# US 12,217,086 B2 · Chain schedule management for machine learning model-based processing in computing environment > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,217,086 B2 - Document ID: US12217086B2 - Status: Granted - Kind: B2 - Application number: 17/681,309 - Filed: 2022-02-25 - Published: 2023-08-31 - Granted: 2025-02-04 - Earliest claimed: 2022-02-25 - Assignees: Dell Products L.P. - Applicants: Dell Products L.P. - Inventors: Victor Fong - Web page: https://patentgrep.com/patent/US12217086B2 ## CPC - G06F 9/4881: {Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues} - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - G06N 3/063: using electronic means - G06N 20/00: Machine learning - G06N 3/0495: Quantised networks; Sparse networks; Compressed networks - G06N 3/082: modifying the architecture, e.g. adding, deleting or silencing nodes or connections - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Techniques are disclosed for chain schedule management for machine learning model-based processing in a computing environment. For example, a method receives a machine learning model-based request and determines a scheduling decision for execution of the machine learning model-based request. Determination of the scheduling decision comprises utilizing a set of one or more scheduling algorithms and comparing results of at least a portion of the set of one or more scheduling algorithms to identify execution environments of a computing environment in which the machine learning model-based request is to be executed. The identified execution environments may then be managed to execute the machine learning model-based request. ## Claims ### Claim 1 A method, comprising: receiving a machine learning model-based request; and determining a scheduling decision for execution of the machine learning model-based request, wherein determining the scheduling decision further comprises utilizing a set of one or more scheduling algorithms and comparing results of at least a portion of the set of one or more scheduling algorithms to identify execution environments of a computing environment in which the machine learning model-based request is to be executed; wherein the receiving and determining steps are performed by at least one processor and at least one memory storing executable computer program instructions. ### Claim 2 (depends on 1) The method of claim 1, wherein the machine learning model-based request comprises an inference serving request. ### Claim 3 (depends on 1) The method of claim 1, wherein determining the scheduling decision further comprises determining that the machine learning model-based request warrants multiple functions to execute in a chain of multiple clusters of the computing environment. ### Claim 4 (depends on 3) The method of claim 3, wherein determining the scheduling decision further comprises identifying the chain of multiple clusters of the computing environment based on scoring values that represent one or more end-to-end metrics associated with execution of the multiple functions across the chain of multiple clusters. ### Claim 5 (depends on 4) The method of claim 4, wherein the scoring values represent one or more of an accuracy metric, a cost metric, and a performance metric. ### Claim 6 (depends on 4) The method of claim 4, wherein comparing results of at least a portion of the set of one or more scheduling algorithms to identify execution environments of a computing environment in which the machine learning model-based request is to be executed further comprises comparing the scoring values associated with the results of at least a portion of the set of one or more scheduling algorithms. ### Claim 7 (depends on 1) The method of claim 1, wherein at least one of the set of one or more scheduling algorithms is configured to perform a heuristic search. ### Claim 8 (depends on 7) The method of claim 7, wherein the heuristic search is configured to consider a given execution environment and one or more variations of the given execution environment when identifying execution environments of a computing environment in which the machine learning model-based request is to be executed. ### Claim 9 (depends on 8) The method of claim 8, wherein the heuristic search is configured to generate respective scoring values for the given execution environment and the one or more variations of the given execution environment. ### Claim 10 (depends on 9) The method of claim 9, wherein the respective scoring values are generated in at least one accelerator. ### Claim 11 (depends on 1) The method of claim 1, further comprising managing the identified execution environments to execute the machine learning model-based request. ### Claim 12 (depends on 1) The method of claim 1, wherein the at least one processor and the at least one memory comprise a node in a virtualized workload orchestration framework. ### Claim 13 (depends on 12) The method of claim 12, wherein the node is part of an edge computing platform. ### Claim 14 An apparatus, comprising: at least one processing platform comprising at least one processor coupled to at least one memory, the at least one memory storing computer program instructions wherein, the at least one processing platform, when executing the computer program instructions, is configured to: receive a machine learning model-based request; and determine a scheduling decision for execution of the machine learning model-based request, wherein determining the scheduling decision further comprises utilizing a set of one or more scheduling algorithms and comparing results of at least a portion of the set of one or more scheduling algorithms to identify execution environments of a computing environment in which the machine learning model-based request is to be executed. ### Claim 15 (depends on 14) The apparatus of claim 14, wherein the machine learning model-based request comprises an inference serving request. ### Claim 16 (depends on 14) The apparatus of claim 14, wherein determining the scheduling decision further comprises determining that the machine learning model-based request warrants multiple functions to execute in a chain of multiple clusters of the computing environment. ### Claim 17 (depends on 16) The apparatus of claim 16, wherein determining the scheduling decision further comprises identifying the chain of multiple clusters of the computing environment based on scoring values that represent one or more end-to-end metrics associated with execution of the multiple functions across the chain of multiple clusters. ### Claim 18 A computer program product stored on a non-transitory computer-readable medium and comprising machine executable instructions, the machine executable instructions, when executed, causing a processing device to perform steps of: receiving a machine learning model-based request; and determining a scheduling decision for execution of the machine learning model-based request, wherein determining the scheduling decision further comprises utilizing a set of one or more scheduling algorithms and comparing results of at least a portion of the set of one or more scheduling algorithms to identify execution environments of a computing environment in which the machine learning model-based request is to be executed. ### Claim 19 (depends on 18) The computer program product of claim 18, wherein the machine learning model-based request comprises an inference serving request. ### Claim 20 (depends on 18) The computer program product of claim 18, wherein determining the scheduling decision further comprises: determining that the machine learning model-based request warrants multiple functions to execute in a chain of multiple clusters of the computing environment; and identifying the chain of multiple clusters of the computing environment based on scoring values that represent one or more end-to-end metrics associated with execution of the multiple functions across the chain of multiple clusters.