Abstract: A method is provided. The method stores a first version of an artifact in a first environment. Further, the method stores a second version of the artifact in the first environment. In addition, the second version of the artifact is distinct from the first version of the artifact. The method also deploys the first version of the artifact and the second version of the artifact to a second environment so that the first artifact and the second artifact can be run simultaneously in the second environment. The second environment is distinct from the first environment.

Abstract: A method, system and computer program product are provided for implementing dynamic adjustment of resources allocated to Single Root Input/Output Virtualization (SRIOV) Remote Direct Memory Access (RMDA) virtual functions (VFs) in Cloud Software Defined Server environments. A hardware management console (HMC), and a hypervisor are used to implement resource allocation to the SRIOV RDMA VFs based on resource usage. The hypervisor checks resource usage for the resource allocations of the SRIOV RDMA VF relative to lower and upper threshold values. Responsive to identifying the resource usage below the lower threshold or above the upper threshold, the hypervisor sends an event to the HMC, and the HMC starts a resource redistribution process.

Abstract: An information processing unit of a communication apparatus includes a non-volatile memory and a volatile memory. A control unit of the communication apparatus loads the boot image from the non-volatile memory to the volatile memory upon activation of the communication apparatus, activates the virtual machine template in the loaded boot image on the volatile memory, deactivates the activated virtual machine template after reconfiguring the activated virtual machine template to minimum required configuration to execute a virtual machine.

Abstract: Systems, methods, and computer program products to perform an operation, the operation comprising processing a first logical partition on a shared processor for the duration of a dispatch cycle, issuing, by a hypervisor, at a predefined time prior to completion of the dispatch cycle, a lightweight hypervisor decrementer (HDEC) interrupt specifying a cache line address buffer location in a virtual processor, and responsive to the lightweight HDEC, writing, by the shared processor, a set of cache line addresses used by the first logical partition to the cache line address buffer location in the virtual processor.

Abstract: Embodiments of the invention relate to scalable policy management in an edge virtual bridging (EVB) environment. One embodiment includes a system including a physical end station including a hypervisor, wherein the physical end station creates at least one virtual machine (VM). A virtual station interface (VSI) database is coupled to a VM manager server. The VSI database stores policy information comprising one or more rules for different VM types and access rules. A policy management module is coupled to a switch adjacent to the physical end station. The policy management module generates a first table using at least a portion of the policy information, generates a second table with a portion of VM information received from the hypervisor for the VM, and uses the first table and the second table to retrieve and apply rules for the VM.

Abstract: Embodiments of the invention relate to scalable policy assignment in an edge virtual bridging (EVB) environment. One embodiment includes a system including a physical end station includes a hypervisor. The physical end station creates at least one virtual machine (VM). A virtual station interface (VSI) database (DB) is coupled to a VM manager server. The VSI DB stores policy information and bandwidth filter information. A policy assignment module is coupled to a switch adjacent to the physical end station. The policy assignment module generates a VSI DB table with at least a portion of the VSI DB information from the VSI DB and a policy discriminator (PD) value for each VSI type ID.

Abstract: A virtualization host may implement variable timeslices for processing latency dependent workloads. Multiple virtual compute instances on a virtualization host may utilize virtual central processing units (vCPUs) to obtain physical processing resources, such as one or more central processing units (CPUs). A vCPU currently utilizing a CPU to performing processing work according to a scheduled timeslice may be preempted by a latency dependent vCPU before completion of the scheduled timeslice. The latency-dependent vCPU may complete processing work, and utilization of the CPU may be returned to the vCPU. A preemption compensation may be determined for the scheduled timeslice to increase the scheduled timeslice for the vCPU such that utilization for the vCPU is performed according to the increased scheduled timeslice.

Abstract: A computer determines that a utilization level of a resource has satisfied a threshold. The computer scales the allocation of the resource to the furthest of the current allocation of the resource plus a parameter and of a historical limit. The computer determines if the scaled allocation of the resource is outside the historical limit and if so, sets the historical limit equal to the scaled allocation of the resource. The computer determines whether the scaling of the allocation of the resource will result in an allocation oscillation. The computer determines if the scaled allocation of the resource is outside a boundary parameter and if so, sets the allocation of the resource equal to the boundary parameter.

Abstract: A workflow is designated for execution across a plurality of autonomous computational entities automatically. Among other things, the cost of computation is balanced with the cost of communication among computational entities to reduce total execution time of a workflow. In other words, a balance is struck between grouping tasks for execution on a single computational entity and segmenting tasks for execution across multiple computational entities.

Abstract: A system and method for dynamically determining parameters of crafting activities for individual users in a virtual space are disclosed. The crafting activity parameters may be determined based on rate of usage of the virtual space by the users. In some examples, a user usage rate of the virtual space may be determined to measure an amount of time a user spend in the virtual space during a period of real-world time. Based on the determined user usage rate of the virtual space, crafting activity parameters may be determined, for example, to balance differences in time spent in the virtual space by the users. In some implementations, for such determinations of the crafting activity parameters, one or more user normalization standards may be obtained.

Abstract: Methods of offering network performance guarantees in multi-tenant datacenters are described. In an embodiment, a request for resources received at a datacenter from a tenant comprises a number of virtual machines and a performance requirement, such as a bandwidth requirement, specified by the tenant. A network manager within the datacenter maps the request onto the datacenter topology and allocates virtual machines within the datacenter based on the available slots for virtual machines within the topology and such that the performance requirement is satisfied. Following allocation, stored residual capacity values for elements within the topology are updated according to the new allocation and this updated stored data is used in mapping subsequent requests onto the datacenter. The allocated virtual machines form part of a virtual network within the datacenter which is allocated in response to the request and two virtual network abstractions are described: virtual clusters and virtual oversubscribed clusters.

Abstract: A preempt of a live migratable workload, or job, in a distributed computing environment is performed, allowing it to release its resources for use by a higher priority workload by moving to another place in the distributed computing environment without interruption. A job scheduler receives a request to schedule a higher priority job, wherein resources needed to run the higher priority job are already dedicated for use by a currently running lower priority job. A dummy job is scheduled at a highest priority that is a copy of the lower priority job. Resources required to run the dummy job are reserved. A live migration of the lower priority job to another host is initiated, and its resources are then released. Upon a successful completion of the live migration of the lower priority job, the higher priority job is then dispatched to run using the now released resources.

Abstract: A method of prioritizing processing units in a system for task scheduling includes, for each processing unit of a plurality of processing units in the system, determining a value that represents a thermal condition of a location of the processing unit. It is determined which of the plurality of processing units is not fully loaded and is in a location with a most favorable thermal condition based on the value of the processing unit that represents thermal conditions of the location of the processing unit. A task is scheduled to the processing unit determined to be not fully loaded and in a location with a most favorable thermal condition based on the value of the processing unit that represents thermal conditions of the location of the processing unit.

Abstract: A plurality of process content is retained, said process content including identifiers of a plurality of part content included in each process and information which denotes dependencies among the plurality of part content. When information is inputted which designates a first process and the part content of a problem portion which is included in the first process, a process similar to the first process is retrieved. On the basis of whether there is a change in any of the plurality of part content which is included in the retrieved process, an evaluation value of the retrieved process is either incremented or decremented, and the information relating to the plurality of processes is outputted on the basis of the evaluation value.

Abstract: A device receives a state chart generated via a technical computing environment. The state chart includes a function block that includes a function that includes function input(s)/output(s). The state chart includes a state block that includes a function call to the function of the function block. The function call includes call input(s)/output(s). The device initiates execution of the state chart, parses the function into the function input(s)/output(s), and parses the function call into the call input(s)/output(s). The device processes, during the execution of the state chart, the function input(s)/output(s) with a graphical engine of the technical computing environment to generate function-related code. The device processes, during the execution of the state chart, the call input(s)/output(s) with a textual engine of the technical computing environment to generate function call-related code, and provides the function-related code and the function call-related code in generated code.

Abstract: Embodiments of the invention relate to scalable policy assignment in an edge virtual bridging (EVB) environment. One embodiment includes fetching virtual machine (VM) information for one or more VMs from a virtual station interface (VSI) database (DB). The VM information includes a VSI type identification (ID) associated with each VM. A policy discriminator (PD) value is associated for each VSI type ID. A VSI DB table is generated with at least a portion of the VM information from the VSI DB and the PD for each VSI type ID. A message is received including virtual machine (VM) information for a created VM. One or more rules and bandwidth filter information associated with a VSI type ID are retrieved from the VSI DB table. The associated rules and filter information are applied based on the PD.

Abstract: A method of executing an original agent application as a virtual agent, the method comprising encapsulating an original agent in a container file to produce a virtual agent; providing the virtual agent to an endpoint machine; and executing the virtual agent, within the container, on the endpoint machine.

Abstract: Methods, systems, and computer program products for restoring a backend after a backend disconnect. The method includes detecting a disconnect of a backend. The method further includes setting a flag and notifying a guest regarding the disconnect. The method further includes resetting a data structure accessed by the backend. The method further includes detecting (or initiating) or a reconnect of the backend. The method further includes sending the backend a state of the data structure.

Abstract: A class preloading mechanism that dynamically preloads classes at runtime in a virtual machine (VM) environment. Data structures representing preloaded classes may be stored in a persistent module corresponding to a classloader. A persistent module can be directly mapped or copied into a memory region at runtime so that the classes may not have to be loaded from the class file container. The preloaded classes are not fully linked and resolved. When a classloader receives a class request, the classloader looks up the preloaded class in the memory region and completes linking and resolution of the class. Persistent modules may be pre-generated and, for example, installed with an application. Alternatively, a persistent module for a class file container may be generated at runtime by preloading classes from the class file container into a memory region and storing data structures representing the classes as a persistent module.

Abstract: A method and apparatus for signature-based detection of kernel data structure modification are disclosed. In the method and apparatus a signature is generated for a kernel data structure, whereby the kernel data structure is capable of being modified based at least in part on access to the kernel data structure. The signature is also updated as a result of access to the kernel data structure due at least in part to one or more identified instructions being executed. The signature is used to determine whether the kernel data structure is accessed by one or more other instructions.