Abstract: Disclosed herein is an apparatus and method for a shuffle manager for a distributed memory object system. In one embodiment, a method includes forming a system cluster comprising a plurality of nodes, wherein each node includes a memory, a processor and a network interface to send and receive messages and data, wherein the network interface operates on remote direct memory access; creating a plurality of sharable memory spaces having partitioned data, wherein each space is a distributed memory object having a compute node, wherein the sharable memory spaces are at least one of persistent memory or DRAM cache; and storing data in an in-memory data structure when there is available memory in a compute node; and if there is an out of memory condition, serializing at least some of the in-memory data and spilling it to a distributed memory object system to persist shuffled data outside the compute node.

Abstract: Described are techniques for synchronizing system times. A request is sent from a first system to a second system to obtain a current remote system time. A first time value denotes a time when the request is sent. A response, including the current remote system time, is received from the second system. A second time value denotes a time when the response is received by the first system. An average of the first time value and the second time value is determined A difference between the current remote system time and the average is determined. It is determined whether the difference is more than an endurance value. Responsive to determining the difference is more than the endurance value, the difference is used to convert a remote system time of the second system to an equivalent local system time of the first system.

Abstract: Disclosed herein is an apparatus and method for a shuffle manager for a distributed memory object system. In one embodiment, a method includes forming a system cluster comprising a plurality of nodes, wherein each node includes a memory, a processor and a network interface to send and receive messages and data, wherein the network interface operates on remote direct memory access; creating a plurality of sharable memory spaces having partitioned data, wherein each space is a distributed memory object having a compute node, wherein the sharable memory spaces are at least one of persistent memory or DRAM cache; and storing data in an in-memory data structure when there is available memory in a compute node; and if there is an out of memory condition, serializing at least some of the in-memory data and spilling it to a distributed memory object system to persist shuffled data outside the compute node.

Abstract: Techniques are disclosed for monitoring the performance of transient virtual volumes created for a virtual machine. Each transient virtual volume is created in response to a first trigger event, used by the virtualization environment solely to perform a single function supporting execution of the virtual machine, and deleted in response to a second trigger event. When creation of a current transient virtual volume is detected, and while the virtualization environment uses the current transient virtual volume to perform a single function supporting execution of the virtual machine, performance data is collected for the current transient virtual volume, and combined with performance data collected for at least one other, previously deleted transient virtual volume that was used solely to perform the same single function supporting execution of the same virtual machine. The combined performance data is represented as performance data for a single monitored virtual volume associated with the virtual machine.

Abstract: Techniques for visualizing performance of file-based VVols and block-based VVols for aiding in administration of a data storage system operating in a virtualization environment allow performance of these different types of VVols to be visualized side-by-side using a common set of performance metrics. Thus, in one embodiment, the performances of file-based and block-based VVols are both converted into a mutually-compatible format and rendered for display together on screen.

Abstract: Techniques for visualizing performance of VVols for aiding in administration of a data storage system operating in a virtualization environment allow performance of these VVols to be visualized in a highly-flexible manner. Thus, in one embodiment, the performances of all VVols within a storage container are aggregated together for easy comparison among the aggregated performances of different storage containers.

Abstract: A method for use in viewing compression and migration status is disclosed. A logical volume is selected for compression into a compressed logical volume. A destination storage pool for the compressed logical volume is selected. The logical volume is compressed during migration of the logical volume to the destination pool. Unified status information of progress of the compression and migration is presented. A system for use in viewing compression and migration status is also disclosed.

Abstract: A method is provided within a storage processor of a data storage system having a fixed storage capacity shared between a plurality of data storage units, The method includes (a) performing a Compression-related operation on a data storage unit of the plurality of data storage units using a free pool of storage space, associated with the fixed storage capacity, as a swap space for the Compression-related operation, (b) detecting a size of the free pool, the size of the free pool representing unallocated storage space of the data storage system, (c) comparing the size of the free pool with a restricted threshold, and (d) if the size of the free pool falls below the restricted threshold, then pausing the Compression-related operation on the data storage unit, otherwise, continuing to perform the Compression-related operation on the data storage unit. Apparatus and computer program product for practicing the method are also provided.

Abstract: A method is provided for testing a feature provider module of a data storage management application (DSMA), the feature provider module providing a feature. The method includes (a) receiving, at a simulation module of the DSMA when the DSMA is not in communication with an external feature implementation component, a test feature command from the feature provider module, (b) at the simulation module, simulating behavior of an interface module in communication with the external feature implementation component, and (c) at the simulation module, returning a feature response to the feature provider module, the feature response communicating the simulated behavior.

Abstract: A method for use in viewing compression and migration status is disclosed. A logical volume is selected for compression into a compressed logical volume. A destination storage pool for the compressed logical volume is selected. The logical volume is compressed during migration of the logical volume to the destination pool. Unified status information of progress of the compression and migration is presented. A system for use in viewing compression and migration status is also disclosed.