EP3743813A2 - Platform for hierarchy cooperative computing - Google Patents
Platform for hierarchy cooperative computingInfo
- Publication number
- EP3743813A2 EP3743813A2 EP19743576.1A EP19743576A EP3743813A2 EP 3743813 A2 EP3743813 A2 EP 3743813A2 EP 19743576 A EP19743576 A EP 19743576A EP 3743813 A2 EP3743813 A2 EP 3743813A2
- Authority
- EP
- European Patent Office
- Prior art keywords
- processor
- vector
- memory
- request
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/5033—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering data affinity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5066—Algorithms for mapping a plurality of inter-dependent sub-tasks onto a plurality of physical CPUs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/101—Collaborative creation, e.g. joint development of products or services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/5015—Service provider selection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/502—Proximity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/54—Indexing scheme relating to G06F9/54
- G06F2209/544—Remote
Definitions
- the disclosure relates to tire field of hierarchical distributed computing systems.
- What is needed is a system that may take a plurality of specified constraints and factors, and automatically determine an optimal plan for executing a user request for computing. Such a system should also be able to procure any additional resources, as well as identify bottlenecks in the system and provide a solution to the bottleneck.
- a system may be configured to operate in a decentralized manner, with a centralized control point.
- die control point may evaluate connections, data localities, processing localities, and the like to determine a best endpoint and plan in executing a user request for cooperative computing with regards to factors such as data and processing localities, any regulations in the aforementioned localities, costs, system available, and the like.
- a system for hierarchical cooperative computing comprising a vector definition service comprising a memory, a processor, and a plurality of programming instructions stored in the memory thereof and operable on the processor thereof, wherein the programmable instructions, when operating on the processor, cause the processor to receive a user-submitted request comprising at least a cooperative computing request, and compile the request into a vector using at least a model definition language; a rules engine comprising a memory, a processor, and a plurality of programming instructions stored in the memory thereof and operable on the processor thereof, wherein the programmable instructions, when operating on the processor, cause the processor to retrieve the vector from the vector definition service, and evaluate the vector for appropriateness based at least on predefined rules, and data and processing localities; a parametric evaluator comprising a memory, a processor, and a plurality of programming instructions stored in the memor ' thereof and operable on the processor thereof, wherein the programmable instructions, when operating on the processor
- a method for hierarchical cooperative computing comprising the steps of: (a) receiving a user-submited request comprising at least a cooperative computing request using a vector definition service; (b) compiling the request into a vector using at least a model definition language using the vector definition service; (c) retrieving the vector from the vector definition service using a rules engine; (d) evaluating the vector for appropriateness based at least on predefined rules, and data and processing localities using the rules engine; (e) parameterizing the vector based at least on intended purpose and predefined constraints of the user-submitted request using a parametric evaluator; (1) generating at least a run from the parameterized vector using the parametric evaluator; (g) retrieving the run from the parametric evaluator using an optimizer; and (h) determining an optimal locality for executing the user-submited request based at least on the stains of connections to the optimal locality and availability of computational resources of the optimal locality
- the system further comprises a data migration service comprising a memory, a processor, and a plurality of programming instructions stored in the memory thereof and operable on the processor thereof, wherein the programmable instructions, when operating on the processor, cause the processor to initiate migration of data associated with the usersubmitted request to a different locality for processing.
- the system further comprises a resource modulation service comprising a memory, a processor, and a plurality of programming instructions stored in the memory thereof and operable on the processor thereof, wherein the programmable instructions, when operating on the processor, cause the processor to automatically acquire additional resources in order to execute the user-submitted request from an external sendee provider.
- the optimizer uses an external simulation service to operate an instanced copy of a compute environment in order to identify bottlenecks in the system.
- the rules engine is further configured to conduct a feasibility analysis on an incoming vector. According to another embodiment of the invention, the rules engine denies a vector and submits a request for additional information.
- FIG. 1 is a block diagram of an exemplar - system architecture for a system for decentralized trading according to various embodiments of the invention.
- FIG. 2 is an illustration of an exemplary' topography of a system employing a plurality' of decentralized trading systems according to various embodiments of the invention.
- FIG. 3 is a block diagram of an exemplary system architecture 300 of a platform for hierarchical cooperative computing according to various embodiments of the invention.
- Fig. 4 is a block diagram of an exemplary optimizer used in a platform for hierarchical cooperative computing according to various embodiments of the invention
- FIG. 5 is an illustration of an exemplary topography of a system employing a platform for hierarchical cooperative computing according to various embodiments of the invention.
- Fig. 6 is a sequence flow diagram illustrating an exemplary sequence 600 for processing a hierarchical cooperative computing-related request according to various embodiments of the invention.
- Fig. 7 is a flow diagram illustrating an exemplary method 700 for verification of a vector using a rules engine according to various embodiments of the invention.
- Fig. 8 is a flow' diagram illustrating an exemplary method 800 for parameterizing vectors according to various embodiments of the invention.
- Fig. 9 is a flow diagram illustrating an exemplary ' method 900 for generating an optimal plan according to various embodiments of the invention.
- FIG. 10 is a block diagram illustrating an exemplary hardware architecture of a computing device used in various embodiments of the invention.
- FIG. 11 is a block diagram illustrating an exemplary logical architecture for a client device, according to various embodiments of the invention.
- Fig. 12 is a block diagram illustrating an exemplary architectural arrangement of clients, servers, and external services, according to various embodiments of the invention.
- FIG. 13 is another block diagram illustrating an exemplary hardware architecture of a computing device used in various embodiments of the invention.
- Devices that are in communication with each other need not be in continuous communication with each other, unless expressly specified otherwise.
- devices that are in communication with each other may communicate directly or indirectly through one or more communication means or intermediaries, logical or physical.
- steps may be performed simultaneously despite being described or implied as occurring non- simultaneously (e.g., because one step is described after the other step).
- the illustration of a process by its depiction in a drawing does not imply that the illustrated process is exclusive of other vaiiations and modifications thereto, does not imply that the illustrated process or any of its steps are necessary to one or more of the aspects, and does not imply that the illustrated process is preferred.
- steps are generally described once per aspect, but this does not mean they must occur once, or that they may only occur once each time a process, method, or algorithm is carried out or executed. Some steps may be omitted in some aspects or some occurrences, or some steps may be executed more than once in a given aspect or occurrence.
- a“vector” may be defined as a container for compute instructions, and may comprise instructions and descriptions for data locality, process locality, priority', type, search, approach, and the like. Vectors may also be used in a search process, and for declaration of constraints regarding the conditions under which specific actions may ' be taken, limitations on inputs, limitations on outputs, limitations on downstream uses to be attached to outputs, and the like.
- a“run” may be considered to be a vector which has been evaluated and processed by a parameterized model execution engine according to various factors contributing to overall utility' and objective function optimization.
- Fig. 1 is a block diagram of an exemplary system architecture for a system 100 for decentralized trading according to various embodiments of the invention.
- System 100 may comprise a parametric evaluator 110, an optimizer 120, a rules engine 130, a model definition language service 140, and a data store 160.
- System 100 may continually' monitor and track current status of connections and system states.
- system 100 may be in logical form, or may be an external service. Other embodiments of system 100 may ' have less components t an what is shown in Fig. 1, while other embodiments may have additional components.
- Parametric evaluator 110 may be configured to assess model performance and bias, and may ' comprise a model execution engine 111. Parametric evaluator 110 may be configured to analyze a plurality of data flow localities and priorities, and compile a list of results according to predefined factors, such as overall associated costs, volatility, profitability, effectiveness of global system optimizations, and the like.
- Model execution engine 111 may be configured to analyze and parameterize a plurality of vectors, and their outcomes when given a plurality of factors relating to a trade, such as overall cost, effectiveness in global system optimization, profitability, volatility, and the like.
- the parameterization of a vector description may result in a“run”, which may be sent to optimizer 120 for further processing and analysis.
- Optimizer 120 may be configured to analyze“runs” that received from parametric evaluator 110, and generate recommendations regarding appropriateness of one or more data flow localities, such as regulatory issues or legality, or utility for one or more sets of exogenous factors or system states. For example, optimizer 120 may recommend a combination of data flow and storage localities based on current global system states to determine a course of action for one or more financial trades resulting in favorable outcomes by choosing whether to migrate data, migrate processes, or call into spot markets to control data and processing locality in order to minimalize latency associated with execution trades across geographically distributed market centers; or analyzing hypothetical system states, such as using a simulation engine to operate an identical instance in simulation to identify current and future bottlenecks.
- optimizer 120 may be configured to define a set of rales pertaining to the appropriateness of data locality and process locality with regards to a system condition for a given purpose, for instance, for determining profitable trades, which may be expressed in a declarative formalism accessible to rules engine 130.
- optimizer 120 may develop an understanding of optimal models, groups of models, or rules defining model appropriateness or performance over time; and may change or restrict ordering of model packages or rules combinations based on the developed understanding.
- Rules engine 130 may be configured to enable management of system rules, and also to evaluate specific elements of a given instance of one or more models when given any definition for the current or future state of said models. For example, rules engine 130 may verify that a request is allowed or appropriate based on the intended use, for example, feasibility or legality of an intended trade; whether a defined confidence requirement or other conditions are met; and evaluate configuration-specific terms and requirements as specified in user-defined operating constraints or guidelines. Rules engine 130 may evaluate rales by executing a forward chaining deduction of data amassed from a set of antecedents derived from model definition language service 140 for a particular application or purpose. Rules engine 130 supports layered“batteries” of modular tests, where functional decomposition of rules supports higher degrees of user productivity and rules re-use.
- Model definition language sendee 140 may be configured to allow user management of models, and defining of vectors using a declarative specification language (DSL).
- DSL declarative specification language
- the use of a DSL for vectorizing the compute environment and data flow descriptions may enable linking of search processes to the rules engine 130, parametric evaluator 110, and feedback loop processes during ongoing operational-use based on the ability to encode appropriateness when combined with rules engine 130, serving as a basis for deep and reinforcement learning to support ongoing improvement to functions of optimizer 120.
- Model definition language service 140 may also enable a user or an autonomous trading system to initiate evaluation of specific pipelines, activities, overall system health, and the like of a specific instance of system 100.
- FIG. 2 is an illustration of an exemplary topography 200 of a system employing a plurality of decentralized trading systems lOOa-d according to various embodiments of the invention.
- Topography 200 is an example of a layout of various components within a geographical area, for example spanning a continent or even on a global scale, and illustrates a plurality of systems lOOa-d connecting with a plurality of user global market centers 210a-e, such as a stock market or foreign exchange markets, through a wide area network connection; and a plurality of user devices 230a-t2, which may be a single user or group of users accessing trading platform 100a through, for example, a web application, mobile device, spatial operating system, AR or VR system, and the like.
- Systems lOOa-d may be flexible in their placement and locale, which may include, for example, as a standalone system 100a; running in a virtual machine of a cloud service provider, such as AMAZON AW5 220, IQOd; residing inside a global market center 210b, 100c; or even submerged in a body of water 240, 100b, for example inside a mobile submersible data center.
- Locations for systems lQQa-d may be strategically chosen, so that they may be useful in operating as an intermediate connection to a trading market.
- Topography 200 utilizes a centralized control point in system 100a for users to communicate with decentralized deployment of a plurality of instances of system lOOb-d. Any particular instance may be chosen by an optimizer of system 100a as the locality for data processing and storage; or system in which to execute a trade based on metrics such as system availability, latency to reach a target global market for trading a certain asset, and the like,
- Fig. 2 is used for demonstration purposes, and does not represent a limitation of the present invention. For example, there may be more than one control point, more decentralized trading system endpoints, more global markets, and She like.
- FIG. 3 is a block diagram of an exemplary system architecture 300 of a platform for hierarchical cooperative computing according to various embodiments of the invention.
- System 300 may comprise an optimizer 320, a rules engine 330, a data migration sendee 335, a resource modulation service 340, a parametric evaluator 350, a vector definition service 355, and a data store 310 for storing data such as rules, vectors, runs, user-defined constraints, and the like.
- the components of system 300 may be implemented in logical form, or may be an external service. Other embodiments of system 300 may have less components than what is shown in Fig. 3, while other embodiments may have additional components.
- Optimizer 320 may be configured to analyze“runs” received from parametric evaluator 350, and generate recommendations regarding appropriateness of one or more data flow localities, such as regulatory' issues or legality ' , or utility for one or more sets of exogenous factors or system states. For example, optimizer 320 may' recommend a combination of data flow ' and storage localities based on current global system states to determine a course of action for one or more financial trades resulting in favorable outcomes by' choosing whether to migrate data, migrate processes, or call into spot markets to control data and processing locality in order to minimalize latency associated with connection latency; or analyzing hypothetical system states, such as using simulation services to operate an identical instance in simulation to identify ' current and future bottlenecks.
- optimizer 320 may develop an understanding of optimal models, groups of models, or rales defining model appropriateness or performance over time. Optimizer 320 may then use this developed knowledge to change or restrict ordering of model packages or rules combinations based on the developed
- Optimizer 320 may comprise an asset analyzer 321, a cost analyzer 322, an appropriateness engine 323, a model selection engine 324, a dimensionality reduction engine 325.
- Asset analyzer 321 may be configured to evaluate available assets and chooses an optimal set of assets based on cost, speed, availability, and the like based on requirements specified by a user submitting a request. Asset manager may also keep track of statuses of all deployed instances of system 300.
- Cost analyzer 322 may be configured to analyze cost associated with using available resources, or acquiring external resources (such as starting a new instance on a cloud computing service such as AMAZON AWS).
- Appropriateness engine 323 may be configured to allow defining a set of rules pertaining to the appropriateness of data locality and process locality with regards to a system condition for a given purpose, for instance, to prohibit migrating and processing data in a region with conflicting data import and export laws.
- Model selection engine 324 ma be configured to choose best-performing models for an particular intended purpose, and also adjust orders of model packages based on developed understanding from processing data over time.
- Dimensionality reduction engine 325 may be configured to utilize a plurality of heuristic search algorithms to reduce dimensionality for optimization purposes. Search algorithms ma include, for instance, grid, brute force, Monte Carlo tree search, simulated annealing, genetic algorithms, and die like.
- Rules engine 330 may be configured to enable management of system rules, and also to evaluate specific elements of a given instance of one or more vectors when given any definition for the current or future state of said vectors. For example, rules engine 330 may verify that a request is allowed or appropriate based on the intended use, for instance, feasibility or legality of an intended computing request from a user; whether a defined confidence requirement or other conditions are met; and evaluate configuration-specific terms and requirements as specified in user-defined operating constraints or guidelines. Rules engine 330 may evaluate vectors by executing a forward chaining deduction of data amassed from a set of antecedents derived from the model definition language for a particular application or purpose.
- Rules engine 330 supports layered“batteries” of modular tests, where functional decomposition of rules supports higher degrees of user productivity and rules re-use.
- Data migration sendee 335 may be configured to trigger migration of data, connect to external services and facilitate the migration of data for computing, such as AMAZON
- SNOWBALL and SNOWMOBILE sendees when required by other components of system 300.
- Resource modulation sendee 340 may be configured to dynamically acquire additional resources when required by other components of system 300. For example, when required by optimizer 320, additional instances may be started a cloud computing platform such as
- AMAZON AWS or additional cloud storage space may be acquired.
- Parametric evaluator 350 may be configured to assess model performance and bias, and may comprise a model execution engine 351.
- Parametric evaluator 350 may be configured to analyze a plurality of data flow localities and priorities, and compile a list of results according to predefined requirements, such as overall associated costs, effectiveness of global sy stem optimizations, and the like.
- Model execution engine 351 may be configured to analyze and parameterize a plurality of vectors, and their generated outcomes when given a plurality of factors relating to an intended purpose, such as overall cost, effectiveness in global system optimization, urgency , and the like. The parameterization of a vector description may result in a “run”, which may be sent to optimizer 320 for further processing and analysis.
- Vector definition sendee 355 may be configured to allo user management of models, and defining of vectors using a model definition language 356 (MDL), which may be a flexible declarative specification language designed to efficiently and uniformly express vectors and models used by system 300.
- MDL model definition language
- the use of MDL 356 for vectorizing the compute environment and data flow descriptions may enable linking of search processes to the rules engine 330, parametric evaluator 350, and feedback loop processes during ongoing operational-use based on the ability to encode appropriateness when combined with rules engine 330, serving as a basis for deep and reinforcement learning to support ongoing improvement to functions of optimizer 320.
- Model definition service 355 may also enable a user or an autonomous intelligent system to initiate evaluation of specific pipelines, activities, overall system health, and the like of a specific instance of system 300.
- FIG. 5 is an illustration of an exemplary topography 500 of a system employing a platform for hierarchical cooperative computing 300 according to various embodiments of the invention.
- Topography 500 is an example of a l ayout of various components within a geographical area, for example spanning a continent or even on a global scale, and illustrates a plurality of systems 3G0a-g used in various configurations, such as deployed on an aerial relay 510, a mobile relay 515, as a stand-alone service as in system 30Qd acting as an intermediary connection to cloud sendee provider 525b and data center 520b, inside of a data center 520a, inside of a cloud service provider 525a, and inside of a submersible data center 530 in a body of water 540 which may all be connected through a wide area network connection,
- a plurality of user devices 505&- ⁇ may provide a single user or group of users a means for accessing control point system 300a through, for example, a web application, mobile device,
- Topography 500 utilizes a centralized control point in system 300a which may receive a user request from any of devices 5Q5a-.n, and determine an endpoint amongst systems 3GGb-g for processing the request.
- Any particular deployment of system 300 may be chosen by an optimizer of system 300a as the locality for storage or processing locality based on specified factors, such as system availability, connection latency, and the like. For example, if a user requires mining of an enormous cache of gathered data, it may be imprudent to transfer data to a distant processing location.
- System 300 may instead utilize a mobile relay to adjust the processing locality to be closer to the processing point. Or system 300 may initiate migration of data to a capable facility for processing the data.
- tire layout and components depicted in Fig. 5 is used for demonstration purposes, and does not represent a limitation of tire present invention.
- Each endpoint may also be configured to be a control point for a plurality of localized endpoints.
- Fig. 6 is a sequence flow ' diagram illustrating an exemplary sequence 600 for processing a hierarchical cooperative computing-related request according to various embodiments of the invention.
- a user submits a user request using a user device.
- vector definition service 355 may compile the request into a vector specified using model definition language 356.
- the vector may be sent to rules engine 330 for processing.
- rules engine 330 processes the vector, which may include verification of the vector, whether system resources are available or acquirable, whether a request is appropriate or allowed based on intended use and regional regulations, whether the request is within user-dehned constraints, and the like.
- the vector may be sent to parametric evaluator 350.
- a process for handling a failed verification is detailed in Fig. 7.
- the vector may be evaluated and parameterized using parametric evaluator 350 and model execution engine 351, which generates a run.
- the run may be sent to optimizer 320 for determining an optimal plan for executing the original request.
- optimizer 320 may determine an optimal course of action to execute tire original user request This ma be based on such metrics as cost, latency, appropriateness, resource availability, and the like.
- a request may be submitted to acquire cloud computer resources for the purposes of processing a large amount of data located at a data center.
- the request is compiled into a vector and sent to the rules engine.
- the rules engine evaluates the request based on requirements, data and processing locality, and the like to determine whether the request is possible based on the provided information, any exisiting regulations, and the like.
- the vector is successfully verified.
- the vector is then sent to the parametric evaluator, where performance for models associated with the vector are measured.
- the vector may then be parameterized based on specified requirements and constraints, which generates one or more runs.
- the runs may then be sent to the optimizer for more in-depth planning, and execution.
- Fig. 7 is a flow diagram illustrating an exemplary method 700 for verification of a vector using a rules engine according to various embodiments of the invention.
- a user submits a hierarchical cooperative computing-related request.
- vector definition service 355 may compile the request into a vector specified using model definition language 356.
- the vector may be sent to rules engine 330 for verification of the vector which may include whether system resources are available or procurable, whether a sufficient amount of information is provided, whether a request is appropriate or allowed based on intended use, whether the request is within user-defined constraints, and the like.
- decision block 704 if die verification was unsuccessful, additional information or adjustments to die request may be requested from the user at step 705. After providing additional information, the user may resubmil the request. If the verification was successful at decision block 704, the vector may be sent to the parametric evaluator for further processing at step 706.
- Fig. 8 is a flow diagram illustrating an exemplary method 800 for parameterizing vectors according to various embodiments of the invention.
- parametric evaluator 350 receives a vector that has been previously verified by rules engine 330.
- the vector may be evaluated by parametric evaluator 350 for model performance and bias.
- a list of results may be generated based on user-provided factors, such as cost restrictions, intended use, data and processing locality requirements, and the like.
- parametric evaluator 350 may use the model execution engine to process and parameterize the vector and evaluation outcomes to generate one or more runs.
- the one or more runs may be sent to an optimizer for determining an optimal plan for execution.
- Fig. 9 is a flo diagram illustrating an exemplary method 900 for generating an optimal plan according to various embodiments of the invention.
- optimizer 320 may receive one or more runs previously generated by a parametric evaluator.
- the runs may be evaluated to determine one or more optimal plans. Evaluation may include, for instance, analysis of available assets and their current status, analysis of associated costs, dimensionality reduction, appropriateness of intended purpose, and the like.
- An optimal plan may include, for example, recommending a certain combination of data flows and storage localities, whether to migrate data localities, whether to migrate processing locality, and the like.
- the optimizer may use components of system 300, for example, the data migration service or resource modulation service, to migrate data, procure resources, and the like in order to execute the determined optimal plan.
- this step and the next step may be withheld until a user authorizes the plan.
- triggers may be predefined (such as allocating a certain budget or completion within a specified time period) and the plan may automatically execute only after one or more triggers are met.
- the plan may be executed.
- the techniques disclosed herein may be implemented on hardware or a combination of software and hardware. For example, they may be implemented in an operating system kernel, in a separate user process, in a library package bound into network applications, on a specially constructed machine, on an application-specific integrated circuit (ASIC), or on a network interface card.
- ASIC application-specific integrated circuit
- Software/hardware hybrid implementations of at least some of the aspects disclosed herein may be implemented on a programmable network-resident machine (which should be understood to include intermittently connected network-aware machines) selectively activated or reconfigured by a computer program stored in memory.
- Such network devices may have multiple network interfaces that may be configured or designed to utilize different types of network communication protocols.
- a general architecture for some of these machines may be described herein in order to illustrate one or more exemplary means by which a given unit of functionality may be implemented.
- At least some of the features or functionalities of the various aspects disclosed herein may be implemented on one or more general-purpose computers associated with one or more networks, such as for example an end- user computer system, a client computer, a network server or other server system, a mobile computing device (e.g., tablet computing device, mobile phone, smartphone, laptop, or other appropriate computing device), a consumer electronic device, a music player, or any other suitable electronic device, router, switch, or other suitable device, or any combination thereof.
- at least some of the features or functionalities of the various aspects disclosed herein may be implemented in one or more virtualized computing environments (e.g., network computing clouds, virtual machines hosted on one or more physical computing machines, or other appropriate virtual environments).
- computing device 10 may be, for example, any one of the computing machines listed in ihe previous paragraph, or indeed any other electronic device capable of executing software- or hardware-based instructions according to one or more programs stored in memory.
- Computing device 10 may be configured to communicate with a plurality of other computing devices, such as clients or servers, over communications networks such as a wide area network a metropolitan area network, a local area network, a wireless network, the Internet, or any oilier network, using known protocols for such communication, whether wireless or wired.
- computing device 10 includes one or more central processing units (CPU) 12, one or more interfaces 15, and one or more busses 14 (such as a peripheral component interconnect (PCI) bus).
- CPU central processing units
- interfaces such as a peripheral component interconnect (PCI) bus
- CPU 12 may be responsible for implementing specific functions associated with the functions of a specifically configured computing device or machine.
- a computing device 10 may be configured or designed to function as a server system utilizing CPU 12, local memory- 11 and/or remote memory '' 16, and interface(s) 15.
- CPU 12 may be caused to perform one or more of the different types of functions and/or operations under the control of software modules or components, which for example, may include an operating system and any appropriate applications software, drivers, and die like.
- CPU 12 may include one or more processors 13 such as, for example, a processor from one of the Intel, ARM, Qualcomm, and AMD families of microprocessors.
- processors 13 may include specially designed hardware such as application-specific integrated circuits (ASICs), electrically erasable programmable read-only memories (EEPROMs), field- programmable gate arrays (FPGAs), anil so forth, for controlling operations of computing device 10.
- ASICs application-specific integrated circuits
- EEPROMs electrically erasable programmable read-only memories
- FPGAs field- programmable gate arrays
- a local memory 11 such as non-volatile random access memory (RAM) and/or read-only memory (ROM), including for example one or more levels of cached memory
- RAM non-volatile random access memory
- ROM read-only memory
- Memory' 11 may be used for a variety of purposes such as, for example, caching and/or storing data, programming instructions, and the like. It should be further appreciated that CPU 12 may be one of a variety of system-on-a-chip (SOC) type hardware that may include additional hardware such as memory or graphics processing chips, such as a QUALCOMM SNAPDRAGONTM or SAMSUNG EXYNOSTM CPU as are becoming increasingly' common in the art, such as for use in mobile devices or integrated devices.
- SOC system-on-a-chip
- processor is not limited merely to those integrated circuits referred to in the art as a processor, a mobile processor, or a microprocessor, but broadly refers to a microcontroller, a microcomputer, a programmable logic controller, an application-specific integrated circuit, and any other programmable circuit.
- interfaces 15 are provided as network interface cards (NICs), Generally ' , NICs control the sending and receiving of data packets over a computer network; other types of interfaces 15 may for example support other peripherals used with computing device 10.
- NICs network interface cards
- NICs control the sending and receiving of data packets over a computer network
- other types of interfaces 15 may for example support other peripherals used with computing device 10.
- the interfaces that may be provided are Ethernet interfaces, frame relay interfaces, cable interfaces, DSL interfaces, token ring interfaces, graphics interfaces, and the like.
- interfaces may be provided such as, for example, universal serial bus (USB), Serial, Ethernet, FIREWIRETM, THUNDERBOLTTM, PCI, parallel, radio frequency (RF), BLUETOOTHTM, near-field communications (e.g., using near-field magnetics), 802.11 (WiFi), frame relay, TCP/IP, ISDN, fast Ethernet interfaces, Gigabit Ethernet interfaces, Serial ATA (SATA) or external SATA (ESATA) interfaces, high-definition multimedia interface (IIDMI), digital visual interface (DVI), analog or digital audio interfaces, asynchronous transfer mode (ATM) interfaces, high-speed serial interface (HSSI) interfaces, Point of Sale (POS) interfaces, fiber data distributed interfaces (FDDIs), and the like.
- USB universal serial bus
- RF radio frequency
- BLUETOOTHTM near-field communications
- near-field communications e.g., using near-field magnetics
- WiFi WiFi
- frame relay e.g., 8
- Such interfaces 15 may include physical ports appropriate for communication with appropriate media. In some cases, they may also include an independent processor (such as a dedicated audio or video processor, as is common in the art for high-fidelity A/V hardware interfaces) and, in some instances, volatile and/or non-volatile memory' (e.g., RAM).
- an independent processor such as a dedicated audio or video processor, as is common in the art for high-fidelity A/V hardware interfaces
- volatile and/or non-volatile memory' e.g., RAM
- FIG. 10 illustrates one specific architecture for a computing device 10 for implementing one or more of the aspects described herein, it is by no means the only device architecture on which at least a portion of ihe features and techniques described herein may be implemented.
- architectures having one or any number of processors 13 may be used, and such processors 13 may' be present in a single device or distributed among any number of devices.
- a single processor 13 handles communications as well as routing computations, while in other aspects a separate dedicated communications processor may be provided.
- different types of features or functionalities may ' be implemented in a system according to the aspect that includes a client device (such as a tablet device or smartphone running client software) and server systems (such as a server system described in more detail below).
- the system of an aspect may' employ one or more memories or memory modules (such as, for example, remote memory block 16 and local memory' 11) configured to store data, program instructions for the general-purpose network operations, or other information relating to the functionality of the aspects described herein (or any combinations of the above).
- Program instructions may control execution of or comprise an operating system and/or one or more applications, for example.
- Memory 16 or memories 11, 16 may also be configured to store data structures, configuration data, encryption data, historical system operations information, or any other specific or generic non-program information described herein.
- At least some network device aspects may include nontransitoiy machine-readable storage media, which, for example, may be configured or designed to store program instructions, state information, and the like for performing various operations described herein.
- nontransitoiy machine-readable storage media include, but are not limited to, magnetic media such as hard disks, floppy disks, and magnetic tape; optical media such as CD-ROM disks; magneto-optical media such as optical disks, and hardware devices that are specially configured to store and perform program instructions, such as read-only memory devices (ROM), flash memory (as is common in mobile devices and integrated systems), solid state drives (SSD) and“hybrid SSD” storage drives that may combine physical components of solid state and hard disk drives in a single hardware device (as are becoming increasingly common in the art with regard to personal computers), memristor memoiy, random access memory (RAM), and the like.
- ROM read-only memory
- flash memory as is common in mobile devices and integrated systems
- SSD solid state drives
- HDD hard disk drives
- RAM random access memory
- such storage means may be integral and non-removable (such as RAM hardware modules that may be soldered onto a motherboard or otherwise integrated into an electronic device), or they ma ' be removable such as swappable flash memory' modules (such as“thumb drives” or other removable media designed for rapidly exchanging physical storage devices),“hot-swappable” hard disk drives or solid state drives, removable optical storage discs, or other such removable media, and that such integral and removable storage media may be utilized interchangeably ' .
- swappable flash memory' modules such as“thumb drives” or other removable media designed for rapidly exchanging physical storage devices
- “hot-swappable” hard disk drives or solid state drives such as “hot-swappable” hard disk drives or solid state drives, removable optical storage discs, or other such removable media
- Examples of program instructions include both object code, such as may be produced by a compiler, machine code, such as may be produced by an assembler or a linker, byte code, such as may ' be generated by for example a JAVATM compiler and may ' be executed using a Java virtual machine or equivalent, or files containing higher level code that may' be executed by' the computer using an interpreter (for example, scripts written in Python, Perl, Ruby, Groovy, or any other scripting language).
- object code such as may be produced by a compiler
- machine code such as may be produced by an assembler or a linker
- byte code such as may ' be generated by for example a JAVATM compiler and may ' be executed using a Java virtual machine or equivalent
- files containing higher level code that may' be executed by' the computer using an interpreter (for example, scripts written in Python, Perl, Ruby, Groovy, or any other scripting language).
- systems may be implemented on a standalone computing system.
- Computing device 20 includes processors 21 that may run software that carry out one or more functions or applications of aspects, such as for example a client application 24.
- Processors 21 may carry out computing instructions under control of an operating system 22 such as, for example, a version of MICROSOFT WINDOWSTM operating system, APPLE macOSTM or iOSTM operating systems, some variety of the Linux operating system, ANDROIDTM operating system, or the like.
- an operating system 22 such as, for example, a version of MICROSOFT WINDOWSTM operating system, APPLE macOSTM or iOSTM operating systems, some variety of the Linux operating system, ANDROIDTM operating system, or the like.
- one or more shared services 23 may be operable in system 20, and may be useful for providing common services to client applications 24.
- Services 23 may for example be WINDOWSTM services, user-space common services in a Linux environment, or any other type of common service architecture used with operating system 21.
- Input devices 28 may be of any type suitable for receiving user input, including for example a keyboard, touchscreen, microphone (for example, for voice input), mouse, touchpad, trackball, or any combination thereof.
- Output devices 27 may be of any type suitable for providing output to one or more users, whether remote or local to system 20, and may include for example one or more screens for visual output, speakers, printers, or any combination thereof.
- Memory 25 may be random- access memory having any structure and architecture known in the art, for use by processors 21, for example to run software.
- Storage devices 26 may be any magnetic, optical, mechanical, memristor, or electrical storage device for storage of data in digital form (such as those described above, referring to Fig, 10). Examples of storage devices 26 include flash memory, magnetic hard drive, CD-ROM, and/or the like.
- systems may be implemented on a distributed computing network, such as one having any number of clients and/or servers.
- a distributed computing network such as one having any number of clients and/or servers.
- FIG. 12 there is shown a block diagram depicting an exemplary architecture 30 for implementing at least a portion of a system according to one aspect on a distributed computing network.
- any number of clients 33 may be provided.
- Each client 33 may run software for implementing client- side portions of a system; clients may comprise a system 20 such as that illustrated in Fig. 11.
- any number of servers 32 may be provided for handling requests received from one or more clients 33.
- Clients 33 and servers 32 may communicate with one another via one or more electronic networks 31, which may be in various aspects any of tire Internet, a wide area network, a mobile telephony network (such as CDMA or GSM cellular networks), a wireless network (such as WiFi, WiMAX, LTE, and so forth), or a local area network (or indeed any network topology known in the art; the aspect does not prefer any one network topology over any other).
- Networks 31 may be implemented using any known network protocols, including for example wired and/or wireless protocols.
- servers 32 may call external services 37 when needed to obtain additional information, or to refer to additional data concerning a particular call.
- external services 37 may' comprise web-enabled sendees or functionality related to or installed on the hardware device itself.
- client applications 24 may obtain information stored in a server system 32 in the cloud or on an external service 37 deployed on one or more of a particular enterprise’s or user’s premises.
- clients 33 or servers 32 may make use of one or more specialized services or appliances that may be deployed locally or remotely across one or more networks 31.
- one or more databases 34 may be used or referred to by one or more aspects. It should be understood by one having ordinar skill in the art that databases 34 may ' be arranged in a wide variety of architectures nil using a wide variety of data access and manipulation means.
- one or more databases 34 may comprise a relational database system using a structured query language (SOL), while others may comprise an alternative data storage technology ' such as those referred to in the art as“NoSQL” (for example, HADOOP CASSANDRATM, GOOGLE BIGTABLETM, and so forth).
- SOL structured query language
- variant database architectures such as column-oriented databases, in-memory' databases, clustered databases, distributed databases, or even flat file data repositories may' be used according to the aspect.
- database any combination of known or future database technologies may be used as appropriate, unless a specific database technology or a specific arrangement of components is specified for a particular aspect described herein.
- database may refer to a physical database machine, a cluster of machines acting as a single database system, or a logical database within an overall database management system.
- security systems 36 and configuration systems 35 may make use of one or more security systems 36 and configuration systems 35.
- Security and configuration management are common information technolog )'' (IT) and web functions, and some amount of each are generally associated with any IT or web systems. It should be understood by one having ordinary skill in the art that any configuration or security subsystems known in She art now or in the future may be used in conjunction with aspects without limitation, unless a specific security 36 or configuration system 35 or approach is specifically required by the description of any specific aspect.
- FIG. 13 shows an exemplary overview of a computer system 4-0 as may be used in any of the various locations throughout the system. It is exemplary of any computer that may execute code to process data.
- Central processor unit (CPU) 41 is connected to bus 42, to which bus is also connected memory 43, nonvolatile memory 44, display 47, input/output (I/O) unit 48, and network interface card (NIC) 53.
- I/O unit 48 may, typically, be connected to keyboard 49, pointing device 50, hard disk 52, and real-time clock 51.
- NIC 53 connects to network 54, which may be the Internet or a local network, which local network may or may not have connections to the Internet.
- power supply unit 45 connected, in this example, to a main alternating current (AC) supply 46.
- AC alternating current
- functionality for implementing systems or methods of various aspects may be distributed among any number of client and/or server components.
- various software modules may be implemented for performing various functions in connection with the system of any particular aspect, and such modules may be variously implemented to run on server and/or client components.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Software Systems (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Devices For Executing Special Programs (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US15/879,182 US10514954B2 (en) | 2015-10-28 | 2018-01-24 | Platform for hierarchy cooperative computing |
PCT/US2019/014875 WO2019147751A2 (en) | 2018-01-24 | 2019-01-24 | Platform for hierarchy cooperative computing |
Publications (2)
Publication Number | Publication Date |
---|---|
EP3743813A2 true EP3743813A2 (en) | 2020-12-02 |
EP3743813A4 EP3743813A4 (en) | 2021-11-10 |
Family
ID=67395675
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP19743576.1A Withdrawn EP3743813A4 (en) | 2018-01-24 | 2019-01-24 | Platform for hierarchy cooperative computing |
Country Status (2)
Country | Link |
---|---|
EP (1) | EP3743813A4 (en) |
WO (1) | WO2019147751A2 (en) |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6356898B2 (en) * | 1998-08-31 | 2002-03-12 | International Business Machines Corporation | Method and system for summarizing topics of documents browsed by a user |
US8086741B2 (en) * | 2003-02-28 | 2011-12-27 | Microsoft Corporation | Method and system for delayed allocation of resources |
US8856333B2 (en) * | 2009-06-16 | 2014-10-07 | Microsoft Corporation | Datacenter execution templates |
US9948514B2 (en) * | 2014-06-30 | 2018-04-17 | Microsoft Technology Licensing, Llc | Opportunistically connecting private computational resources to external services |
CN105512723B (en) * | 2016-01-20 | 2018-02-16 | 南京艾溪信息科技有限公司 | A kind of artificial neural networks apparatus and method for partially connected |
RU2632126C1 (en) * | 2016-04-07 | 2017-10-02 | Общество С Ограниченной Ответственностью "Яндекс" | Method and system of providing contextual information |
CN107368895A (en) * | 2016-05-13 | 2017-11-21 | 扬州大学 | A kind of combination machine learning and the action knowledge extraction method planned automatically |
CN107301084B (en) * | 2017-07-05 | 2020-04-21 | 深圳先进技术研究院 | Virtual machine migration method and device of cluster server, server and storage medium |
-
2019
- 2019-01-24 WO PCT/US2019/014875 patent/WO2019147751A2/en unknown
- 2019-01-24 EP EP19743576.1A patent/EP3743813A4/en not_active Withdrawn
Also Published As
Publication number | Publication date |
---|---|
WO2019147751A2 (en) | 2019-08-01 |
EP3743813A4 (en) | 2021-11-10 |
WO2019147751A3 (en) | 2019-09-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11055140B2 (en) | Platform for hierarchy cooperative computing | |
US10310908B2 (en) | Dynamic usage balance of central processing units and accelerators | |
US9401835B2 (en) | Data integration on retargetable engines in a networked environment | |
US9256460B2 (en) | Selective checkpointing of links in a data flow based on a set of predefined criteria | |
US9262220B2 (en) | Scheduling workloads and making provision decisions of computer resources in a computing environment | |
US10324754B2 (en) | Managing virtual machine patterns | |
US11403131B2 (en) | Data analysis for predictive scaling of container(s) based on prior user transaction(s) | |
US11501232B2 (en) | System and method for intelligent sales engagement | |
US20150186427A1 (en) | Method and system of analyzing dynamic graphs | |
US20240020582A1 (en) | Parameter data sharing for multi-learner training of machine learning applications | |
US20210034374A1 (en) | Determining optimal compute resources for distributed batch based optimization applications | |
US10783441B2 (en) | Goal-driven composition with preferences method and system | |
US10885127B2 (en) | Machine-learning to alarm or pre-empt query execution | |
US20190258964A1 (en) | Runtime estimation for machine learning tasks | |
US11954564B2 (en) | Implementing dynamically and automatically altering user profile for enhanced performance | |
US11782704B1 (en) | Application refactoring with explainability | |
US20220004434A1 (en) | Platform for hierarchy cooperative computing application deployment | |
US11700306B2 (en) | Context aware edge computing | |
EP3743813A2 (en) | Platform for hierarchy cooperative computing | |
CN118176490A (en) | Task failover | |
US11163603B1 (en) | Managing asynchronous operations in cloud computing environments | |
US20240062069A1 (en) | Intelligent workload routing for microservices | |
US10771329B1 (en) | Automated service tuning | |
US20240143692A1 (en) | Method of Intelligent Matrix Solving Approach Enhanced with Integrated Realtime Machine Learning Training and Inference | |
US20230409922A1 (en) | Optimising evolutionary algorithm storage usage |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE |
|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
17P | Request for examination filed |
Effective date: 20200824 |
|
AK | Designated contracting states |
Kind code of ref document: A2 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
AX | Request for extension of the european patent |
Extension state: BA ME |
|
DAV | Request for validation of the european patent (deleted) | ||
DAX | Request for extension of the european patent (deleted) | ||
A4 | Supplementary search report drawn up and despatched |
Effective date: 20211012 |
|
RIC1 | Information provided on ipc code assigned before grant |
Ipc: G06Q 40/04 20120101ALI20211006BHEP Ipc: G06F 9/50 20060101AFI20211006BHEP |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE APPLICATION HAS BEEN WITHDRAWN |
|
18W | Application withdrawn |
Effective date: 20211201 |