Abstract: Parallel training of a machine learning model on a computerized system may be provided. Computing tasks can be assigned to multiple workers of a system. A method may include accessing training data. A parallel training of the machine learning model can be started based on the accessed training data, so as for the training to be distributed through a first number K of workers, K>1. Responsive to detecting a change in a temporal evolution of a quantity indicative of a convergence rate of the parallel training (e.g., where said change reflects a deterioration of the convergence rate), the parallel training of the machine learning model is scaled-in, so as for the parallel training to be subsequently distributed through a second number K? of workers, where K>K??1. Related computerized systems and computer program products may be provided.

Abstract: Embodiments of the present invention provide computer-implemented methods, computer program products and systems. Embodiments of the present invention can run preemptable tasks distributed according to a distributed environment, wherein each task of a plurality of preemptable tasks has been assigned two or more of the training data samples to process during each iteration. Embodiments of the present invention can, upon verifying that a preemption condition for each iteration is satisfied: preempt any task of the preemptable tasks that have started processing training data samples assigned to it, and update the cognitive model based on outputs obtained from completed tasks, including outputs obtained from both the preempted tasks and completed tasks that have finished processing all training data samples as assigned to it.

Abstract: Embodiments of the present invention provide computer-implemented methods, computer program products and systems. Embodiments of the present invention can run preemptable tasks distributed according to a distributed environment, wherein each task of a plurality of preemptable tasks has been assigned two or more of the training data samples to process during each iteration. Embodiments of the present invention can, upon verifying that a preemption condition for each iteration is satisfied: preempt any task of the preemptable tasks that have started processing training data samples assigned to it, and update the cognitive model based on outputs obtained from completed tasks, including outputs obtained from both the preempted tasks and completed tasks that have finished processing all training data samples as assigned to it.

Abstract: Parallel training of a machine learning model on a computerized system may be provided. Computing tasks can be assigned to multiple workers of a system. A method may include accessing training data. A parallel training of the machine learning model can be started based on the accessed training data, so as for the training to be distributed through a first number K of workers, K>1. Responsive to detecting a change in a temporal evolution of a quantity indicative of a convergence rate of the parallel training (e.g., where said change reflects a deterioration of the convergence rate), the parallel training of the machine learning model is scaled-in, so as for the parallel training to be subsequently distributed through a second number K? of workers, where K>K??1. Related computerized systems and computer program products may be provided.

Abstract: Performance of a data cache is controlled; the cache implements a garbage collection process for maintaining free storage blocks in a data store of the cache and an eviction policy for selecting data to be evicted from the cache. A cache performance control method defines a performance target for operation of the cache and, in operation of the cache, monitors performance of the cache in relation to the performance target. The garbage collection process is selectively performed in a relocation mode and an eviction mode so as to promote compliance with the performance target. In the relocation mode, data contained in a set of storage blocks selected for garbage collection is relocated in the data store. In the eviction mode, a set of storage blocks for garbage collection is selected in dependence on the eviction policy and data contained in each selected storage block is evicted from the cache.

Abstract: Performance of a data cache is controlled; the cache implements a garbage collection process for maintaining free storage blocks in a data store of the cache and an eviction policy for selecting data to be evicted from the cache. A cache performance control method defines a performance target for operation of the cache and, in operation of the cache, monitors performance of the cache in relation to the performance target. The garbage collection process is selectively performed in a relocation mode and an eviction mode so as to promote compliance with the performance target. In the relocation mode, data contained in a set of storage blocks selected for garbage collection is relocated in the data store. In the eviction mode, a set of storage blocks for garbage collection is selected in dependence on the eviction policy and data contained in each selected storage block is evicted from the cache.