# US 2019/0356555 A1 · Machine learning microservice architecture design tools and methods > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2019/0356555 A1 - Document ID: US20190356555A1 - Status: Published application - Kind: A1 - Application number: 15/982,243 - Filed: 2018-05-17 - Published: 2019-11-21 - Earliest claimed: 2018-05-17 - Assignees: Microsoft Technology Licensing, LLC - Applicants: Microsoft Technology Licensing, LLC - Inventors: Haishi BAI - Web page: https://patentgrep.com/patent/US20190356555A1 ## CPC - H04L 41/145: {involving simulating, designing, planning or modelling of a network} - H04L 41/142: using statistical or mathematical methods - H04L 41/16: using machine learning or artificial intelligence - H04L 41/5009: Determining service level performance parameters or violations of service level contracts, e.g. violations of agreed response time or mean time between failures [MTBF] - H04L 41/5048: {Automatic or semi-automatic definitions, e.g. definition templates} - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) ## Abstract According to some embodiments, a machine learning architecture design platform may access a microservice architecture design data store that contains existing microservice architecture designs, and a graphic abstraction computing component may automatically create existing graph models of the existing designs. A pattern recognition computing component may then execute a machine learning algorithm to access the existing graph models and automatically detect existing design patterns. A designer interface computing component may interactively and iteratively exchange information with a designer, including receipt of at least one design requirement from the designer. Based on the at least one received design requirement and the automatically detected existing design patterns, a dynamic recommendation computing component may automatically construct a recommended microservice architecture for the cloud computing environment. In some embodiments, a quality assessment computing component may evaluate a potential microservice architecture design. ## Claims ### Claim 1 A system associated with microservice architecture design for a cloud computing environment, comprising: at least one microservice architecture design data store containing existing microservice architecture designs; and a machine learning architecture design platform, coupled to the at least one microservice architecture design data store, including: a graphic abstraction computing component to access the existing microservice architecture designs and automatically create existing graph models of the existing designs, a pattern recognition computing component to execute a machine learning algorithm that accesses the existing graph models and automatically detects existing design patterns, a designer interface computing component to interactively and iteratively exchange information with a designer, and a dynamic recommendation computing component to: receive, from the designer interface computing component, at least one design requirement, and based on the at least one received design requirement and the automatically detected existing design patterns, automatically construct a recommended microservice architecture for the cloud computing environment. ### Claim 2 (depends on 1) The system of claim 1, wherein the at least one microservice architecture design data store is associated with at least one of: (i) a public design data store, (ii) a data store associated with the designer, (iii) a data store associated with other designers, (iv) a data store associated with the cloud computing environment, (v) a data store associated with other cloud computing environments, (vi) an architecture requirements data store, (vii) a platform-specific knowledge data store, (viii) a library of design templates, and (ix) textual design descriptions. ### Claim 3 (depends on 1) The system of claim 1, wherein the existing graph models comprise enterprise topography graphs. ### Claim 4 (depends on 3) The system of claim 3, wherein the graphic abstraction computing component executes at least one of: (i) a process to automatically acquire knowledge by crawling a plurality of data stores, (ii) a process utilizing a horizontal-scale property, (iii) a process utilizing a vertical-scale property, (iv) a process utilizing a state-feature property, (v) a process utilizing a proxy property, (vi) a process utilizing a communication pattern property, (vii) a graph generalization process, (viii) a graph specialization process, (ix) a graph substitution process, and (x) a graph acquisition process. ### Claim 5 (depends on 1) The system of claim 1, wherein the pattern recognition computing component executes at least one of: (i) a process to identify efficient patterns, and (ii) a process to identify inefficient anti-patterns. ### Claim 6 (depends on 1) The system of claim 1, wherein the designer interface computing component exchanges information with the designer via at least one of: (i) a natural language interface, (ii) a virtual expert, (iii) a chat bot, (iv) voice recognition, and (v) a video stream. ### Claim 7 (depends on 1) The system of claim 1, wherein the dynamic recommendation computing component automatically constructs a series of recommended microservice design patterns for the designer. ### Claim 8 (depends on 1) The system of claim 1, wherein the machine learning architecture design platform further includes: a quality assessment computing component to evaluate a potential microservice architecture design via at least one of: (i) a Quality of Service (“QoS”) merit analysis, (ii) a modulization analysis, (iii) a robustness analysis, (iv) a scalability analysis, and (v) a single-point-of-failure analysis. ### Claim 9 (depends on 8) The system of claim 8, wherein the quality assessment computing component further evaluate a currently deployed microservice architecture design. ### Claim 10 (depends on 1) The system of claim 1, wherein the machine learning architecture design platform utilizes at least one of: (i) decision tree learning, (ii) association rule learning, (iii) an artificial neural network, (iv) deep learning, (v) inductive logic programming, (vi) a Support Vector Machine (“SVM”), (vii) clustering, (viii) a Bayesian network, (ix) reinforcement learning, (x) representation learning, (xi) similarity and metric learning, (xii) sparse dictionary learning, and (xiii) a genetic algorithm. ### Claim 11 (depends on 1) The system of claim 1, wherein the machine learning architecture designer platform is to transmit a deployable package associated with a microservice architecture design to the designer. ### Claim 12 A computer-implemented method associated with microservice architecture design for a cloud computing environment, comprising: accessing, by a machine learning architecture design platform, at least one microservice architecture design data store containing existing microservice architecture designs; automatically creating, by a graphic abstraction computing component, existing graph models of the existing designs; executing, by a pattern recognition computing component, a machine learning algorithm that accesses the existing graph models and automatically detects existing design patterns; interactively and iteratively exchanging information, by the designer interface computing component, with a designer including a receipt of at least one design requirement from the designer; and based on the at least one received design requirement and the automatically detected existing design patterns, automatically constructing, by a dynamic recommendation computing component, a recommended microservice architecture for the cloud computing environment. ### Claim 13 (depends on 12) The method of claim 12, wherein the at least one microservice architecture design data store is associated with at least one of: (i) a public design data store, (ii) a data store associated with the designer, (iii) a data store associated with other designers, (iv) a data store associated with the cloud computing environment, (v) a data store associated with other cloud computing environments, (vi) an architecture requirements data store, (vii) a platform-specific knowledge data store, (viii) a library of design templates, and (ix) textual design descriptions. ### Claim 14 (depends on 12) The method of claim 12, wherein the existing graph models comprise enterprise topography graphs. ### Claim 15 (depends on 14) The method of claim 14, wherein the graphic abstraction computing component executes at least one of: (i) a process to automatically acquire knowledge by crawling a plurality of data stores, (ii) a process utilizing a horizontal-scale property, (iii) a process utilizing a vertical-scale property, (iv) a process utilizing a state-feature property, (v) a process utilizing a proxy property, (vi) a process utilizing a communication pattern property, (vii) a graph generalization process, (viii) a graph specialization process, (ix) a graph substitution process, and (x) a graph acquisition process. ### Claim 16 (depends on 12) The method of claim 12, wherein the pattern recognition computing component executes at least one of: (i) a process to identify efficient patterns, and (ii) a process to identify inefficient anti-patterns. ### Claim 17 A non-transient, computer-readable medium storing instructions to be executed by a processor to perform a method associated with microservice architecture design for a cloud computing environment, the method comprising: accessing, by a machine learning architecture design platform, at least one microservice architecture design data store containing existing microservice architecture designs; automatically creating, by a graphic abstraction computing component, existing graph models of the existing designs; executing, by a pattern recognition computing component, a machine learning algorithm that accesses the existing graph models and automatically detects existing design patterns; interactively and iteratively exchanging information, by the designer interface computing component, with a designer including a receipt of at least one design requirement from the designer; and based on the at least one received design requirement and the automatically detected existing design patterns, automatically constructing, by a dynamic recommendation computing component, a recommended microservice architecture for the cloud computing environment. ### Claim 18 (depends on 17) The medium of claim 17, wherein the designer interface computing component exchanges information with the designer via at least one of: (i) a natural language interface, (ii) a virtual expert, (iii) a chat bot, (iv) voice recognition, and (v) a video stream. ### Claim 19 (depends on 17) The medium of claim 17, wherein the dynamic recommendation computing component automatically constructs a series of recommended microservice design patterns for the designer. ### Claim 20 (depends on 17) The medium of claim 17, further comprising: evaluating, by a quality assessment computing component, a potential microservice architecture design via at least one of: (i) a Quality of Service (“QoS”) merit analysis, (ii) a modulization analysis, (iii) a robustness analysis, (iv) a scalability analysis, and (v) a single-point-of-failure analysis.