Abstract: Techniques are described for eliminating backpressure in a distributed system by changing the rate data flows through a processing element. Backpressure occurs when data throughput in a processing element begins to decrease, for example, if new processing elements are added to the operating chart or if the distributed system is required to process more data. Indicators of backpressure (current or future) may be monitored. Once current backpressure or potential backpressure is identified, the operator graph or data rates may be altered to alleviate the backpressure. For example, a processing element may reduce the data rates it sends to processing elements that are downstream in the operator graph, or processing elements and/or data paths may be eliminated. In one embodiment, processing elements and associate data paths may be prioritized so that more important execution paths are maintained.

Abstract: Computing platforms may enable clients to use services to perform data processing jobs. A given job may utilize several services in a service call pattern. Each of these services consumes resources during execution. Described herein are systems, devices and methods for determining costs associated with performance of a service call pattern. Weight data may be generated for each service using various metrics. As service calls are made in the service call pattern, an accumulated weight value is generated which is indicative of the weight data of the services in the service call pattern. The accumulated weight data may represent metrics such as amounts of resources consumed, latency, and transactions per second.

Abstract: The present disclosure is directed to fast approximate conflict detection. A device may comprise, for example, a memory, a processor and a fast conflict detection module (FCDM) to cause the processor to perform fast conflict detection. The FCDM may cause the processor to read a first and second vector from memory, and to then generate summaries based on the first and second vectors. The summaries may be, for example, shortened versions of write and read addresses in the first and second vectors. The FCDM may then cause the processor to distribute the summaries into first and second summary vectors, and may then determine potential conflicts between the first and second vectors by comparing the first and second summary vectors. The summaries may be distributed into the first and second summary vectors in a manner allowing all of the summaries to be compared to each other in one vector comparison transaction.

Abstract: An apparatus comprises at least one processing platform implemented using at least one processing device comprising a processor coupled to a memory. The processing platform comprises virtualization infrastructure, an assurance layer and an analytic layer. The assurance and analytic layers are configured to provide data collection and event management functionality to support automation relating to resources of the virtualization infrastructure and associated workloads. By way of example, the assurance and analytic layers illustratively comprise respective deterministic and indeterministic functional groupings of components. The functional groupings of components of the assurance and analytic layers are utilized to implement closed-loop remediation workflows and other types of automation relating to the virtualization infrastructure resources and their associated workloads.

Abstract: The workload profiler and performance interference (WPPI) system uses a test suite of recognized workloads, a resource estimation profiler and influence matrix to characterize un-profiled workloads, and affiliation rules to identify optimal and sub-optimal workload assignments to achieve consumer Quality of Service (QoS) guarantees and/or provider revenue goals. The WPPI system uses a performance interference model to forecast the performance impact to workloads of various consolidation schemes (e.g., consolidation strategies) usable to achieve cloud provider and/or cloud consumer goals, and uses the test suite of recognized workloads, the resource estimation profiler and influence matrix, affiliation rules, and performance interference model to perform off-line modeling to determine the initial assignment selections and consolidation strategy to use to deploy the workloads.

Abstract: A scheduling method executed by a scheduler that manages multiple processors, includes detecting based on an application information table when a first application is started up, a processor that executes a second application that is not executed concurrently with the first application; and assigning the first application to the processor.

Abstract: A parallel computer system includes a plurality of calculation nodes and a job management apparatus to allocate jobs to the plurality of calculation nodes. The job management apparatus includes a controller to execute a process including searching a vacant resource generated as a result of job allocation, the vacant resource including one or more of the plurality of calculation nodes each having a vacant state in a range between current time and a scheduled start time point of a certain allocated job, retrieving the job allocatable to the vacant resource, from the unallocated jobs when the vacant resource is found, and allocating the retrieved job to the vacant resource.

Abstract: A data model for application to a constraint programming solver is generated. The data model includes a set of data model elements. A particular data model element corresponds to a particular request. The particular data model element also corresponds to one or more resources that may be assigned to the request. The data model also includes a set of constraints. One or more sort/search algorithms may be applied with the data model to the constraint programming solver. The sort/search algorithms may direct the constraint programming solver to output certain preferred assignments of resources to requests.

Abstract: A method, system, and computer program product for of configuring cloud computing platforms. One such method serves for recommending alternative computing architectures for a selected application using automated instrumentation of the application under an abstracted workload. The method commences by measuring workload characteristics of the selected application using pre-determined performance parameters. Additional performance parameters to be measured are selected based on previous measurements, and further analysis includes instrumenting the application to provide measurement instruments corresponding to the respective selected additional performance parameters. Such hardware- or software-based instruments are used for determining a baseline set of performance metrics by measuring the performance of the selected application on a first (e.g., currently-in-use) computing architecture, then, measuring the performance of the application on other proposed computing architectures.

Abstract: Virtual machine resources may be monitored for optimal allocation. One example method may include monitoring a virtual machine operating in a network to determine whether at least one predefined service tier threshold has been exceeded for a predefined amount of time, initiating a query to determine current performance threshold data of the at least one predefined service tier threshold from a database, determining at least one component state of at least one component of the virtual machine based on the at least one service tier threshold assigned to the at least one component, and reallocating the resource provided by the virtual machine when the component state indicates a high warning state.

Abstract: Methods and systems for using a scheduler in a data pipeline are disclosed. A plurality of objects in a first layer are created, each representing a respective regularly scheduled task. A plurality of objects in a second layer are created, each representing a respective scheduled instance of a regularly scheduled task. It is determined whether each object in the second layer is ready to execute. For at least one object in the second layer, it is determined if the object has received notifications from any objects on which it depends. For each object that is ready to execute, the regularly scheduled task associated with the object is performed. For each object that is not ready to execute, the object is put to sleep.

Abstract: A system and method for distributed computing, including executing a job of distributed computing on compute nodes. The speed of parallel tasks of the job executing on the compute nodes are adjusted to increase performance of the job or to lower power consumption of the job, or both, wherein the adjusting is based on imbalances of respective speeds of the parallel tasks.

Abstract: A system and methods of communicating events includes detecting an event at a first embedded application, the first embedded application being embedded in an application; triggering the detected event on an event aggregator of the application; determining, by the application, whether a second embedded application is embedded in the application; and if a second embedded application is determined to be embedded in the application, transmitting the detected event from the application to the second embedded application.

Abstract: A system and methods of communicating events includes detecting, on a code space of an application, an event at the application; transmitting the detected event from the application to an embedded application, the embedded application being embedded in the application; and triggering the detected event on an event aggregator of the embedded application based upon data associated with the detected event.

Abstract: Embodiments of the invention provide a programming model for CPU-GPU platforms. In particular, embodiments of the invention provide a uniform programming model for both integrated and discrete devices. The model also works uniformly for multiple GPU cards and hybrid GPU systems (discrete and integrated). This allows software vendors to write a single application stack and target it to all the different platforms. Additionally, embodiments of the invention provide a shared memory model between the CPU and GPU. Instead of sharing the entire virtual address space, only a part of the virtual address space needs to be shared. This allows efficient implementation in both discrete and integrated settings.

Abstract: Embodiments of the present invention provide a method, system and computer program product for single program code message retrieval for message queues. In an embodiment of the invention, a message queue data processing system can be configured for single program code message retrieval for message queues. The system can include a message queue executing in a host server and providing an API to applications communicatively coupled to the message queue over a computer communications network. The API exposed by the message queue can include a single program call including program code enabled to open a queuing resource in the message queue, to retrieve all messages in a message buffer from the queuing resource and to close the queuing resource.

Abstract: A system and method for enabling the interchange of enterprise data through an open platform is disclosed. This open platform can be based on a standardized interface that enables parties to easily connect to and use the network. Services operating as senders, recipients, and in-transit parties can therefore leverage a framework that overlays a public network.

Abstract: A method of protecting a computer against malware infection. The method includes during operation of the computer, reading master boot record code from a removable storage device into the computer and inspecting said code to identify any instructions associated with suspicious behaviour. In the event that suspicious instructions are identified, the master boot record code on the removable storage device is modified and/or the behaviour of the computer adapted in order to prevent said master boot record code installing malware into the computer. Examples of suspicious behaviour include hard disk read or write operations.

Abstract: A system and method for local survivability in a distributed contact center environment has a first processor in a first contact center node receiving a first request for interaction. The first processor transmits a first message to a second contact center node in response to the request for interaction. The first message is configured to invoke a first resource associated with the second contact center node for handling the interaction via the first resource. The first processor monitors connection with the second contact center node. The first processor receives a second request for interaction, and further determines lack of connection with the second contact center node. In response to determining lack of connection with the second contact center node, the first processor refrains from transmitting a second message to the second contact center node.

Abstract: Systems, methods, and computer program products to perform an operation comprising receiving, by a management console managing a server, a request to perform an operation on the server, determining that performance of the operation is associated with a deterministic failure in the server, and blocking performance of the operation by the management console, thereby preventing an occurrence of the deterministic failure in the server.

Abstract: There are provided a data preprocessing device and a method thereof. A data preprocessing device according to an embodiment of the present disclosure includes a risk level analyzing unit configured to calculate a failure risk level of a target system using failure history information of the target system and characteristic information of the target system; a collection period setting unit configured to determine a data collection period from the target system according to the calculated failure risk level; and a preprocessing unit configured to preprocess data collected from the target system according to the data collection period.

Abstract: Provided is a method of detecting a warning sign indicating the occurrence of a system failure targeting a plurality of varied information processing systems. A system for monitoring a plurality of monitored systems receives measured values regarding a plurality of indexes, designates, from a plurality of prediction models, a prediction model for predicting a future measured value from the measured value that was received regarding a reference index, predicts a predicted value of the reference index based on the designated prediction model, generates or updates a Bayesian network targeting the reference index and a target index, and calculates a probability that a measured value of the target index will become a predetermined value or fall within a predetermined value range.

Abstract: A method includes receiving a signal indicative of a value of a data point associated with a compute device at a first time. The value of the data point includes a data point category that is correlated with a fault category. The compute device is operatively coupled to a record module having a protected mode and an unprotected mode. A signal is received indicative of a value of the data point at a second time, after the first time. When a characteristic of a change in the value of the data point at the first time to value of the data point at the second time crosses a threshold in a first direction, a signal is sent to the record module indicative of an instruction to record data associated with the compute device in the protected mode to define a protected data set.

Abstract: An enterprise level data element review system including a data access event collection subsystem operative to collect data access event notifications relating to ones of a multiplicity of data elements, a data element metadata modification subassembly receiving an output from the data access event collection subsystem and providing a script indicating which data elements have had a metadata modification over a given period of time, and a data element dancer operative to collect at least one of metadata and access permissions for a plurality of data elements which is substantially less than the multiplicity of data elements and is selected on the basis of the script.

Abstract: Methods and systems for modular system awareness in virtualized information handling systems (IHSs) include generating, by a chassis management controller (CMC), dependency information indicating which ones of modular IHSs included in a chassis are dependent on individual shared components in the chassis. When a fault occurs at one of the shared components, the CMC may determine, using the dependency information, which particular ones of the modular IHSs are operationally dependent on the fault. The CMC may send information indicative of the fault and of the dependent modular IHSs to a virtual machine manager, which may generate a ranking of virtual machines. The virtual machine manager may use the rankings to make migration decisions for virtual machines among modular IHSs.

Abstract: A method, computer program product, and system that discards unneeded elements when generating a fault tree of an object to be analyzed. Configuration information identifies a plurality of functional blocks comprised by the object and a plurality of signal lines that connect the functional blocks in logical relationships. Exclusion target information identifies a signal line that may be excluded from the plurality of signal lines without loss of information or a functional block that may be excluded from the plurality of functional blocks without loss of information. Exclusion of a block or signal line may be determined by detecting a redundant functional block or by detecting a circular signal path traversed by two or more signal lines. The generated fault tree omits the excluded block or signal line and identifies the existence of a redundant block or of a circular signal path.

Abstract: An apparatus for comparing strings comprises a first and a second set of input registers, a matrix of comparator circuits wherein rows of the matrix correspond to the first set of input registers and columns of the matrix correspond to the second set of input registers, a set of row control registers wherein each register corresponds to a row of the matrix of the comparator circuits, the matrix of comparator circuits configured to compare data provided by the first set of registers with data provided by the second set of registers according to the row control registers, and an error detection circuit configured to compare results from four comparator circuits corresponding to two adjacent rows and two columns of the matrix of comparator circuits. A method that invokes the above mentioned apparatus is also disclosed herein, along with a computer program product corresponding to the method.

Abstract: Methods and systems for network devices is provided. In one aspect, a network device includes a plurality of ports, where the plurality of ports are configured to operate in a first operating mode as a single port at a first speed and in a second operating mode where each of the plurality of ports operate as an independent port at a second operating speed; a shared memory device for staging information received from a network for the plurality of ports operating in the first operating mode and the second operating mode; a receive port selector that selects information from the shared memory device when the plurality of ports are operating in the second operating mode; and a shared error correction code module for decoding and performing error correction on information received via the network for the first operating mode and the second operating mode.

Abstract: A method of operating a memory device includes: generating an internal read command in response to a received masked write command, the internal read command being generated one of (i) during a write latency associated with the received masked write command, (ii) after receipt of a first bit of masked write data among a plurality of bits of masked write data, and (iii) in synchronization with a rising or falling edge of a clock signal received with an address signal corresponding to the masked write command; reading, in response to the internal read command, a plurality of bits of data stored in a plurality of memory cells, the plurality of memory cells corresponding to the address signal; and storing, in response to an internal write command, the plurality of bits of masked write data in the plurality of memory cells.

Abstract: Provided are a method, system, and apparatus using reliability information from multiple storage units and a parity storage unit to recover data for a failed one of the storage units. A decoding operation of the codeword is performed in each of the storage units comprising the data storage units other than the target data storage unit and the parity storage unit to produce reliability information. In response to the decoding operation failing for at least one additional failed storage unit comprising the data and/or parity storage units other than the target data storage unit that failed to decode, reliability information is obtained for the data portion of the at least one additional failed storage unit. The reliability information obtained from the storage units other than the target data storage unit is used to produce corrected data for the data unit in the target data storage unit.

Abstract: A system and method for efficiently distributing data among multiple storage devices. A data storage array receives read and write requests from multiple client computers. The data storage array includes multiple storage devices, each with multiple allocation units (AUs). A storage controller within the data storage array determines a RAID layout for use in storing data. In response to determining a failure of a first AU, the storage controller begins reconstructing in a second AU the data stored in the first AU. For read and write requests targeting data in the first AU, the request is serviced by the first AU responsive to determining no error occurs when accessing the first AU.

Abstract: In one embodiment, a method includes iteratively selecting an original, unmoved stride from an original array until all original strides have been moved, determining a target stride location in a distributed array, determining a state of the target stride from the following states: an old state indicating unmoved data, a blank state indicating no data, and a new state indicating migrated data, determining that all target stripes are blank, moving data from the original stride to the target stripes when all the target stripes are blank, delaying the moving of the data from the original stride to the target stripes in the target stride when any of the target stripes of the target stride are in the old state and waiting until all the target stripes of the target stride are in the blank state, and determining that all original strides from the original array have been moved.

Abstract: In one aspect, the invention relates to a fault tolerant computing system. The system includes a primary virtual machine and a secondary virtual machine, wherein the primary and secondary virtual machines are in communication, wherein the primary virtual machine comprises a first checkpointing engine and a first network interface, wherein the secondary virtual machine comprises a second network interface, wherein the first checkpointing engine forwards a page of memory of the primary virtual machine to the second virtual machine such that the first checkpointing engine can checkpoint the page of memory without pausing the primary virtual machine.

Abstract: A processor includes a storage configured to receive a snapshot of a state of the processor prior to performing a set of computations in an approximating manner. The processor also includes an indicator that indicates an amount of error accumulated while the set of computations is performed in the approximating manner. When the processor detects that the amount of error accumulated has exceeded an error bound, the processor is configured to restore the state of the processor to the snapshot from the storage.

Abstract: A mobile terminal and controlling method thereof are disclosed, by which restoration and backup operations are performed for the mobile terminal. For example, the mobile terminal includes a controller performing a backup operation and a restoration operation on the mobile terminal, the controller including an application module for activating a backup restoration application and at least one or more applications; and a Linux kernel module for managing a basic hardware function of the mobile terminal, the Linux kernel module including a backup restoration service unit having an access authority for accessing a file system of each of the at least one or more applications, wherein the backup restoration application accesses the file system of each of the at least one or more applications and then performs the backup and restoration operations on data of each of the at least one or more applications.

Abstract: A method and system for recovering a corrupt virtual disk is discussed. A request to recover the disk may be received. A recovery snapshot for the last point-in-time the system was stable may be synthesized. A difference may be identified between that recovery snapshot and the corrupt disk. A virtual machine may communicate with both the difference and the corrupt disk, and the difference may be merged with the corrupt disk.

Abstract: Disclosed is a system and method for restoring modified data. An example method includes intercepting, by an activity tracking module, a request from a program to modify data; determining, by an analysis module, parameters of the intercepted request; generating, by the analysis module, a request to generate a backup copy of the data based on at least one of the determined parameters of the intercepted request; and generating and storing, by a backup module, the backup copy of the data in an electronic database.

Abstract: Disclosed methods and systems leverage resources in a storage management system to partially synchronize primary data files based on synchronizing selected portions thereof without regard to changes that may be occurring in other non-synchronized portions. Accordingly, a number of primary data files may be partially synchronized by synchronizing designated portions thereof via auto-restore operations from backup data. This approach relies on storage management resources to designate portions of source data that is to be kept synchronized across any number of targets; detect changes to the designated portions; back up changes to secondary storage; and distribute the changes from secondary storage to the associated targets, with minimal impact to the primary data environment. The approach may be mutually applied, so that changes in any one of an associated group of source data files may be likewise detected, backed up, and distributed to the other members of the group.

Abstract: A system is described that includes a first network controller and a second network controller. The first controller operates as a master controller and the second controller operates as a standby controller for a set of access points. Using a set of VRRP advertisements between the first and second controllers, the second controller may (1) determine that the first controller has failed independent of any determination by the access points and (2) send a failover request to the access points. The failover request may cause the access points to use previously established tunnels between the second controller and each of the access points. By transmitting a failover request message from the second controller to the access points upon the detection by the second controller that the first controller has failed and independent of any determination by the access points, the system reduces network access downtime for the access points.

Abstract: Disclosed are various embodiments for distributing data items within a plurality of nodes. A data item that is subject to a data item update request is updated from a master node to a plurality of slave notes. The update of the data item is determined to be locality-based durable based at least in part on acknowledgements received from the slave nodes. Upon detection that the master node has failed, a new master candidate is determined via an election among the plurality of slave nodes.

Abstract: Embodiments include a method for temporary pipeline marking for processor error workarounds. The method includes monitoring a pipeline of a processor for an event that is predetermined to place the processor in a stuck state that results in an errant instruction execution result due to the stuck state or repeated resource contention causing performance degradation. The pipeline is marked for a workaround action based on detecting the event. A clearing action is triggered based on the marking of the pipeline. The marking of the pipeline is cleared based on the triggering of the clearing action.

Abstract: In embodiments of the invention LPARs can be run on any server in a group of servers. Upon detecting a server has failed, each LPAR then running on the failed server is identified, and servers in the group that are available for restarting the identified LPARs are determined. Identified LPARs are assigned to an available server for restarting, wherein each LPAR has a value associated with a specified LPAR priority criterion, and a given LPAR is assigned in accordance with its value. Responsive to assigning the given LPAR to an available server, a specified storage resource is connected for use by the server in association with the given LPAR, wherein the specified storage resource was previously connected for use by the failed server in association with the given LPAR.

Abstract: Systems and methods are disclosed for providing a secondary website with content mirrored from a primary website. One method includes accessing a generated primary website provided by one or more primary servers; generating a secondary website based on the generated primary website; storing the secondary website on a secondary server; and providing the secondary website from the secondary server when primary website is unavailable.

Abstract: Fault domains are defined which reflect, either physically or virtually, the topology of a networked computing environment. These defined fault domains are then used to control where cached data is replicated when running in a write back cache mode. Unlike known replication approaches, the present approach replicates such data according to a user's defined data policy and based on the defined fault domains thereby avoiding the user having to keep track of changes in computing system configurations or update their data policy when virtual machines migrate from one host computing system to another.

Abstract: Provided are a system, computer program, and method for restoring redundancy in a storage group when a storage device in the storage group fails. In response to detecting a failure of a first storage device in a storage group, wherein the storage group stores each of a plurality of extents in the first storage device and a second storage device to provide redundancy, a determination is made whether a spare storage device that has a storage capacity less than that of the storage group. One of the extents in a storage location in the second storage device that is beyond an upper limit of positions in the spare storage device is moved to a new storage location. The spare drive is incorporated into the storage group to provide redundant storage for the storage group, wherein the extents in the storage group are copied to the spare drive.

Abstract: When a media error occurs on a storage device of a number of storage devices of a redundant array, the logical stripe of data affected by the media error is determined. A portion of non-volatile memory is reserved and the logical stripe is backed up to this portion of non-volatile memory. A read request is subsequently serviced from the non-volatile memory and not from the storage devices. When a write request is received, it is first serviced to the storage devices. If successful, then the previously reserved portion of non-volatile memory is freed up, and subsequent requests are serviced using the storage devices. If unsuccessful, then the write request is serviced using the non-volatile memory.

Abstract: Systems and methods for replicating data from a production server to a backup server include recording at least one operation on one or more data items stored in a volume of a production server. The operation may be recorded as at least one journal event in a memory. A determination may then be made regarding whether a system malfunction incident has occurred in the production server and if so, a first set of journal events may be transferred from the memory to an auxiliary storage at a first time instant. At a second time instant, a second set of journal events recorded in the memory between the first and second time instants may be transferred to the auxiliary storage. At one journal event stored in the auxiliary storage unit may then be sent for replication to a backup server.

Abstract: A detecting circuit includes: a testing signal generator, arranged to selectively generate a testing signal having a first signal edge or a second signal edge to a connecting port; and a detector, arranged to detect a first detect signal on the connecting port after the testing signal having the first signal edge is coupled to the connecting port, and to detect a second detect signal on the connecting port after the testing signal having the second signal edge is coupled to the connecting port; wherein the detector is further arranged to determine if an external circuit element is coupled between the connecting port and a reference voltage according to the first detect signal and the second detect signal.

Abstract: Systems and methods are provided for generating a pseudo-random bit sequence at an output frequency using a clock signal operating at a first frequency that is lower than the output frequency. A first bit sequence of a particular type is generated using a clock signal operating at a first frequency. A second bit sequence is generated using the clock signal operating at the first frequency, where the second bit sequence is a delayed version of the first bit sequence. A delayed version of the first bit sequence is generated using the second bit sequence and another bit sequence, wherein the delayed version is delayed based on the particular type and a difference between the output frequency and the first frequency. The first bit sequence and the delayed version are combined to generate a pseudo-random bit sequence at the output frequency.

Abstract: One embodiment of the present invention provides a system that facilitates automatically identifying components to monitor in an enterprise environment. During operation, the system receives a designation of the enterprise environment. The system then identifies an enterprise application in the enterprise environment. The system also scans the enterprise application for one or more software components, wherein a software component can include a service, a database, or any other object that provides functionality to the enterprise application. The system then adds each software component to a monitor list. Next, the system determines each hardware component hosting the enterprise application and the software components. The system then adds each hardware component to the monitor list. Finally, the system stores the monitor list to facilitate subsequent monitoring of the enterprise application.