Abstract: An information processing apparatus includes: a memory configured to store an information processing program; and a plurality of processor cores configured to acquire and execute a task from a storage region which is provided for each of the processor cores and including a first processor core configured to execute the information processing program, wherein the first processor core: performs, in work steal in which a task stored in a storage region of the first processor core is acquired by a second processor core, a writing process for an abort region, which is provided corresponding to the task, for detecting access contention by the first processor core and the second processor core using a transactional memory function; and performs a reading process for the abort region when the task is to be acquired from the storage region.

Abstract: Disclosed is a computer program stored in a computer readable storage medium including encoded commands according to an exemplary embodiment of the present disclosure. When the computer program is executed by one or more processors of a computer system, the computer program causes the one or more processors to perform the following procedures for the distributed processing. The procedures may include: analyzing queries included in a process by a processor to classify the queries into a distributed processing query and a centralized processing query; registering the distributed processing query in one or more distributed processing nodes among two or more nodes; registering the centralized processing query in a centralized processing node among two or more nodes; and receiving an event by each node.

Abstract: A distributed task system based on internet of things and a related service processing method are provided. The system can include a trigger for receiving data sent by a smart device or a user in the internet of things, a task scheduling module and a plurality of processing units. The task scheduling module can instantiate a service processing flow according to the data received by the trigger, and can sequentially schedule and start multiple processing units according to the service processing flow. Each processing units can execute a certain stage task of the service processing flow, and send an execution result of the certain stage of the service processing flow back to the task scheduling module. And the task scheduling module can notify the smart device or the user of a final execution result of one processing unit that executes a final stage task of the service processing flow.

Abstract: Cognitive management of application request processing in a multi-threading application server environment is provided by, in part, obtaining multiple threads based on receiving multiple application execution requests, and determining whether to consolidate processing of at least two threads of the multiple threads based on one or more specified characteristics related to processing of the at least two threads. Based on determining to consolidate processing, the cognitive management approach includes pausing the at least two threads and generating a consolidated thread to determine a common result for the at least two paused threads. Thereafter, execution of the at least two threads is resumed using the common result provided by processing of the consolidated thread. In one embodiment, the at least two threads may be at least two request-level threads, while in another the threads may be at least two method-level threads.

Abstract: Providing dynamic latency in an integration flow. An integration flow connecting multiple integration nodes and including at least one repeat node is managed, wherein a repeat node repeats by sending a message, a variation of a message, or an element of a message to multiple paths or to a single path multiple times. A repeat node is registered with a workload manager component for monitoring. Data relating to the effect on performance by the integration flow including the monitored repeat node is collected. Required latencies or variations of latencies between firing of repeats of the repeat node based on the collected data are determined. The latencies or variations of latencies between firing of the repeats of the repeat node are instructed to influence dynamically a flow of elements through the repeat node.

Abstract: A device may determine values for a set of metrics related to a set of event messages being processed, by a set of server devices, from a set of queues. The values for the set of metrics may be determined as the set of event messages are being processed. Each of the set of queues may be associated with a different subset of event messages and a different subset of server devices. The device may determine to scale a quantity of server devices included in the set of server devices. The quantity of server devices may be scaled to increase the quantity of server devices or to decrease the quantity of server devices. The device may provide a set of instructions to scale the quantity of server devices. The device may perform an action to facilitate accessibility of data related to processing of the set of event messages.

Abstract: Embodiments of the invention are described being used to distribute a database across a plurality of compute nodes of a parallel computing system. That is, a method is described for creating a fully in-memory database on the parallel computing system. Further, data compression techniques are described being used to increase the performance of the in-memory database by compressing some portions of the database to fit within a single node or a logically or physically related group of nodes.

Abstract: Managing data backup during workload migration is provided. A set of workloads for migration from a source environment to a target environment is identified in response to receiving a request to migrate the set of workloads. The migration of the set of workloads is initiated from the source environment to the target environment along with migration of backup data corresponding to the set of workloads. A backup configuration transformation from a backup configuration corresponding to the source environment to a set of backup configurations corresponding to the target environment is determined based on semantic matching between characteristics of the backup configuration corresponding to the source environment and characteristics of the set of backup configurations corresponding to the target environment, a state of the source environment, backup configuration transformation actions, and a goal state of the target environment.

Abstract: An abnormality detection apparatus (2000) handles tasks allocated to a plurality of processing servers (3200) as processing targets in a distribution system (3000) having the processing servers (3200). A history acquisition unit (2020) acquires progress history information which is information regarding progress of the plurality of tasks at a plurality of time point of recording. A target range determination unit (2040) determines a target range. A distribution calculation unit (2060) calculates a task speed distribution which is a probability distribution of processing speeds of the tasks using the progress history information regarding the plurality of tasks. An abnormality determination unit (2080) compares a processing speed of a task to be determined with the task speed distribution to thereby determine whether or not the processing speed of the task to be determined is abnormal.

Abstract: Provided are methods and systems for load balancing client requests between sites associated with a domain name. A method comprises determining a first active response delay time between a Domain Name System server and a first site. The method further comprises determining a first application response delay time between the first site and one or more first servers associated with the first site. According to the method, the first active response delay time and the first application response delay time are compounded to produce a first compounded response delay time. The method further comprises determining a second active response delay time and a second application response delay for a second site to produce a second compounded response delay time. The first compounded response delay time and the second compounded response delay time are compared to perform load balancing between the first site and the second site.

Abstract: Various embodiments for managing migration between server systems are presented. Indications of a plurality of migration scripts to be executed during a migration of a database system from a source server system to a target server system may be received. An execution order for the plurality of migration scripts may be determined based upon characteristics of a plurality of database objects to be migrated using the plurality of migration scripts and characteristics of the plurality of migration scripts. Execution of a first migration script of the plurality of migration scripts may be initiated according to the determined execution order. The first migration script may cause copying of a database object of the plurality of database objects from the source server system to the target server system.

Abstract: A method and an apparatus for designing a cloud based radio access network, C-RAN, are provided. The solution comprises dividing (302) the access network into two or more tiers having different processing capabilities; receiving as an input information about the physical location of one or more remote radio heads, RRH, of the network and required quality of service of the remote radio heads; and determining (306) the physical location of the tier, BBU, having most efficient processing capability for serving the remote radio heads on the basis of the information about the location and required quality of service of at least part of the one or more remote radio heads.

Abstract: Disclosed aspects relate to query plan management associated with a shared pool of configurable computing resources. A query, which relates to a set of data located on the shared pool of configurable computing resources, is detected. A virtual machine includes the set of data. With respect to the virtual machine, a set of burden values of performing a set of asset actions is determined. Based on the set of burden values, a query plan to access the set of data is established. Using at least one asset action of the set of asset actions, the query plan is processed.

Abstract: Embodiments directed toward a method, and a system. A method can be directed toward deciding whether to accept a virtual machine migration on a host. The method can include receiving a request to host a migrated virtual machine. The method can include determining if the host is overcommitted. The method can include selecting a low priority virtual machine in response to the host being overcommitted. The method can include determining if a resource for the low priority virtual machine can be reduced to accommodate the migrated virtual machine. The method can include reducing the resources for the low priority virtual machine in response to the resources being able to be reduced. The method can include accepting the request to host the migrated virtual machine in response to the reduction of the resources.

Abstract: Virtual machines (VMs) and a social networking graph where relationships between VM nodes are based on communications between VM nodes. VMs using social networking graphs to help manage a set of VMs. A system where a VM uses a social networking graph to query other VMs to query regarding “health” issues, problem solutions and/or performance improvements.

Abstract: Provided is a process having steps including obtaining a specification of a data analysis to be performed in parallel on a computing cluster; parsing the specification of the data analysis; determining which data is implicated in each portion of the data analysis to be assigned to a plurality of computing nodes of the computing cluster; determining that a portion of the implicated data is not already present in memory of at least some of the plurality of computing nodes of the computing cluster; distributing the portion of the implicated data according to an index that positions related values of the data on the same computing nodes of the computing cluster; determining which computing nodes of the computing cluster have data relevant to which rules in the data analysis and send relevant rules to the corresponding computing nodes; executing the rules on the computing nodes; and aggregating results of executing the rules.

Abstract: An e-mail system identifies actions or activities that a user takes relative to an e-mail message. The activity is logged as an activity log entry in an activity log. When the user accesses the activity log, the activities and corresponding e-mail messages are accessible to the user through the activity log entry, along with a user interface element that allows the user to take additional actions with respect to the e-mail message in the activity log entry.

Abstract: A computer implemented method, system, and computer program product comprising computing at a cluster manager and node manager a realized goal state by examining a goal state stack for a set of nodes of a cluster, determining differences between the realized goal state and the actual state of the cluster and nodes, and based on the differences, determining a set of actions to align the actual state with the realized goal state.

Abstract: A system for collecting and analyzing metadata from peer-to-peer (P2P) media networks. The present invention gathers metadata about P2P media from the P2P network directly as well as relevant 3rd party social networks where that media is discussed. Users access the system via any Internet enabled device. The users' media streaming behavior is collected so that it can determine what they like. That data is indexed for future analysis. The system also monitors social network dialog keyed to the media (e.g. titles, actors, singers). The system then can identify related media on P2P networks and source it so that it is immediately available. The social data collection agents then extract data about the media that users of social networks are providing for the computing of demand for the media and/or performers.

Abstract: A method for a live migration in a multitenant clustered database aimed at providing efficient migration of transactional online databases with no service interruption or failed requests and minimal impact on quality of service. The migration is facilitated by a client connection pool, which migrates connections from the source node to the target node using a self-adapting algorithm to control the migration rate. The algorithm calculates a base migration rate according to the relative number of connections transferred to the target node. Once base migration rate is calculated, the algorithm accelerates or throttles the base migration rate based on workload behavior and system load, attempting to minimize impact on the migrated tenant, while keeping overall migration time low.

Abstract: Methods and systems of managing a resource in a distributed resource management system can include: monitoring, by at least one processor, utilization of the resource being used by at least one workload; and performing an enforcement action on a particular workload of the at least one workload when a utilization condition is met, and when the particular workload has a current resource utilization exceeding its associated resource allocation limit.

Abstract: A method using L processers includes: receiving image data encoded by delta row encoding; dividing the two dimensionally arranged plurality of pixels by M (M?L) to make M blocks of pixels; assigning M processers to perform a parallel processing of accumulating the delta data for all the row lines of each of the M blocks of the image data to obtain the accumulated delta data, the accumulated delta data including a total delta between the first row line and the last row line in each of the blocks; obtaining the first row lines in the respective blocks using the total delta one by one starting from the second block; and assigning M+1 processers to perform a parallel processing of decoding using the obtained first row lines to obtain the decoded data and a parallel processing of rendering the image data using the decoded data.

Abstract: Exemplary methods, apparatuses, and systems include a client virtual machine processing a system call for a device driver to instruct a physical device to perform a function and transmitting the system call to an appliance virtual machine to execute the system call. The client virtual machine determines, in response to the system call, that an established connection with the appliance virtual machine has switched from a first protocol to a second protocol, the first and second protocols including a high-performance transmission protocol and Transmission Control Protocol and Internet Protocol (TCP/IP). The client virtual machine transmits the system call to the appliance virtual machine according to the second protocol. For example, the established connection may switch to the second protocol in response to the client virtual machine migrating to the first host device from a second host device.

Abstract: In one embodiment of the present invention, one or more computer processors create update commands that describe a state change of a resource and include the update commands in a change event for the resource in a change log. Embodiments of the present invention can obviate the need for an indexer to retrieve all triples of a changed resource via HTTP GET requests, and obviate the need for the indexer to replace all triples of a prior version of a changed resource with newly obtained triples.

Abstract: Managing workload in a computing environment, in one aspect, may include monitoring an execution behavior of a portion of a job executing on the external computing resources. The execution behavior of the portion of the job executing on the external computing resources may be compared to the behavior required to attain at least one performance objective. If the comparison indicates that a performance objective is not being met, whether allocating additional internal resources would improve the overall performance as compared to the objective may be determined. If the determination indicates that allocating additional internal resources would improve the overall performance, at least one additional resource may be allocated.

Abstract: Methods and apparatuses are described for real-time monitoring of computer system processor and transaction performance during an ongoing performance test. A plurality of monitoring agent processes executing on a processor of a first computing device in a computer system testing environment extracts performance data relating to the ongoing performance test in the computer system testing environment by scraping CPU usage data, analyzing submitted transaction data as stored in a transaction log file and executing search queries against a transaction database in the computing system testing environment to generate a transaction lifecycle dataset.

Abstract: Systems, methods and media are shown for equitable job processing of asynchronous messages for multiple tenants in a multi-tenant platform that involve receiving messages of a given message type in an input buffer from at least one front end tier (FET) device, providing messages of the given message type from an output buffer to at least one back end tier (BET) device, determining a priority weight W for each message in the input buffer based on a tenant identifier and message type for the message, and evaluating the priority weight W for the message and delaying the message if the priority weight W is below a priority threshold and moving the message to the output buffer if the priority weight W is not below the priority threshold.

Abstract: A comprehensive approach to streaming backups for virtual machines (“VMs”) in a storage management system comprises improvements to the assignment of data agent proxies for VM secondary copy operations. New considerations in performing a VM streaming backup job include without limitation: determining and enforcing a system-wide per-proxy limit of concurrent data streams; generating an ordered priority list of the VMs to be backed up as a basis for choosing which proxies will back up the respective VM, though the illustrative system may not strictly adhere to the priority list based on further considerations; identifying a next available proxy based on data stream utilization at the proxy; and dynamically re-generating the priority list and re-evaluating considerations if some VMs become “stranded” due to a failure to be backed up. Secondary copy operations are distributed to proxies in ways that improve the chances of successfully completing VM streaming backups.

Abstract: Described herein is an end-to-end solution (i.e., capable of constructing a content delivery network all the way from an origin server to end user devices) which provides a managerial overview to what has been a very disjoint build-out of networks. In the past, networks have evolved to serve particular needs, as opposed to being architected from a strategic view. What this does is to allow someone (e.g., content provider, network operator, etc.) to impose a strategy over those networks. Rather than having to construct solutions that deliver a particular kind of content, a particular security requirement, a particular latency requirement, etc., one can simply provision particular types of nodes as needed when needed.

Abstract: A content distribution service may provide lead information to content providers or other entities associated with a broadcast. The service may include both registered and unregistered (e.g., anonymous) users or subscribers that may each provide data associated with one or more subscriber attributes. Subscriber attribute information may be collected from a subscriber in a progressive manner such that sequentially small portions of data are collected over a period of time. The content distribution service may generate leads by collecting and determining subscriber attributes for each subscriber or anonymous user associated with the service. The leads may be generated in response to a request by a content provider or other requestor, or in response to an event such as the creation of a new broadcast.

Abstract: A method, apparatus, and machine readable storage medium is disclosed for balancing loads among a plurality of virtual machines (VMs) from a central dispatcher, wherein the dispatcher receives data packets and maps the data packets to VMs selected from the plurality of VMs, using a weighted hash function, having an associated weighting for each VM and forwarding each packet to a VM accordingly, wherein a load balancer decrements a weighting for a VM, responsive to an indication of the load on the VM exceeding a first load threshold. Weightings can correspond to a number of bins associated with each VM. Weightings are adjusted in response to receiving invite and disinvite messages from the VMs, representing their respective loads.

Abstract: A machine learning service implements programmatic interfaces for a variety of operations on several entity types, such as data sources, statistics, feature processing recipes, models, and aliases. A first request to perform an operation on an instance of a particular entity type is received, and a first job corresponding to the requested operation is inserted in a job queue. Prior to the completion of the first job, a second request to perform another operation is received, where the second operation depends on a result of the operation represented by the first job. A second job, indicating a dependency on the first job, is stored in the job queue. The second job is initiated when the first job completes.

Abstract: A data mining method and a node are provided. The method includes: obtaining predicted execution time of each computing subnode in a current round of iteration of a task, and allocating a corresponding volume of task data to the computing subnode; after the current round of iteration is executed, collecting execution status information of each computing subnode in the current round of iteration, and accordingly determining whether the task data volume of each computing subnode needs to be adjusted in a next round of iteration; and performing the next round of iteration according to the adjusted task data volume. Therefore, a corresponding volume of task data can be allocated according to a capability of each computing subnode. In this way, some unnecessary load balancing processes can be avoided, network consumption can be reduced, and data mining performance of a system can be improved.

Abstract: A system that provides an improved way of scheduling a plurality of imaging tasks in a medical imaging system is disclosed. The system enables a user (i.e. a technician or medical expert) to group different imaging tasks and executes them simultaneously to significantly reduce scan time. These imaging tasks are of different types or are related to different imaging techniques. If different imaging tasks need to be performed substantially at the same location and have substantially same scan time then they can be scheduled simultaneously so that overall scan time can be reduced.

Abstract: A parallel-computing system includes: computation nodes connected in a mesh or torus connection and configured to perform parallel-computation; and a job management node including a computer, being coupled to the computation nodes, and being configured to: manage a plurality of jobs, select a first job from the jobs being executed by any of the computation nodes, search for a second area adjoining a first area occupied by a first computation node of the computation nodes executing the first job, the second area being occupied by a free computation node not executing any job, determine a fourth area of a same size as the first area at a location contacting an end in a predetermined direction in a third area comprising the first area and the second area, determine a fifth area of a predetermined shape in a range not overlapping the fourth area in the third area.

Abstract: Methods, systems, and computer readable mediums for optimizing a system configuration are disclosed. In some examples, a method includes determining whether a system configuration for executing a workload using a distributed computer system is optimizable and in response to determining that the system configuration is optimizable, modifying the system configuration such that at least one storage resource for storing workload data is located at a server node that is executing the workload in the distributed computer system.

Abstract: This disclosure sets forth systems and methods for recommending candidate computing platforms for migration of data and data-related workload from an original computing platform. The systems and methods further describe determining recommendations of candidate computing platforms based on a comparison of key performance and utilization statistics of the original computing platform under a user-generated workload with candidate computing platforms under a synthetic workload. Key performance and utilization statistics may relate to CPU, memory, file I/O, network I/O, and database I/O operations on the respective computing platforms. The synthetic workload may be defined by parameters that simulate the key performance and utilization statistics of the original computing platform under the user-generated workload. Further, the synthetic workloads may be executed on individual candidate computing platforms to determine service level capabilities that are ultimately used to form the recommendation.

Abstract: One aspect is a method that includes analyzing, by a processor of an analysis system, an instruction set architecture of a targeted complex-instruction set computer (CISC) processor to generate an instruction set profile for each CISC architectural instruction variant of the instruction set architecture. A combination of instruction sequences for the targeted CISC processor is determined from the instruction set profile that corresponds to a desired stressmark type. The desired stressmark type defines a metric representative of functionality of interest of the targeted CISC processor. Performance of the targeted CISC processor is monitored with respect to the desired stressmark type while executing each of the instruction sequences. The targeted CISC processor is stress tested based on executing at least one of the instruction sequences identified as most closely aligning with the desired stressmark type.

Abstract: Capacity of a database system and/or a computing system that includes a database can be effectively changed from a current computing capacity to another computing capacity. This can be achieved by causing usage capacity of at least one of resource to be changed when the database is active. By way of example, capacity of a database system can be regulated by delaying the execution of the database work based on a target capacity. As a result, database work can take relatively longer to complete when the capacity of a database is effectively regulated to be below its full capacity. In effect, a portion of available capacity (excess capacity) can be made inaccessible to the database. However, excess capacity can be used to manage various services of the database in accordance with one or more service criteria (e.g., Service Level Goals or Agreements).

Abstract: Example data management systems and methods are described. In one implementation, a method identifies multiple files to process based on a received query and identifies multiple execution nodes available to process the multiple files. The method initially creates multiple scansets, each including a portion of the multiple files, and assigns each scanset to one of the execution nodes based on a file assignment model. The multiple scansets are processed by the multiple execution nodes. If the method determines that a particular execution node has finished processing all files in its assigned scanset, an unprocessed file is reassigned from another execution node to the particular execution node.

Abstract: In various embodiments, techniques for improving function of in-memory processing systems include identifying data stored in a distributed filesystem, including data to be processed by an in-memory processing application and data not to be processed by the in-memory processing application; identifying one or more partitions of the data corresponding to the data to be processed; and selectively transferring the one or more partitions from the distributed filesystem to a memory of the in-memory processing application. Techniques may also include determining an average remote execution time of the in-memory processing system; tracking a waiting time of a job pending processing on a local node of the in-memory processing system; comparing the tracked waiting time of the job pending processing; and demoting a locality of the job if waiting time of the job exceeds the average remote execution time of the in-memory processing system. Corresponding systems and computer-program products are also disclosed.

Abstract: A method includes determining a capacity model that configures computing resource capacity for a capacity container. The computing resource capacity includes a first storage attribute for an amount of storage in a storage component. A load model is determined that configures load for the capacity container. The load includes a second storage attribute for a storage requirement for a virtual machine. A profile of a virtual machine unit is determined for estimating available capacity in a capacity container. The profile is determined using virtual machine attributes for a set of virtual machines, wherein the virtual machine unit includes a storage requirement based on storage requirements for the set of virtual machines. The profile of the virtual machine unit is fit into available capacity. A number of virtual machine units is determined based on the fitting, the number of virtual machine units being a measure of available capacity.

Abstract: Provided are methods and systems for implementing a distributed database in a data network. The method comprises receiving node data associated with one or more nodes of a plurality of nodes, updating the distributed database and replicating the distributed database to each of the plurality of nodes. The plurality of nodes comprises one or more cluster device, a cluster master, a traffic classification engine, a service node, and an orchestrator. The node data comprises node health, a number of total connections, node processing unit utilization, node memory status, destination server address, destination server capacity, destination server network connectivity, node dynamic state, node responsiveness. The distributed database comprises tables containing traffic map, node health information, traffic classification mapping, and service policy.

Abstract: The disclosure herein describes a system for providing distributed global server load balancing (GSLB) over resources across multiple data centers. The system includes a directory group comprising one or more directory nodes and a plurality of GSLB nodes registered to the directory group. A respective GSLB node is configured to provide GSLB services over a respective portion of the resources. A directory node includes a domain name system (DNS) query-receiving module configured to receive a DNS query from a client, a node-selecting module configured to select from the plurality of GSLB nodes a first GSLB node based at least on the DNS query, and a DNS query-responding module configured to respond to the DNS query to the client using an address of the selected first GSLB node, thereby facilitating the selected first GSLB node in performing GSLB while resolving the DNS query.

Abstract: A streams manager monitors performance of a streaming application and determines if operators are underperforming according to a threshold. When the performance needs to be improved, the streams manager automatically modifies the flow graph to offload or reroute a stream of data, or part of a stream of data, to a similar operator to more efficiently utilize streaming resources. Operators are provided with multiple ports to allow the streams manager to send additional streams to the operator.

Abstract: Analytics and visualization for logical dependency mapping of computer servers may be provided. Data collected from computer servers is received and analyzed to discover logical dependencies among the computer servers, for example, to discover communications between one or more of operating systems, middleware and applications associated with the computer servers. The computer servers may be grouped into one or more dependency groups based at least on a user-defined grouping criterion. The one or more dependency groups and their logical dependencies may be output.

Abstract: Technology for detecting processor usage of a physical host hosting computing instances is provided. In one example, a method may include identifying a location of a hardware register of a physical host containing data in a model specific register. The physical host hosts a plurality of computing instances using a hypervisor. The method may include determining power consumption by a processor of the physical host for the computing instances from the data. An upper level of acceptable processor power consumption may be identified. Processor usage may be determined based on the power consumption determined and the upper level of acceptable processor power consumption. The processor usage may be provided as data through the hypervisor.

Abstract: A streams manager monitors performance of a streaming application and determines if operators are underperforming according to a threshold. When the performance needs to be improved, the streams manager automatically modifies the flow graph to offload or reroute a stream of data, or part of a stream of data, to a similar operator to more efficiently utilize streaming resources. Operators are provided with multiple ports to allow the streams manager to send additional streams to the operator.

Abstract: A batch scheduler receives a task. Further, a processor, at the batch scheduler, generates a virtual machine corresponding to the task. In addition, the virtual machine is deployed from the batch scheduler to a compute node. The task is also dispatched from the batch scheduler to the virtual machine at the compute node.

Abstract: In a system in which documents are generated dynamically in response to user requests, historical data is collected regarding data retrieval subtasks, such as service requests, that are performed to generate such documents. This data is used to predict the specific subtasks that will be performed to respond to specific document requests, such that these subtasks may be initiated preemptively at or near the outset of the associated document generation task. In one embodiment, the historical data is included within, or is used to generate, a mapping table that maps document generation tasks (which may correspond to specific URLs) to the data retrieval subtasks that are frequently performed within such tasks.