Scalable edge computing

Abstract

There is disclosed in one example an application-specific integrated circuit (ASIC), including: an artificial intelligence (AI) circuit; and circuitry to: identify a flow, the flow including traffic diverted from a core cloud service of a network to be serviced by an edge node closer to an edge of the network than to the core of the network; receive telemetry related to the flow, the telemetry including fine-grained and flow-level network monitoring data for the flow; operate the AI circuit to predict, from the telemetry, a future service-level demand for the edge node; and cause a service parameter of the edge node to be tuned according to the prediction.

What is claimed is: 
     
         1 . An application-specific integrated circuit (ASIC), comprising:
 an artificial intelligence (AI) circuit; and   circuitry to:
 identify a flow, the flow comprising traffic diverted from a core cloud service of a network to be serviced by an edge node closer to an edge of the network than to the core of the network; 
 receive telemetry related to the flow, the telemetry comprising fine-grained and flow-level network monitoring data for the flow; 
 operate the AI circuit to predict, from the telemetry, a future service-level demand for the edge node; and 
 cause a service parameter of the edge node to be tuned according to the prediction. 
   
     
     
         2 . The ASIC of  claim 1 , wherein the AI circuit comprises a neural network. 
     
     
         3 . The ASIC of  claim 1 , wherein the network is a 5G network. 
     
     
         4 . The ASIC of  claim 1 , wherein the network is a wide area network (WAN). 
     
     
         5 . The ASIC of  claim 1 , wherein the telemetry includes internet of things (IoT) telemetry. 
     
     
         6 . The ASIC of  claim 1 , wherein the ASIC further comprises circuitry to instigate a network repair according to the telemetry. 
     
     
         7 . The ASIC of  claim 1 , wherein the ASIC further comprises circuitry to instigate network self-healing according to the telemetry. 
     
     
         8 . A network switch, comprising:
 a high-speed switching circuit, including one or more ingress ports and one or more egress ports, including circuitry to program at least one egress port as a diverted port to divert traffic from a cloud service node to an edge node located closer to an edge of a cloud network than to the cloud service node;   an artificial intelligence (AI) circuit; and   a programmable control circuit including circuitry to:
 monitor flows of the high-speed switching circuit, and identify a flow to be diverted via the diverted port; 
 receive telemetry about the flow; 
 operate the AI circuit to predict a future network condition for the edge node; and 
 cause a parameter of the edge node to be tuned according to the prediction. 
   
     
     
         9 . The network switch of  claim 9 , wherein the programmable control circuit is an application-specific integrated circuit (ASIC). 
     
     
         10 . The network switch of  claim 9 , wherein the programmable control circuit comprises an intellectual property (IP) block. 
     
     
         11 . The network switch of  claim 9 , wherein the programmable control circuit comprises a field-programmable gate array (FPGA). 
     
     
         12 . The network switch of  claim 9 , wherein the AI circuit comprises a neural network. 
     
     
         13 . The network switch of  claim 9 , wherein the network is a 5G network. 
     
     
         14 . The network switch of  claim 9 , wherein the network is a wide area network (WAN). 
     
     
         15 . The network switch of  claim 9 , wherein the telemetry includes internet of things (IoT) telemetry. 
     
     
         16 . The network switch of  claim 9 , wherein the ASIC further comprises circuitry to instigate a network repair according to the telemetry. 
     
     
         17 . The network switch of  claim 9 , wherein the ASIC further comprises circuitry to instigate network self-healing according to the telemetry. 
     
     
         18 . A cloud service controller, comprising:
 a control circuit, including a machine learning (ML) circuit;   network analysis software to analyze, via the ML circuit, a cloud network, including one or more edge processing nodes to handle traffic flows diverted from a primary cloud service node to an edge processing node, and automatically predictively scale cloud network resources, including the one or more edge processing nodes, according to the analysis.   
     
     
         19 . The cloud service controller of  claim 18 , wherein the ML circuit comprises a neural network. 
     
     
         20 . The cloud service controller of  claim 18 , wherein the one or more edge processing nodes are physically located closer to an edge of the cloud network than to the primary cloud service node.