations for the plurality of server computing devices;
updating, by the one or more processors included in the cluster coordination subsystem based on the networking topology and the workload operations forecast for the plurality of server computing devices, a first subset of the plurality of switch computing devices; and
updating, by the one or more processors included in the cluster coordination subsystem following the updating of the first subset of the plurality of switch computing devices, a second subset of the plurality of switch computing devices.
15. The method of claim 14 , wherein the networking topology provided by connections between the plurality of server computing devices and the plurality of switch computing devices includes one of: a fully-converged networking topology or a non-converged networking topology.
16. The method of claim 14 , wherein the forecasting the workload operations for the plurality of server computing devices includes:
monitoring at least one Network Interface Controller (NIC) device in each server computing device; and
forecasting, based on network traffic provided via the at least one Network Interface Controller (NIC) device in each server computing device, the workload operations for the plurality of server computing devices.
17. The method of claim 14 , wherein the forecasting the workload operations for the plurality of server computing devices includes:
accessing a historical workload operation database that includes historical workload operations for the plurality of server computing devices; and
forecasting, based on the historical workload operations for the plurality of server computing devices, the workload operations for the plurality of server computing devices.
18. The method of claim 14 , wherein the updating the first subset of the plurality of switch computing devices based on the networking topology and the workload operations forecast for the plurality of server computing devices includes:
determining that current workload operations for the plurality of server computing devices during a current time period are above a workload threshold for the first subset of the plurality of switch computing devices;
scheduling a subsequent time period for performing the updating the first subset of the plurality of switch computing devices; and
updating the first subset of the plurality of switch computing devices during the subsequent time period.
19. The method of claim 18 , further comprising:
providing, by the cluster coordination subsystem, switch computing device update information on each of the first subset of the plurality of switch computing devices prior to the subsequent time period.
20. The method of claim 14 , wherein the first subset of the plurality switch computing devices include a first management networking switch device providing access to a management network and a first storage networking switch device providing access to a storage network, and wherein the second subset of the plurality switch computing devices include a second management networking switch device providing access to the management network and a second storage networking switch device providing access to the storage network.