Artificial intelligence delivery edge network

Abstract

Approaches, techniques, and mechanisms are disclosed for accessing AI services from one region to another region. An artificial intelligence (AI) service director is configured with mappings from domain names of AI cloud engines to IP addresses of edge nodes of an AI delivery edge network. The AI cloud engines are located in an AI source region. The AI delivery edge network is deployed in a non-AI-source region. An AI application, which accesses AI services using a domain name of an AI cloud engine in the AI cloud engines located in the AI source region, is redirected to an edge node in the edge nodes of the AI delivery edge network located in the non-AI-source region. The AI application is hosted in the non-AI-source region. The AI services is then provided, by way of the edge node located in the non-AI-source region, to the AI application.

What is claimed is: 
     
         1 . A computer-implemented method comprising:
 receiving, from an artificial intelligence (AI) application located in a non-AI-source region by an edge node in one or more edge nodes of an AI delivery edge network, an AI service request directed to an AI cloud engine located in an AI source region;   sending, by the edge node to an AI service proxy agent of the AI cloud engine, the AI service request, wherein the AI service request is forwarded by the AI service proxy agent to the AI cloud engine;   receiving, by the edge node from the AI service proxy agent, an AI service response to the AI service request, wherein the AI service response is generated by the AI cloud engine and forwarded by the AI service proxy agent;   causing the AI application to invoke an AI service to train a prediction model locally at the edge node.   
     
     
         2 . The method of  claim 1 , wherein the AI service is provided in real time or near real time by one or more AI cloud engines in the AI source region. 
     
     
         3 . The method of  claim 1 , wherein the AI service is provided without communication mediation of an AI proxy service agent. 
     
     
         4 . The method of  claim 1 , wherein the AI service is provided in real time or near real time from a locally cached AI service stack accessible by the edge node. 
     
     
         5 . The method of  claim 1 , wherein AI data is exchanged between the AI application and the edge node, and wherein the AI data comprises one or more of: texts, messages, hyperlinks, images, audio, dialogs, audio scenes, videos, documents, AI model parameters, or machine learning (ML) model parameters. 
     
     
         6 . The method of  claim 1 , wherein the AI service is based on one or more of: linear regression analyses, decision trees, non-linear regression analyses, Gaussian process regression analyses, data clustering, face detection, face recognition, neural networks, trend prediction, natural language processing, deep learning, auto learning one-stop AI algorithm stores, healthcare related AI models, or autonomous driving related AI models. 
     
     
         7 . The method of  claim 1 , wherein the AI application is in one or more non-AI-source computer networks in the non-AI-source region that are separated from one or more AI source computer networks located in the AI source region. 
     
     
         8 . A non-transitory computer readable medium that stores computer instructions which, when executed by one or more computing processors, cause the one or more computing processors to perform:
 receiving, from an artificial intelligence (AI) application located in a non-AI-source region by an edge node in one or more edge nodes of an AI delivery edge network, an AI service request directed to an AI cloud engine located in an AI source region;   sending, by the edge node to an AI service proxy agent of the AI cloud engine, the AI service request, wherein the AI service request is forwarded by the AI service proxy agent to the AI cloud engine;   receiving, by the edge node from the AI service proxy agent, an AI service response to the AI service request, wherein the AI service response is generated by the AI cloud engine and forwarded by the AI service proxy agent;   causing the AI application to invoke an AI service to train a prediction model locally at the edge node.   
     
     
         9 . The medium of  claim 8 , wherein the AI service is provided in real time or near real time by one or more AI cloud engines in the AI source region. 
     
     
         10 . The medium of  claim 8 , wherein the AI service is provided without communication mediation of an AI proxy service agent. 
     
     
         11 . The medium of  claim 8 , wherein the AI service is provided in real time or near real time from a locally cached AI service stack accessible by the edge node. 
     
     
         12 . The medium of  claim 8 , wherein AI data is exchanged between the AI application and the edge node, and wherein the AI data comprises one or more of: texts, messages, hyperlinks, images, audio, dialogs, audio scenes, videos, documents, AI model parameters, or machine learning (ML) model parameters. 
     
     
         13 . The medium of  claim 8 , wherein the AI service is based on one or more of: linear regression analyses, decision trees, non-linear regression analyses, Gaussian process regression analyses, data clustering, face detection, face recognition, neural networks, trend prediction, natural language processing, deep learning, auto learning one-stop AI algorithm stores, healthcare related AI models, or autonomous driving related AI models. 
     
     
         14 . The medium of  claim 8 , wherein the AI application is in one or more non-AI-source computer networks in the non-AI-source region that are separated from one or more AI source computer networks located in the AI source region. 
     
     
         15 . An apparatus, comprising:
 one or more computing processors;   a non-transitory computer readable medium that stores computer instructions which, when executed by the one or more computing processors, cause the one or more computing processors to perform: 
 receiving, from an artificial intelligence (AI) application located in a non-AI-source region by an edge node in one or more edge nodes of an AI delivery edge network, an AI service request directed to an AI cloud engine located in an AI source region; 
 sending, by the edge node to an AI service proxy agent of the AI cloud engine, the AI service request, wherein the AI service request is forwarded by the AI service proxy agent to the AI cloud engine; 
 receiving, by the edge node from the AI service proxy agent, an AI service response to the AI service request, wherein the AI service response is generated by the AI cloud engine and forwarded by the AI service proxy agent; 
 causing the AI application to invoke an AI service to train a prediction model locally at the edge node. 
   
     
     
         16 . The apparatus of  claim 15 , wherein the AI service is provided in real time or near real time by one or more AI cloud engines in the AI source region. 
     
     
         17 . The apparatus of  claim 15 , wherein the AI service is provided without communication mediation of an AI proxy service agent. 
     
     
         18 . The apparatus of  claim 15 , wherein the AI service is provided in real time or near real time from a locally cached AI service stack accessible by the edge node. 
     
     
         19 . The apparatus of  claim 15 , wherein AI data is exchanged between the AI application and the edge node, and wherein the AI data comprises one or more of: texts, messages, hyperlinks, images, audio, dialogs, audio scenes, videos, documents, AI model parameters, or machine learning (ML) model parameters. 
     
     
         20 . The apparatus of  claim 15 , wherein the AI service is based on one or more of: linear regression analyses, decision trees, non-linear regression analyses, Gaussian process c regression analyses, data clustering, face detection, face recognition, neural networks, trend prediction, natural language processing, deep learning, auto learning one-stop AI algorithm stores, healthcare related AI models, or autonomous driving related AI models. 
     
     
         21 . The apparatus of  claim 15 , wherein the AI application is in one or more non-AI-source computer networks in the non-AI-source region that are separated from one or more AI source computer networks located in the AI source region.