Abstract: The present invention provides a method and a device for service management, wherein the method comprises: detecting popularity of use of a remote target service by an application; and when the popularity reaches a first threshold value, loading the target service to the application locally. The present invention improves the response efficiency and adaptive capability of a distributed system.

Abstract: Simplifying understanding of procedure dependencies in a form definition used in user interfaces. In an embodiment, a form definition including a set of procedures (identified by a corresponding identifier) are examined to determine pairs of procedures such that a caller procedure of a pair invokes/refers a called procedure in the pair. For each of pair of procedures determined, a link associated with an identifier identifying the called procedure is included in the caller procedure. The link is designed to display the called procedure in response to a user action associated with the identifier in a display of the caller procedure.

Abstract: Provided is a method for entering a desktop of an operating system, including: entering a screen locking interface, the screen locking interface displaying a prompt for password input; acquiring an unlocking password input on the screen locking interface; comparing the unlocking password with preset passwords; if the unlocking password is a privacy information setting password, entering a setting interface for setting a hiding function for personal privacy information; or, if the unlocking password is a switching password, entering a desktop of an operating system corresponding to the switching password.

Abstract: An input device according to an exemplary embodiment of the present invention includes: a band including a contracted or extended flexible area; a sensor which detects a change of the flexible area and outputs a signal corresponding to the detected change of the flexible area; and a main body which is connected to the band and determines a cause operation for the change of the flexible area based on the signal output from the sensor, in which the flexible area includes fixed units, and connection units which connect the fixed units and are contracted or extended, and a distance between the fixed units is changed according to the cause operation.

Abstract: Systems and methods are provided for generating interactive user interface pages from a data store. At least a portion of metadata is received from a data store, wherein the appearance and functionality of a plurality of interactive user interface pages is based on the metadata. The display and functionality of an interactive maintenance page is generated based on at least a portion of the metadata, wherein the interactive maintenance page is one of the plurality of interactive user interface pages, and wherein the interactive maintenance page maintains the plurality of interactive user interface pages. One or more instructions are received from the interactive maintenance page. The metadata is updated based upon the one or more instructions, and the appearance and functionality of at least one of the interactive user interface pages and/or the interactive maintenance page is modified based on the metadata.

Abstract: In one example, a multi-dimensional data set is converted into a single-dimensional data set that includes structural information associated with the multi-dimensional data set, such that root nodes and each node in the single-dimensional node for which each ancestor node of the node is an expanded node is assigned a reference. A view is caused to be displayed, including using the single-dimensional data set to render at least a selected portion of the single-dimensional data set.

Abstract: Systems and methods using one or more processors of a user electronic device for receiving user interactions and sending data related to those user interactions to another device. This may include receiving input comprising an interaction with the user device, identifying an application installed on the user device to which the interaction is directed to, and sending data over a network comprising an indication of a model of the user device, an identifier of the application, and an indication of the interaction. Sending the data may initiate one of a video playback, a tutorial, a demonstration, or a simulation, based on the data, on a device different than the user device. In some arrangements, the interaction is directed to a process running on the user device instead of an application.

Abstract: A computer-implemented method and system creates an interactive learning environment. Windows are created for guiding a user through a series of steps to perform a task supported by a computer software application. The windows contain textual and/or visual content that informs the user of the elements to accomplish the task. At least one of the windows contains a pointer indicating a location of a command in a user interface of the computer software application. User interaction is enabled during the execution of the series of steps, allowing for user input.

Abstract: A computerized personal assistant communicatively couples to a computer database including a plurality of available skills for the computerized personal assistant. The computerized personal assistant recognizes a current context of the user. The computerized personal assistant operates a previously-trained machine learning classifier to assess a match confidence for a candidate skill, the match confidence indicating a quality of match between the current context and a reference context previously associated with the candidate skill. The computerized personal assistant executes instructions defining an assistive action associated with the candidate skill responsive to the match confidence exceeding a predefined match confidence threshold. The computerized personal assistant executes the instructions defining a complementary help action associated with the candidate skill responsive to the match confidence not exceeding the predefined match confidence threshold.

Abstract: Methods, systems, and computer program products for adapting the tone of the user interface of a cloud-hosted application based on user behavior patterns are provided herein. A computer-implemented method includes analyzing behavior of a user with respect to one or more software applications; automatically detecting, from a pre-established collection of multiple software tone settings, one or more appropriate software tone settings to be applied to the one or more software applications based on the analyzed behavior; dynamically updating the software tone settings of the one or more software applications, wherein updating comprises (i) defining the value for one or more strings of the one or more software applications as one or more run-time attributes and (ii) resolving the one or more run-time attributes upon detecting the one or more appropriate software tone settings; and outputting the one or more dynamically updated software applications to at least a display.

Abstract: A PC, which is an information processing device, simulates a control program and displays the result of the simulation on a monitor. The PC acquires a value, and identification information identifying a variable for at least one variable at an arbitrary time during the simulation. The PC modifies the control program in accordance with an instruction from a user. The PC simulates the modified control program with the value of the variable identified by the acquired identification information as the value of the acquired variable. The PC displays the result of simulating the modified control program on the monitor.

Abstract: Methods and apparatuses relating to preventing the execution of a modified instruction. In one embodiment, an apparatus includes a hardware binary translator to translate an instruction to a translated instruction, and a consistency hardware manager to prevent execution of the translated instruction by a hardware processor on detection of a modification to a virtual to physical address mapping of the instruction after the translation.

Abstract: Described herein are systems, methods, and software to provide enhanced security when opening applications. In one implementation, an application service receives, over a network, an application request from an end user device to support a file open request on the end user device, wherein the application request occurs based on a security status of a file associated with the file open request. The application service further allocates a virtual node with an application to the end user device to support the application request, provides a remote connection for the application to the end user device, and opens the file in the application of the virtual node to support the file open request.

Abstract: Aspects of the present disclosure involve systems and methods for the automatic packaging and deployment of virtual machines in a virtual environment. More particularly, the disclosure involves a system that automatically packages two virtual machines, a console virtual machine and a template virtual machine, in a virtual application. The template virtual machine initiates communication with the console virtual machine to generate multiple virtual machine instances.

Abstract: A computing device for displaying images includes one or more processors to obtain graphical data that has been generated by one or more applications associated with a first virtual machine, and to map one or more virtual apertures or one or more actual buffers associated with the first virtual machine to a textual buffer associated with a second virtual machine, and process the graphical data based on the mapping. The virtual apertures are in a virtual memory space and the actual buffers being in a physical memory space. The graphical data includes rendered composite graphical data representing a desktop and one or more application windows. The one or more processors allocate physical memory space in the textual buffer associated with the second virtual machine to support the mapping. A display device displays images using the processed graphical data.

Abstract: A system and method for providing dynamic I/O virtualization is herein disclosed. According to one embodiment, a device capable of performing hypervisor-agnostic and device-agnostic I/O virtualization includes a host computer interface, memory, I/O devices (GPU, disk, NIC), and efficient communication mechanisms for virtual machines to communicate their intention to perform I/O operations on the device. According to one embodiment, the communication mechanism may use shared memory. According to some embodiments, the device may be implemented purely in hardware, in software, or using a combination of hardware and software. According to some embodiments, the device may share its memory with guest processes to perform optimizations including but not limited to a shared page cache and a shared heap.

Abstract: Implementations describe hibernating virtual machines (VMs). In one implementation, a method includes receiving, by a first virtual processor of a VM executed by a processing device, an instruction from a sleep-handling module of the VM to invoke a sleep function of a hypervisor executed by the processing device to manage the VM, transmitting, by the first virtual processor of the VM, a request to the hypervisor to put the VM to sleep, wherein the hypervisor, in response to the request, to stop the first virtual processor subsequent to stopping a second virtual processor of the VM, and initiating execution on the first virtual processor and on the second virtual processor responsive to the hypervisor re-starting the first and second virtual processors in response to a wake event.

Abstract: According to an embodiment, a processing network is configured to access a reputation system, provide a first user interface, provide a second user interface, receive a request from a user through the second user interface, retrieve user information from the reputation system, assign a plurality of human intelligence tasks based on the request, receive a plurality of results from the plurality of human intelligence tasks, confirm an identity and a status of the user based on the user information and the plurality of results, execute the request based on confirming the identity and the status of the user, and store evidence of execution of the request as an immutable data entry in a permanent database. The first user interface includes information from the reputation system, and the plurality of human intelligence tasks are assigned to a plurality of participants through the first user interface.

Abstract: In one approach, an import mechanism allows new hardware intrinsics to be utilized by writing or updating a library of source code, rather than specifically modifying the virtual machine for each new intrinsic. Thus, once the architecture is in place to allow the import mechanism to function, the virtual machine itself (e.g. the code which implements the virtual machine) no longer needs to be modified in order to allow new intrinsics to be utilized by end user programmers. Since source code is typically more convenient to write than the language used to implement the virtual machine and the risk of miscoding the virtual machine is minimized when introducing new intrinsics, the import mechanism described herein increases the efficiency at which new hardware intrinsics can be introduced.

Abstract: A setting support program related to monitoring of a system causes a computer to execute a process including (1) receiving an input of first system information including a first configuration element group of a first system, (2) referring to a storage unit that stores a second configuration element group of a second system and setting information related to monitoring of the second system for a plurality of the second systems to identify setting information related to the monitoring of the second system corresponding to the second configuration element group having predetermined similarity to the first configuration element group included in the first system information, and (3) outputting the identified setting information as setting information of the first system.

Abstract: A new approach is proposed that contemplates systems and methods to support virtio-based data packet path optimization for live virtual machine (VM) migration for Linux. Specifically, a data packet receiving (Rx) path and a data packet transmitting (Tx) path between a VM running on a host and a virtual function (VF) driver configured to interact with a physical network device of the host to receive and transmit communications dedicated to the VM are both optimized to implement a zero-copy solution to reduce overheads in packet processing. Under the proposed approach, the data packet Tx path utilizes a zero-copy mechanism provided by Linux kernel to avoid copying from virtio memory rings/Tx vrings in memory of the VM. The data packet Rx path also implements a zero-copy solution, which allows a virtio device of the VM to communicate directly with the VF driver of the network device while bypassing a macvtap driver entirely from the data packet Rx path.

Abstract: Problem ticket analysis and consolidation are performed by monitoring a virtual computing environment including at least a first physical host, a second physical host, and one or more virtual machines, to detect one or more problems. A problem ticket is opened in response to detecting at least one problem of the one or more problems at the one or more virtual machines, the first physical host or the second physical host. A migration is identified for a virtual machine of the one or more virtual machines from the first physical host to the second physical host. A plurality of problem tickets are analyzed to identify at least two correlated problem tickets. The identified at least two problem tickets are consolidated into a single problem ticket.

Abstract: Apparatus and methods can be implemented to perform software testing or to perform emulated hardware testing using a cloud architecture that can utilize centralized testing technology and can enable scaling up to test for multiple tenants and scaling up to arbitrary numbers of programs tested for each tenant. A user can configure an initial test virtual machine on a cloud platform for a cloud service over a physical network such as the Internet. Components of the cloud architecture can create a set of clones of the initial test virtual machine and inject tools into each clone for testing. Testing of one or more clones of the set can be conducted in an environment isolated from the physical network and isolated from a backend of the cloud service. Additional apparatus, systems, and methods are disclosed.

Abstract: Systems and methods for disabling in-memory caching of a virtual machine during migration are disclosed. An example method may include receiving, at a migration agent of a source host machine, a request to migrate a virtual machine (VM) hosted by the source host machine to a destination host machine, determining that a migration caching attribute of the VM is enabled, instructing a VM agent executing on the VM to disable caching of the VM in view of determining that the migration caching attribute of the VM is enabled, receiving acknowledgment of successful disablement of caching on the VM, and responsive to receiving the acknowledgment, initiating migration of the VM from the source host machine to the destination host machine.

Abstract: A parallel processing method for a traffic having an order, which supports virtual core automatic scaling in a virtual environment having at least one virtual machine (VM) and at least one virtual core queue for each VM, may comprise identifying a flow of the traffic according to attributes of the traffic; determining a virtual core queue to which the traffic is mapped according to the identified flow and queuing the traffic to the determined virtual core queue; and transferring the traffic from the determined virtual core queue to a VM corresponding to the determined virtual core queue through a virtual switch. Also, traffic belonging to a same flow may be queued to a same virtual core queue and processed by a same virtual core.

Abstract: Described systems and methods allow protecting a hardware virtualization system from malicious software. Some embodiments use a hybrid event notification/analysis system, wherein a first component executing within a protected virtual machine (VM) registers as a handler for processor exceptions triggered by violations of memory access permissions, and wherein a second component executing outside the respective VM registers as a handler for VM exit events. The first component filters permission violation events according to a set of rules and only notifies the second component about events which are deemed relevant to security. The second component analyzes notified events to detect malicious software.

Abstract: Techniques for zero-loss workload mobility with segment routing for virtual machines are presented. The techniques include receiving, by a virtual router, an electronic message destined for a first virtual machine running on a first physical machine and checking a first virtual machine state for the first virtual machine. In response to determining that it is associated with a running state indicating the first physical machine, inserting a segment routing header including an indication of the source virtual machine, the first physical machine, and the first virtual machine. In response to determining that it is associated with a migration state, inserting, by the virtual router, a segment routing header indicating the source virtual machine, an END.S for the first physical machine, the first virtual machine; and an END.SBUF for a second physical machine. The message is then routed based at least in part on the inserted segment routing header.

Abstract: Some embodiments of the present invention include a method comprising: accessing units of network storage that encode state data of respective virtual machines, wherein the state data for respective ones of the virtual machines are stored in distinct ones of the network storage units such that the state data for more than one virtual machine are not commingled in any one of the network storage units.

Abstract: Controlling usage of virtual disks in a data-processing center hosting a set of virtual machines is described where each virtual disk comprises a set of software artifacts. A corresponding method comprises detecting a request of adding a new virtual disk comprising a set of new software artifacts to a selected one of the virtual machines, determining composition information of the new virtual disk, which includes an indication of the new software artifacts, retrieving an indication of usage entitlements of the new software artifacts in the data-processing center, enabling or preventing the attachment of the new virtual disk to the selected virtual machine according to a comparison between the composition information and the usage entitlements, attaching the new virtual disk to the selected virtual machine in response to the enabling thereof, and updating the usage entitlements in response to the attachment of the new virtual disk.

Abstract: Aspects of the disclosure provide for mechanisms for automatic generating synopsis data of command-line commands. A method of the disclosure includes processing source code implementing a command; identifying, in view of the processing, a plurality of command options related to the command; generating, by a processing device, relationship data representing dependencies of the command options; and generating, by the processing device, synopsis data for the command in view of the relationship data. In some embodiments, the relationship data may include a graph, wherein the graph including an arc that associates a first node of the graph with a second node of the graph. The first node may correspond to the first command option. The second node may correspond to the second command option.

Abstract: A multi-core processing device includes a first processor core and a second processor core. The computation capability of the second processor core is greater than that of the first processor core. When loading of a task is lower than an hmp_down_migration threshold, the multi-core processing device allocates the task to the first processor core. When the loading of a task is higher than an hmp_up_migration threshold, the multi-core processing device allocates the task to the second processor core. At least one of the hmp_down_threshold and the hmp_up_threshold changes from a first value to a second value during a run time of the multi-core processing device.

Abstract: An object node apparatus, system, and method are described. An apparatus can include a lockless-mode controller configured to communicatively couple to a plurality of storage resources and to a plurality of processor cores each preassigned to process a specific type of sub-task at a different preassigned storage resource that is configured to receive object input/output (I/O) only from the preassigned core, the lockless-mode controller being further configured to receive a plurality of object I/O messages from one or more clients, each to perform an object I/O task, divide each object I/O task into a plurality of sub-tasks, identify a specific sub-task type for each sub-task, and send each sub-task for each specific sub-task type to a processor core preassigned to process the specific sub-task type, wherein the sub-tasks include storage operations related to storing sub-object data in, or retrieving sub-object data from, the preassigned storage resource for each processor core.

Abstract: The invention provides for a method for processing a plurality of data sets (105; 106; 108; 110-113; DB1; DB2) in a data repository (104) for storing at least unstructured data, the method comprising: —providing (302) a set of agents (150-168), each agent being operable to trigger the processing of one or more of the data sets, the execution of each of said agents being automatically triggered in case one or more conditions assigned to said agent are met, at least one of the conditions relating to the existence, structure, content and/or annotations of the data set whose processing can be triggered by said agent; —executing (304) a first one of the agents; —updating (306) the annotations (115) of the first data set by the first agent; and —executing (308) a second one of the agents, said execution being triggered by the updated annotations of the first data set meeting the conditions of the second agent, thereby triggering a further updating of the annotations of the first data set.

Abstract: Methods and systems are disclosed for executing tasks in a partially out-of-order execution environment. Input is received indicating a task and task type for execution within an environment. Functions associated with the task and type of task may be selected. An instruction may be generated for each function indicating that the function is configured for static scheduling or dynamic scheduling. A schedule for instantiating each function may be generated, where functions configured for static scheduling are scheduled for instantiation according to a position of the function within the list and functions configured for dynamic scheduling are scheduled for instantiation at runtime based on an environment in which the function is instantiated and a position of the function of the subset of the set of functions within the list. A thread specification may then be generated using the functions and list. The thread specification may transmitted to remote devices.

Abstract: A data scraping system includes: a management module having a service program that calls inquiry or execution requiring scraping to receive user information for scraping and allocate user information and a task using an internal scheduling algorithm; and a scraping module connected to a plurality of different institutions, allowing communication, based on the user information and the task allocated by the management module, to scrape data corresponding to the called inquiry or execution from the plurality of institutions. The scraping module includes a meta database to designate a data item to be scraped, and is configured to extract data to scrape based on the meta database from a data set provided by the institution. According to the data scraping system, it is possible to enable data scraping in various platforms such as mobile computing devices, and collect multiple scraping information in one device without constraints of a security module.

Abstract: In various embodiments, methods and systems, for implementing proxy objects are provided. A task request directed to a driver component is intercepted at a proxy object component. The task request is an Input/Output or a session task. A proxy object is generated for the task request at the proxy object component. The task request is communicated from the proxy object component to the driver component. A modification indication to modify the driver component is received. A standby mode is initiated at the proxy object component and the driver component, the standby mode comprises processing task requests and proxy objects based on standby operations. A determination that the driver component has been modified to a modified version of the driver component is made. A recovery mode is initiated at the proxy object component and the driver component, the recovery mode comprises processing task requests and proxy objects based on recovery operations.

Abstract: An aspect includes optimizing an application workflow. The optimizing includes characterizing the application workflow by determining at least one baseline metric related to an operational control knob of an embedded system processor. The application workflow performs a real-time computational task encountered by at least one mobile embedded system of a wirelessly connected cluster of systems supported by a server system. The optimizing of the application workflow further includes performing an optimization operation on the at least one baseline metric of the application workflow while satisfying at least one runtime constraint. An annotated workflow that is the result of performing the optimization operation is output.

Abstract: In accordance with an embodiment, described herein is a system and method for use of a multi-tenant application server with a multitasking virtual machine. The multi-tenant application server enables the definition of one or more partitions that provide a runtime and administrative subdivision or slice of an application server domain; while the multitasking virtual machine enables safe, secure, and scalable multitasking of software applications. Applications can be deployed a resource group scoped either to a domain or to a particular partition, or to a resource group template at the domain level. Partitions running on a managed server instance within the domain can share the multitasking virtual machine, which provides each partition with its own subdivision or slice of that multitasking virtual machine.

Abstract: Methods, systems, and computer program products for leveraging shared work to enhance job performance across analytics platforms are provided herein. A computer-implemented method includes comparing one or more task characteristics of multiple tasks across multiple jobs to be executed within a given environment, wherein each of the multiple jobs comprises one or more tasks; identifying, based on said comparing, one or more of the multiple tasks that can be shared by two or more of the multiple jobs; scheduling the multiple jobs for execution within the given environment, wherein said scheduling is based on (i) the identified tasks that can be shared by two or more of the multiple jobs, and (ii) one or more performance metrics of the given environment; and allocating resources to the multiple jobs based on said scheduling.

Abstract: A method by a computing device to allocate computing resources for processing items in a queue using a list of computing resource allocation steps. The method includes determining a total amount of computing resources that are to be allocated for processing items in the queue based on a current number of items in the queue and the list of computing resource allocation steps, where the list of computing resource allocation steps specifies a plurality of ordered steps. Each step in the plurality of ordered steps defines a rate at which additional computing resources are allocated for processing items in the queue when the number of items in the queue is within a particular range. The method further includes adjusting an amount of computing resources allocated for processing items in the queue to match the determined total amount.

Abstract: An apparatus includes processing cores, memory blocks, a connection between each of processing core and memory block, chip selection circuit, and chip selection circuit busses between the chip selection circuit and each of the memory blocks. Each memory block includes a data port and a memory check port. The chip selection circuit is configured to enable writing data from a highest priority core through respective data ports of the memory blocks. The chip selection circuit is further configured to enable writing data from other cores through respective memory check ports of the memory blocks.

Abstract: A method of registering devices (2), in particular access control devices or payment or vending machines, on a server (3) of a system where the device (2) which is to be registered generates a public and a private key with asymmetrical cryptology. The device (2) subsequently transmits a registration request, signed with the device's private key, containing at least one device ID, the device's public key and the login details of a technician. The server (3) verifies the registration request using the device's public key and, after verification, assigns the device (2) to a ‘data reception room’ where no operational communication with the server (3) occurs. Subsequently, the device (2) is cleared for communication by a technician logged in to the server (3). The device is assigned to an area of the server memory for operationally cleared devices, and a notification of registration is transmitted to the device (2).

Abstract: A method and system to assign a task to a first thread pool in view of a determination that a previous execution time associated with the task is less than or equal to a threshold execution time. A second thread pool is designated to execute one or more tasks that have an execution time that exceeds the threshold execution time. The method and system may determine that a first execution time associated with the task exceeds the threshold execution time. The task is reassigned to the second thread pool in view of the determination that the first execution time exceeds the threshold execution time.

Abstract: A method and a system embodying the method for job pre-scheduling in a processing system comprising distributed job management, encompassing: determining a maximum amount of pre-schedulable jobs for each of a plurality of engines; setting for each of the plurality of engines a threshold less than or equal to the maximum amount; pre-scheduling by a scheduler an amount of jobs less than or equal to the threshold to at least one of a plurality of job managers; determining at the at least one of the plurality of job managers managing one of the plurality of engines one of a plurality of data processing devices in order for each pre-scheduled job; and assigning the job to the determined data processing device.

Abstract: A management resource and method monitor management operations associated with a group of managed devices, assign significance factors the management operations, and record instance information, identifying the managed device and the management operation, for each management operation performed. For each managed device, an interest factor is determined for each management operation. Weighting factors are determined for each managed device based on the interest factors and the significance factors. A management console display may be generated wherein the display indicates, whether in graphical or textual form, high priority managed devices, where the high priority devices are selected from the entire population of managed devices, based on the weighting factors. The IT administrator may adjust a weighting factor threshold to adjust the magnitude or extent of the filtering.

Abstract: Methods and systems for supporting multifunction virtualization within SR-IOV in a multi-partition virtualization system are described. One method includes instantiating a physical function of an I/O interface device within an interconnect partition of a multi-partition virtualization system implemented at least in part on the computing device, and instantiating a plurality of virtual functions within a guest partition of the multi-partition virtualization system, each of the plurality of virtual functions associated with a physical function. The method includes assigning a bus identifier, a device identifier, and a function identifier to each of the plurality of virtual functions, the plurality of virtual functions including a first virtual function associated with a first bus identifier and a first device identifier and a second virtual function associated with the first bus identifier and the first device identifier.

Abstract: Systems and methods are provided that enable a general framework for partitioning application-defined computations (e.g., jobs) in a scalable environment. The general framework decouples partitioning of a computation from the other aspects of the computation. As a result, the effort required to define an application-defined job is reduced or minimized, as the user is not required to provide a partitioning algorithm. A user can optionally take advantage of a partitioning framework by providing application-defined interfaces to perform the desired job. Optionally, a user can provide additional information to allow for modification of how partitions are assigned.

Abstract: Embodiments for computing and implementing automatic scaling operations in a distributed computing environment. A current allocation of a specific resource used by a specific application instance of an application. A high watermark is configured as a low bound for a high tier of consumption of the specific resource used by the specific application instance, and a low watermark is configured as a high bound for a low tier of consumption of the specific resource used by the specific application instance. An increase in allocation of the specific resource for the specific application instance is computed and implemented when sustained consumption of the specific resource at the high tier is identified, and a decrease in allocation of the specific resource for the specific application instance is computed and implemented when the sustained consumption of the specific resource at the low tier is identified.

Abstract: Implementations of the present disclosure include methods, systems, and computer-readable storage mediums for training ML models in aPaaS architectures including actions of receiving, at a training master provided using a first VM, a training request to train a ML model, the training request being received from a first tenant in a multi-tenant, aPaaS architecture, initiating, by the training master, training of the ML model by a training worker provided using a second VM, during training of the ML model, periodically updating training storage metadata based on metadata describing progress of the training of the ML model, and in response to completion of the training of the ML model, storing a trained ML model in a model storage of the aPaaS architecture.

Abstract: A system for controlling shared computing resources is provided including a semaphore for controlling concurrent access to one or more computing resources by one or more worker processing threads, the semaphore including a semaphore counter corresponding to vacancies of one or more computing resources and a semaphore queue for resource requests of the one or more worker processing threads waiting for the one or more computing resources. The semaphore is programmed and configured to decrement the semaphore counter when allocating one of the computing resources to one of the worker threads and configured to increment the semaphore counter when releasing one of the computing resources from one of the worker threads. The semaphore increments the queue of requests from the one or more worker threads if a request for resource access occurs when the counter of the semaphore counter is zero or less than zero.