Abstract: An anomaly detection system according to an embodiment is an anomaly detection system that executes anomaly detection of each of one or a plurality of simulators by using a controller. The controller causes each simulator to start a process of simulation, to transmit to the controller an existence notification indicative of existence of the simulator, at a predetermined cycle until an end of the process of simulation, and to transmit to the controller an end notification indicative of an end of the process if the process ends; determines that the simulator that is a transmission source of the existence notification is normal; determines that the process of the simulator that is a transmission source of the end notification ends, upon receiving the end notification; and detects that an anomaly occurs in the simulator.

Abstract: A method, device, and system for configuring a coating machine for coating a surface of a product using a coating substance are provided. The method includes determining a value associated with one or more parameters from a plurality of parameters associated with the coating operation. The method also includes predicting a value associated with at least one attribute associable with the coating substance based on the determined value associated with the one or more parameters using a trained machine learning model. The method includes configuring the coating machine for coating the surface using the coating substance based on the predicted value associated with the at least one attribute associable with the coating substance. The method also includes initiating a coating operation at the configured coating machine for coating the surface of the product using the coating substance.

Abstract: A method is provided for creating and executing a control program for controlling an automation system having a controller and a web server connected to the controller. The method includes creating a first version of a program code of a control program for the automation system in an input module of a web-based development environment executed in a web browser, in a code creating step; executing a translation module of the web-based development environment on the web server and translating the program code into a program code of a binary language, in a translating step; and executing the program code in the binary language with the aid of the controller of the automation system, in an executing step. An automation system is also provided.

Abstract: It is disclosed a PLC Program analysis method where a program (PROG) is translated (TRANS) into a model (MOD) in a logical framework, from which properties (Prop) are determined. Said properties coupled with user specifications (IntProp) are verified by an automated solver (SMT). If contraposition of a property (Prop) is satisfiable, counter-examples (PROOF NOK) representative of model inputs and internal memory values is provided. Counter-examples (PROOF NOK) are translated into error initial configurations (IniConf) of said model. Execution of the model is simulated (EXE) with said model error initial configurations (IniConf), and error intermediary configurations (AST-IntConf) of said model simulation are recorded up to said property violation. Error initial and intermediary configurations (Lad-IniConf, Lad-IntConf) of said original program (PROG) are derived from error initial configurations (IniConf) of said model and error intermediary configurations (AST-IntConf) of said model simulation and displayed.

Abstract: A method includes: accessing an attack record defining actions representing a previous known attack on a second computer network; initializing an attack graph; for each action, defining a set of behaviors—analogous to the action and executable by an asset on a target network to emulate an effect of the action on the second computer network—and storing the set of behaviors in a node in the attack graph; connecting nodes in the attack graph according to an order of actions in the known attack; scheduling the asset to selectively execute analogous behaviors stored in the set of nodes in the attack graph; accessing alerts generated by a set of security tools deployed on the target network; and characterizing vulnerability of the target network based on alerts, in the set of alerts, indicating detection and prevention of behaviors executed by the asset according to the attack graph.

Abstract: Systems and methods may generate code, for a model, with one or more service access points generated and at locations in the code based on an analysis of model constraints and deployment specifications (e.g., RTE specifications or OS specifications). The systems and methods may analyze the model and identify a functionality that needs an RTE service. The system and methods may receive deployment specifications. The systems and methods may generate code for the model, where an analysis of model constraints and the deployment specifications determine which service access points are generated and where in the code the service access points are located. In an embodiment, the code may be executed by different RTEs. In an embodiment, the systems and methods may determine, based on the analysis of the model constraints and the deployment specification, one or more admissible implementations for an RTE service that may be implemented in different ways.

Abstract: Techniques described herein relate to implementing mutation testing of software applications associated with continuous integration (CI) systems. A mutation test system may determine one or more portions of modified source code within an application codebase. Mutated applications may be generated based on the modified source code, and a mutation test system may determine subsets application test suites for execution based on the portions of the modified source code and/or other factors. In various examples, the mutation test system may use mappings between portions of source code and test subsets, and/or machine-learned models or heuristics-based techniques to determine subsets of test suites based on discreet source code modifications.

Abstract: A method of providing a calendar service includes obtaining time-series data, determining whether the time-series data correspond to an outlier, obtaining, based on the determination that the time-series data correspond to an outlier, non-time-series data linked to the time-series data corresponding to the outlier, and displaying at least one of the non-time-series data and the time-series data corresponding to the outlier on a calendar page.

Abstract: Aspects are described herein that are capable of providing a vehicle configuration for a vehicle that is optimized to maximize the customer's performance priorities while additionally complying with regulatory emissions requirements and equipment regulations. A machine learning (ML) predictive model is trained based on simulations run on combinations of vehicle configurations and routes and on real-world telematics data, and used to determine a vehicle configuration optimized for a representative route.

Abstract: Systems and methods automatically rescale an original electric motor model so that it models an electric motor of a different size. The original electric motor model may be coupled to a motor controller model, and the systems and methods may also rescale the original motor controller model to produce a rescaled motor controller model matched to the rescaled electric motor model. The original electric motor model may include motor parameters and motor lookup tables, and the original motor controller model may include controller parameters and controller lookup tables. Rescaling factors indicating the size of the new electric motor being modeled may be received, and ratios may be computed as a function of the rescaling factors. Original motor parameters and controller parameters may be rescaled based on the ratios. Original motor lookup tables and controller lookup tables may be reshaped based on the ratios.

Abstract: An API classification unit (231) extracts as omittable internal data processing, from among internal data processing included in execution data processing which is data processing executed by an outside data processing device which is a data processing device outside, internal data processing that is required to be executed when the outside data processing device executes the execution data processing, but is omittable from the execution when a data processing apparatus (10) executes the execution data processing, due to a difference in an execution environment between the outside data processing device and the data processing apparatus (10). An execution unit (235) executes the execution data processing while omitting the execution of the omittable internal data processing extracted by the API classification unit (231).

Abstract: The present disclosure provides a scalable acceleration of data processing in Machine Learning pipeline which is unavailable in conventional methods. Initially, the system receives a dataset and a data processing code. A plurality of sample datasets are obtained based on the received dataset using a sampling technique. A plurality of performance parameters corresponding to each of the plurality of sample datasets are obtained based on the data processing code using a profiling technique. A plurality of scalable performance parameters corresponding to each of a plurality of larger datasets are predicted based on the plurality of performance parameters and the data processing code using a curve fitting technique. Simultaneously, a plurality of anti-patterns are located in the data processing code using a pattern matching technique.

Abstract: A system and method for detecting code initialization actions is utilized for performing static analysis of an application code including an external initialization component. The method includes receiving an application code including an external initialization code component, the application code deployed in a cloud computing environment; detecting in the application code an anchor point, the anchor point including a call to another code object; inserting a hook into the application code based on a location of the anchor point in the application code; and receiving a result from the hook in response to executing the application code.

Abstract: In a computer-implemented method for improving performance of a database optimizer, a processor calculates a first execution time for a database query executing through a first access path of a database system. A processor generates a second access path. A processor determines a second execution time via a simulation of the database query through the second access path. A processor determines that the second execution time is less than the first execution time. A processor, responsive to determining that the second execution time is less than the first execution time, sends the second access path to the database optimizer.

Abstract: The present technology is directed to visualizing a Wi-Fi signal propagation in 3-D at various heights and locations. The present technology can calculate a radio frequency (RF) propagation pattern for a Wi-Fi access point (AP) based on a RF propagation model for the Wi-Fi AP and overlay the RF propagation pattern for the Wi-Fi AP over a visualization of the building plan to present a 3-D visualization of the RF propagation pattern of the 3-D space. In particular, the present technology can project a plurality of ray-paths in various directions in a 3-D space originated from the Wi-Fi AP and determine whether the ray-paths interface with objects defined in the building plan. The present technology can segment the respective ray-path into contiguous segments of substantially uniform mediums for each ray-path that interface with the objects and determine a RF signal strength at points along the contiguous segments of the ray-paths.

Abstract: Devices and techniques for loading contexts in a coarse-grained reconfigurable array processor are described herein. A system or apparatus may include context load circuitry operable to load context for a coarse-grained reconfigurable array processor, where the context load circuitry is configured to: (a) receive a kernel identifier; (b) access a first registry to obtain a context mask base address; (c) determine a context mask address from the context mask base address and the kernel identifier; (d) access a second registry to obtain a context state base address; (e) determine a context state address from the context state base address and the kernel identifier; (f) use a context mask at the context mask address to determine corresponding active context state; and (g) load the corresponding active context state into the coarse-grained reconfigurable array processor.

Abstract: An approach is provided that receives event data that correspond to detected activities performed by a user on one of a set of one or more computer systems. The detected activities are performed by the user over a time duration. The approach analyzes the event data using time-based models. Each of the time-based models correspond to a different time interval that is included in the time duration. The analysis results in time-based risk scores pertaining to the user for each of the different time intervals. An action is then performed based on an overall security risk score of the user with the overall security risk score of the user being calculated based on the different time-based risk scores.

Abstract: A system and method for generating a netlist of a memory device includes receiving a logical netlist file including memory instances and placement information for the memory device. Each memory instance includes leaf cells. Further, a location of a first leaf cell and a location of a second leaf cell of the leaf cells of a first memory instance is determined based on the placement information. A first net segment between the first leaf cell and the second leaf cell is generated based on the location and parasitic elements of the first leaf cell and the location and parasitic elements of the second leaf cell. A parasitic netlist is generated based on the first net segment and the parasitic elements of the first leaf cell and parasitic elements of the second leaf cell.

Abstract: A method for generating an upgrade recommendation for a communication network includes: analyzing a dataset to generate a request category, in which the dataset comprises at least a request, a response to the request, and a corresponding response time, in which the request and the response are communicated over the communication network; generating a network slice corresponding to the request category; executing an emulation using the network slice and the request to generate an emulated response time for the request; comparing the emulated response time against the corresponding response time; determining, based on the comparison, a response time variation for the request; and generating the upgrade recommendation based on the response time variation.

Abstract: An example system includes a processor to monitor activity on a database server to generate an events stream. The processor can convert the events stream into a time series that approximates activity load at the database server using an exponential smoothing. The processor can also send the time series to a streaming analytics engine.

Abstract: The present disclosure relates to a method and apparatus for predicting application service response time in a communication system. Particularly, the present disclosure relates to a method, apparatus and system for predicting response time required to provide an application service to a user in response to the user's request of the application service in a server of a communication system. A method for predicting application service response time according to an embodiment of the present disclosure includes: data collection for collecting network data associated with a user application service, in response to a request for the application service by a user device; and data analysis for performing, by utilizing the collected data, a response time predictive analysis, in order to predict a user's application service response time. The data analysis selects an optimal prediction model by initially applying a response time prediction model and testing the initial prediction model that is applied.

Abstract: An attribute-value extraction unit (101) extracts as a plurality of model-generation attribute values, a plurality of attribute values belonging to an attribute associated with a monitoring subject for anomaly detection. A division-data generation unit (102) extracts for each model-generation attribute value, a normal event associated with the model-generation attribute value from normal data indicating a plurality of normal events each of which is found out to be normal, each of which is associated with one attribute value of the plurality of attribute values, and each of which includes a plurality of characteristics, and generates for each model-generation attribute value, division data indicating the extracted normal event. A characteristic selection unit (103) selects a combination of characteristics to be used for generation of a normal model to be used for anomaly detection, from a plurality of characteristics included in a plurality of normal events indicated in a plurality of pieces of division data.

Abstract: A method for evaluating an updated analytical procedure in a monitoring system, comprising a plurality of monitoring device arranged to monitor similar environments, is provided. The method comprises identifying available processing resources, selecting a first monitoring device for which available processing resources have been identified, selecting a second monitoring device, acquiring monitoring data by the second monitoring device, and performing a current analytical procedure on the monitoring data. The method further comprises sending the monitoring data to the first monitoring device, performing, in the first monitoring device, an updated analytical procedure on the monitoring data, and evaluating the updated analytical procedure based on the outcome of the current analytical procedure and the updated analytical procedure.

Abstract: A program is executed in a first mode of operation in a controlled environment in accordance with normal operations without malicious behavior. An acceptable behavior model is generated based on a plurality of sequences of events that occur during the normal operation of the program. The acceptable behavior model is indicative of normal behavior of the program that occurs during the normal operation. Then the program is executed in a second mode of operation in an operational environment. An operational sequence of events (determined during the second mode of operation) is compared with the acceptable behavior model. When there is a match between the operational sequence of events and the acceptable behavior model, execution in the second mode of operation continues. When there is not a match between the operational sequence of events and the acceptable behavior model, execution in the second mode of operation is halted.

Abstract: In an approach for predicting an optimal system upgrade, a processor detects an upgrade is available for a system of a first user computing device. A processor gathers a set of live data from an environment of the first user computing device using a process mining software. A processor compares the set of live data to a set of historical data stored in a knowledge corpus to identify a second user computing device similar to the first user computing device. Responsive to identifying the second user computing device, a processor performs a pattern recognition to generate one or more critical recommendations associated with the upgrade for the system of the first user computing device. A processor schedules the upgrade for the system of the first user computing device during a period of time, wherein the period of time is scheduled consistent with the one or more critical recommendations.

Abstract: There is provided a method for programming an industrial robot, where distributors and integrators can present accessories that run successfully at end users. Also the developer can define customized installation screens and program nodes for the end user. There is provided a software platform, where the developer can define customized installation screens and program nodes for the end user thereby extending an existing robot system with customized functionalities by still using the software platform available in the robot system. Hereby a robot developer can define customized installation screens and program nodes for the end user. These can, for example, encapsulate complex new robot programming concepts, or provide friendly hardware configuration interfaces.

Abstract: This application relates to a non-communication electronic warfare system design analysis system based on engineering modeling. In one aspect, the system includes a scenario unit transmitting a simulation threat signal corresponding to an input scenario and a threat signal simulator transmitting an actual threat signal. The system may also include an electronic warfare support receiving model unit allocating a jamming technique. The system may further include an electronic warfare support receiving analyzer allocating the jamming technique to the received actual threat signal and an electronic attack jamming model unit modelling a device for generating a simulation jamming signal. The system may further include an electronic attack jamming generator generating an actual jamming signal and a simulation situation demonstration controller analyzing performance by using the actual jamming signal and the actual threat signal.

Abstract: A computer-implemented method of machine-learning is described that includes obtaining a test dataset of scenes. The test dataset belongs to a test domain. The method includes obtaining a domain-adaptive neural network. The domain-adaptive neural network is a machine-learned neural network taught using data obtained from a training domain. The domain-adaptive neural network is configured for inference of spatially reconfigurable objects in a scene of the test domain. The method further includes determining an intermediary domain. The intermediary domain is closer to the training domain than the test domain in terms of data distributions. The method further includes inferring, by applying the domain-adaptive neural network, a spatially reconfigurable object from a scene of the test domain transferred on the intermediary domain. Such a method constitutes an improved method of machine learning with a dataset of scenes comprising spatially reconfigurable objects.

Abstract: The disclosure is to constitute, while reducing a cost for collecting training data used in machine learning that makes a control module acquire an ability to control a robot device, the control module operatable in an actual environment by the machine learning. A learning device according to one aspect of the present invention executes machine learning of an extractor by using a first learning data set constituted by a combination of simulation data and first environmental information and a second learning data set constituted by a combination of actual data and second environmental information. Further, a learning device according to one aspect of the present invention executes machine learning of a controller by using a third learning data set constituted by a combination of third environmental information, state information, and a control command.

Abstract: A method for testing a basic parameterization of a component in an automation system is provided. The method includes: starting a test run of the component in the automation system with the basic parameterization, measuring of a measured value data record during trial operation, access to a machine learning module comprising a pre-trained neural network, wherein the pre-trained neural network is pre-trained to calculate a target parameterization for the respective component for a measured value data set, wherein the basic parameterization is compared with the calculated target parameterization and in the event of deviation a result message for adapting the basic parameterization is provided, and receipt of the provided result message for adaptation of the basic parameterization.

Abstract: The following description is directed to a logic repository service. In one example, a method of a logic repository service can include receiving a first request to generate configuration data for configurable hardware using a specification for application logic of the configurable hardware. The method can include generating the configuration data for the configurable hardware. The configuration data can include data for implementing the application logic. The method can include encrypting the configuration data to generate encrypted configuration data. The method can include signing the encrypted configuration data using a private key. The method can include transmitting the signed encrypted configuration data in response to the request.

Abstract: Discussed herein are devices, systems, and methods for dynamic IPB product updates or generation. A method can include storing, by a memory device, information used to generate IPB products, receiving, by a server coupled to the memory device, a request for an IPB product of the IPB products, receiving, by the server and from the memory device, information that satisfies conditions of a query, generating the requested IPB product based on the received information, and providing the requested IPB product to a client device.

Abstract: A method for iteratively autotuning a high-performance computing system that depends on a set of parameters. Performance is first evaluated two or more times with the current values of the parameters. Afterward at least two evaluations, the median performance is evaluated. The median is then tested against a rule based on a filtering threshold. If the median does not the rule, the current values of the parameters are discarded, and the method is restarted with at least one other value generated by an optimization module; otherwise, a resampling method is performed based on the median and on a confidence interval that decreases with the number of steps of the optimization method.

Abstract: Test scenarios can be generated for decision models on a computer according to some examples. In one example, a system can determine groupings of user input tests for a decision model based on outputs produced by inputting the user input tests into the decision model. Each grouping can include a portion of the user input tests. For each grouping, the system can determine an impact score for each input for each user input test. The system can determine sub-groupings within each grouping based on the impact scores for the inputs. The system can then identify a particular user input test within each sub-grouping. The system can then generate a test scenario file for the decision model based on the particular user input test for use in testing the decision model.

Abstract: A system for developing a simulation of a process. In one aspect, a system creates a first model within the simulation. The first model represents a part of the process and comprises a first port to which other models may be connected. The system also creates a second model within the simulation. The second model represents another part of the process and comprises a second port to which other models may be connected. The system then connects the first port and the second port together. Upon connection, the system allocates a memory location as a connection variable that represents a type of information transfer between the first and second ports. A first port variable, which represents a value transferrable through the first port, is set to reference the value at the allocated memory location. Similarly, a second port variable, which represents a value transferrable through the second port, is also set to reference the value at the allocated memory location.

Abstract: Examples of device-driven management are described. A management console can include a set of workflow objects to use in a workflow creation user interface. Workflow objects can be positioned in the workflow creation user interface area based on user manipulation. A device state criteria overlay can be painted on a connector workflow object to indicates that a branch of executable instructions corresponding to the connector workflow object is performed where a client device corresponds to the specified device state criteria.

Abstract: A method, a computer program product, and a system for pre-pooling virtual machine components prior to application startup. The method includes reading a virtual machine pool manifest by a virtual machine pool manager. The virtual machine pool manifest includes an initial number of virtual machines to launch, how many virtual machines to allocate resources to, and an amount of resources to allocate based on a virtual machine resource definition. The method also includes launching a plurality of base virtual machines into a virtual machine pool and based on the initial number provided by the virtual machine pool manifest. The base virtual machines lack initial application assignments. The method further includes allocating initial resources to a portion of the base virtual machines based on a virtual machine resource definition in the virtual machine pool manifest. The method includes loading core program packages into the portion of the base virtual machines.

Abstract: One implementation of the present disclosure is a building system of a building including one or more memory devices having instructions stored thereon, that, when executed by one or more processors, cause the one or more processors to receive a selection including a context and a twin function and generate an operation to monitor a building graph based on the context, the operation identifying whether one or more new events are added to the building graph, the one or more new events affecting a processing result. The instructions cause the one or more processors to cause the twin function to execute responsive to identifying the one or more new events added to the building graph that affect the processing result, execute the twin function based on the one or more new events and the context to generate the processing result.

Abstract: Methods and systems for generating a recovery scheduling solution in response to a scheduling disruption are disclosed.

Abstract: Various embodiments of the present technology generally relate to the characterization and improvement of software applications. More specifically, some embodiments relate to systems and methods for modeling code behavior and generating new versions of the code based on the code behavior models. In some embodiments, a method of improving a codebase includes recording a run of the existing code, characterizing the code behavior via one or more models, prototyping new code according to a target language and target environment, deploying the new code to the target environment, and comparing the behavior of the new code to the behavior of the existing code. In some implementations, generating new code based on the behavior models includes using one or more machine learning techniques for code generation based on the target language and environment.

Abstract: Systems and methods for training prediction models are illustrated. One embodiment includes a method for training a prediction model in a network. The method includes steps for receiving context data for a portable device in a system, wherein the context data includes localization data that describes a location of the portable device, identifying a predicted stationary device from several stationary devices that is predicted based on the context data using a prediction model, receiving input identifying a target stationary device from the several stationary devices, generating training data based on the predicted stationary device and the received input, updating the prediction model based on the generated training data.

Abstract: A caching system including a first sub-cache, and a second sub-cache, coupled in parallel with the first cache, for storing cache data evicted from the first sub-cache and write-memory commands that are not cached in the first sub-cache, and wherein the second sub-cache includes: color tag bits configured to store an indication that a corresponding cache line of the second sub-cache storing write miss data is associated with a color tag, and an eviction controller configured to evict cache lines of the second sub-cache storing write-miss data based on the color tag associated with the cache line.

Abstract: Systems and methods for automatically commissioning and operating a heating, ventilation, or air conditioning (HVAC) system for a building site are provided. An exemplary method includes constructing a model using physical equipment of the HVAC system and relationships between the physical equipment. The model indicates connections between the physical equipment and one or more resources produced or consumed by the physical equipment. The method includes generating a mapping between points of the physical equipment at the building site and corresponding variables of the model, using the model to generate values of one or more control variables of the model, and operating the physical equipment by providing the values of the control variables to corresponding points of the physical equipment based on the mapping.

Abstract: One variation of a method for emulating a known attack on a computer network includes: generating a set of data packets by recombining packet fragments within a packet capture file representing packet fragments transmitted between machines during a prior malicious attack on a second network; defining transmission triggers for transmission of the set of data packets between pairs of agents connected to a target network based on timestamps of packet fragments in the packet capture file; initiating transmission of the set of data packets between the pairs agents according to the set of transmission triggers to simulate the malicious attack on the target network; and, in response to absence of a security event related to the simulation in a log of a security technology deployed on the target network, generating a prompt to reconfigure the security technology to respond to the malicious attack.

Abstract: A processor may receive at least one test application corresponding to an application profile. The processor may simulate the at least one test application in a non-production environment for a plurality of infrastructure configurations to generate a plurality of test performance results. The processor may evaluate the plurality of test performance results to identify an optimal infrastructure configuration from among the plurality of infrastructure configurations for the application profile. The processor may apply the optimal infrastructure configuration to an application corresponding to the application profile that is deployed in a production environment.

Abstract: A method retrains a cognitive interaction information system. One or more processors monitors an interactive communication between a user and the cognitive interactive information system. The processor(s) identifies an incorrect answer returned by the cognitive interactive information system to a question asked by the user; reevaluates the question utilizing alternative intent classifications to form alternative answers; and presents the alternative answers to the user. The processor(s) receive a selection from the alternative answers from the user; updates processing rules to generate updated processing rules that are based on the selection; and retrains the cognitive interaction information system based on the updated processing rules.

Abstract: A system for read-access of a regulated system, the system comprising a specialized data store, at least one memory, and a flexible reader. The specialized data store able to receive at least a portion of a set of procedures that define a respective set of systematic data and executable operations. The at least one memory including at least one set of data related to the set of procedures.

Abstract: Methods and systems for managing security in a cloud computing environment are provided. Exemplary methods include: gathering data about workloads and applications in the cloud computing environment; updating a graph database using the data, the graph database representing the workloads of the cloud computing environment as nodes and relationships between the workloads as edges; receiving a security template, the security template logically describing targets in the cloud computing environment to be protected and how to protect the targets; creating a security policy using the security template and information in the graph database; and deploying the security policy in the cloud computing environment.

Abstract: A thread management circuit of a processing system stores a thread identifier table and a thread completion table. The thread management circuit receives, from a processor core, a request for execution of a portion of an application by an accelerator circuit. The thread management circuit allocates a thread identifier available in the thread identifier table to the processor core for the execution of the portion by the accelerator circuit. The thread management circuit communicates a response and an acceleration request, both including the allocated thread identifier, to the processor core and the accelerator circuit, respectively. The thread management circuit communicates a thread joining response to the processor core based on a received thread joining request and an indication by the thread completion table that the execution of the portion by the accelerator circuit is complete. The executed portion is integrated with the application based on the thread joining response.

Abstract: A digital environment includes multiple computing nodes and a scheduling system that assigns workloads to computing nodes. The scheduling system includes an equivalence-class-based resource usage prediction system that receives a workload request and predicts an equivalence class for that workload request based on resource usage over time by the workload request or metadata associated with the workload request. The scheduling system also includes a workload assignment system that assigns the workload request to one or more of the computing nodes based on the predicted equivalence class. The number of equivalence classes is small relative to the total number of workloads that are scheduled (as an example, 10 to 15 equivalence classes for a total number of workloads in the tens or hundreds of thousands).