Abstract: A method for a machine learning model training is provided which operates in a mixed CPU/GPU environment. The amount of general processing unit memory is larger than the amount of special processing unit memory. The method includes loading a complete training data set into the memory of the general processing unit, determining importance values relating to training data vectors in the provided training data set of the training data vectors, dynamically transferring training data vectors of the training data set from the general processing unit memory to a special processing unit memory using as decision criteria the importance value of the training data vector, wherein the importance value used is taken from an earlier training round of the machine learning model, and executing a training algorithm on the special processing unit with the training data vectors having the highest available importance values of one of the earlier training rounds.

Abstract: The invention relates to a storage system comprising one or more storage controllers, two or more storage devices and two or more storage clients, the storage controllers, the storage devices and the storage clients being coupled via a network in order to exchange information between the storage controllers, said storage devices and said storage clients, wherein each storage client is adapted to provide data-path-storage commands to the storage devices via the network bypassing the storage controllers, wherein the storage system is adapted to provide data redundancy using a data redundancy scheme higher than RAID 1.

Abstract: A method for a machine learning model training is provided which operates in a mixed CPU/GPU environment. The amount of general processing unit memory is larger than the amount of special processing unit memory. The method includes loading a complete training data set into the memory of the general processing unit, determining importance values relating to training data vectors in the provided training data set of the training data vectors, dynamically transferring training data vectors of the training data set from the general processing unit memory to a special processing unit memory using as decision criteria the importance value of the training data vector, wherein the importance value used is taken from an earlier training round of the machine learning model, and executing a training algorithm on the special processing unit with the training data vectors having the highest available importance values of one of the earlier training rounds.

Abstract: The invention relates to a storage system comprising one or more storage controllers, two or more storage devices and two or more storage clients, the storage controllers, the storage devices and the storage clients being coupled via a network in order to exchange information between the storage controllers, said storage devices and said storage clients, wherein each storage client is adapted to provide data-path-storage commands to the storage devices via the network bypassing the storage controllers, wherein the storage system is adapted to provide data redundancy using a data redundancy scheme higher than RAID 1.