Abstract: Embodiments include a method, system, and computer program product for allocating data to storage in a network. A data item accessed by a server in the network is identified. A controller classifies the identified data item based on at least one of: a frequency of access requests for the data item by the server and an access time associated with providing the data item to the server once the server requests the data item. A memory of a switch in the network is selected for storing the data item based on the classification of the data item. The controller causes the data item to be stored in the memory of the switch, from which the data item is accessed by the server upon request.

Abstract: In a distributed file system with a number of replicas of a file set to two or more replicas, a computer sets a timer to track a time since a last access to the file, wherein the replicas of the file are distributed across two or more nodes within the distributed file system. Responsive to an access to the file prior to the timer reaching a first timer window threshold, the computer resets the timer. Responsive to the timer matching a first timer window threshold, the computer automatically reduces a number of replicas of the file within the distributed file system, wherein the probability that the file will be accessed prior to the first timer window threshold is greater than the probability that the file will be accessed after the first timer window threshold.

Abstract: In a distributed file system with a number of replicas of a file set to two or more replicas, a computer sets a timer to track a time since a last access to the file, wherein the replicas of the file are distributed across two or more nodes within the distributed file system. Responsive to an access to the file prior to the timer reaching a first timer window threshold, the computer resets the timer. Responsive to the timer matching a first timer window threshold, the computer automatically reduces a number of replicas of the file within the distributed file system, wherein the probability that the file will be accessed prior to the first timer window threshold is greater than the probability that the file will be accessed after the first timer window threshold.

Abstract: Embodiments include a method, system, and computer program product for managing workloads in a network. A switch receives data associated with a workload. The received data is tagged with an identifier that associates the data with the workload. The received data is compressed based on determining that second data stored in a buffer of the switch exceeds a threshold. The switch stores the compressed data in the buffer. The compressed data is transmitted to a second network based on a determination that the switch is over-subscribed.

Abstract: Embodiments include a method, system, and computer program product for allocating data to storage in a network. A data item accessed by a server in the network is identified. A controller classifies the identified data item based on at least one of: a frequency of access requests for the data item by the server and an access time associated with providing the data item to the server once the server requests the data item. A memory of a switch in the network is selected for storing the data item based on the classification of the data item. The controller causes the data item to be stored in the memory of the switch, from which the data item is accessed by the server upon request.

Abstract: Embodiments include a method, system, and computer program product for managing workloads in a network. A switch receives data associated with a workload. The received data is tagged with an identifier that associates the data with the workload. The received data is compressed based on determining that second data stored in a buffer of the switch exceeds a threshold. The switch stores the compressed data in the buffer. The compressed data is transmitted to a second network based on a determination that the switch is over-subscribed.

Abstract: In a data processing system, a selection is made, based at least on addresses of previously detected errors in a memory subsystem, between at least a first timing and a second timing of data transmission with respect to completion of error detection processing on a target memory block of the memory access request. In response to receipt of the memory access request and selection of the first timing, data from the target memory block is transmitted to a requestor prior to completion of error detection processing on the target memory block. In response to receipt of the memory access request and selection of the second timing, data from the target memory block is transmitted to the requestor after and in response to completion of error detection processing on the target memory block.

Abstract: In a data processing system, a selection is made, based at least on addresses of previously detected errors in a memory subsystem, between at least a first timing and a second timing of data transmission with respect to completion of error detection processing on a target memory block of the memory access request. In response to receipt of the memory access request and selection of the first timing, data from the target memory block is transmitted to a requestor prior to completion of error detection processing on the target memory block. In response to receipt of the memory access request and selection of the second timing, data from the target memory block is transmitted to the requestor after and in response to completion of error detection processing on the target memory block.

Abstract: In a data processing system, a selection is made, based at least on addresses of previously detected errors in a memory subsystem, between at least a first timing and a second timing of data transmission with respect to completion of error detection processing on a target memory block of the memory access request. In response to receipt of the memory access request and selection of the first timing, data from the target memory block is transmitted to a requestor prior to completion of error detection processing on the target memory block. In response to receipt of the memory access request and selection of the second timing, data from the target memory block is transmitted to the requestor after and in response to completion of error detection processing on the target memory block.

Abstract: A method includes receiving a query for data, the query including a deadline for completion. The method includes determining a plurality of access plans for the query. The method includes selecting an access plan from the plurality of access plans, wherein the access plan is selected according to an execution time for the access plan that indicates the query can be completed prior to the deadline for completion and wherein the access plan is selected to provide a lowest resource usage among access plans that can be completed prior to the deadline for completion.

Abstract: A data management system receives a query for data. A deadline for completing the query is supplied along with the query. The data management system determines access plans for the query and selects one of the access plans for execution to return a results set for the query. The selected access plan is selected according to an estimated ability to complete the query prior to the deadline. Further, the access plan is selected according to a lowest resource usage among the plans that can be completed the query prior to the deadline.

Abstract: A data management system receives a query. A query optimizer determines an access plan for the query. The access plan has a plurality of stages, where each stage may have a different degree of parallelism for processing threads executing the stage. The query optimizer determines a maximum degree of parallelism for the plurality of stages and reserves processor cores based on the maximum degree of parallelism. The data management system schedules threads on the reserved processor cores for a first stage of the access plan. The data management system changes simultaneous multi-threading (SMT) mode for the processor cores when a second stage of the access plan having a different degree of parallelism is executed.

Abstract: A data management system receives a query for data. A deadline for completing the query is supplied along with the query. The data management system determines access plans for the query and selects one of the access plans for execution to return a results set for the query. The selected access plan is selected according to an estimated ability to complete the query prior to the deadline. Further, the access plan is selected according to a lowest resource usage among the plans that can be completed the query prior to the deadline.

Abstract: A method includes receiving a query for data, the query including a deadline for completion. The method includes determining a plurality of access plans for the query. The method includes selecting an access plan from the plurality of access plans, wherein the access plan is selected according to an execution time for the access plan that indicates the query can be completed prior to the deadline for completion and wherein the access plan is selected to provide a lowest resource usage among access plans that can be completed prior to the deadline for completion.

Abstract: An approach is provided that retrieves a time spent value corresponding to a selected partition that is selected from a group of partitions included in a virtualized environment running on a computer system. The virtualized environment is provided by a Hypervisor. The time spent value corresponds to an amount of time the selected partition has spent processing interrupts. A number of virtual CPUs have been assigned to the selected partition. The time spent value (e.g., a percentage of the time that the selected partition spends processing interrupts) is compared to one or more interrupt threshold values. If the comparison reveals that the time that the partition is spending processing interrupts exceeds a threshold, then the number of virtual CPUs assigned to the selected partition is increased.

Abstract: A method, system and computer program product for optimizing allocation of resources to partitions of a data processing system are disclosed. The method includes creating a first virtual central processing unit and a second virtual central processing unit, wherein at least one of the set of the first virtual processing unit and the second virtual processing spans across a first physical processing unit and a second physical processing unit. One or more resources from the first and second virtual central processing units are allocated to a first partition and a second partition. Whether one or more processes running on the first partition can utilize additional resources is determined. One or more resources from the first virtual central processing unit and resources from the second virtual central processing unit are reallocated to the first partition, wherein at least one of the resources was previously allocated to the second partition.

Abstract: A system and method for allowing jobs originating from different partitions to simultaneously utilize different hardware threads on a processor by concatenating partition identifiers with virtual page identifiers within a processor's translation lookaside buffer is presented. The device includes a translation lookaside buffer that translates concatenated virtual addresses to system-wide real addresses. The device generates concatenated virtual addresses using a partition identifier, which corresponds to a job's originating partition, and a virtual page identifier, which corresponds to the executing instruction, such as an instruction address or data address. In turn, each concatenated virtual address is different, which translates in the translation lookaside buffer to a unique system-wide real address. As such, jobs originating from different partitions are able to simultaneously execute on the device and, therefore, fully utilize each of the device's hardware threads.

Abstract: The inventive method includes creating a first virtual central processing unit (CPU) and a second virtual CPU, where at least one of the set of the first virtual CPU and the second virtual CPU spans across a first physical processing unit and a second physical processing unit. One or more resources from the first and second virtual CPUs are allocated to a first partition and a second partition. Whether one or more processes running on the first partition can utilize additional resources is determined. One or more resources from the first virtual CPU and resources from the second virtual CPU are reallocated to the first partition, where at least one of the resources was previously allocated to the second partition.

Abstract: Administration of locks for critical sections of computer programs in a computer that supports a multiplicity of logical partitions that include determining by a thread executing on a virtual processor executing in a time slice on a physical processor whether an expected lock time for a critical section of the thread exceeds a remaining entitlement of the virtual processor in the time slice and deferring acquisition of a lock if the expected lock time exceeds the remaining entitlement.

Abstract: An approach is provided that retrieves a time spent value corresponding to a selected partition that is selected from a group of partitions included in a virtualized environment running on a computer system. The virtualized environment is provided by a Hypervisor. The time spent value corresponds to an amount of time the selected partition has spent processing interrupts. A number of virtual CPUs have been assigned to the selected partition. The time spent value (e.g., a percentage of the time that the selected partition spends processing interrupts) is compared to one or more interrupt threshold values. If the comparison reveals that the time that the partition is spending processing interrupts exceeds a threshold, then the number of virtual CPUs assigned to the selected partition is increased.