International Business Machines Corporation

Michael Keller

Rivka Mayraz Matosevich

Orit Nissan-Messing

Ofer Leneman

Eliyahu Weissbrem

Lior Chen

Osnat Shasha

Moshe Weiss

Dani Shemesh

Itzhack Goldberg

Zah Barzik

Ehood Garmiza

Jonathan Amit

Daniel Lereya

Daniel Gan-Levi

Amit Bareket

Moriel Lechtman

Yoni Raveh

Lior Shlomov

Gal Sinay

Ariel Waizel

Ram Elron

Avraham Sabzerou

Yuval A Ben-Horin

Yuval Ben-Horin

Daniel M Aizer

Provided are a computer program product, system, and method for determining the location for volumes of data being initially stored within a storage space, regardless of the physical location of the data. The storage space includes stripes composed of volumes, which can be logically represented as a utilization histogram of stripe locations offset from one another. Sometime the stripes are fully allocated with one large volume or partially allocated with multiple, arbitrary-sized smaller volumes. When there are multiple smaller volumes that do not utilize all of the available stripe space, gaps form. To minimize the creation of such gaps, when a volume of data is initially stored, a start location to place the volume of data is selected by using selection criteria as guidance.

Selection and placement of volumes in a storage system using stripes

Methods, computing systems and computer program products implement embodiments of the present invention that include associating a token with a computing device, defining preferences for the computing device, and conveying, by the computing device, the token and the preferences to an event processing system. Upon the event processing system, an event message from a computing system via a one-way firewall and matching the computing device preferences to the event message, the event processing system can convey the token and the event message to a push notification system. In some embodiments, upon the push notification service receiving the token and the event message, the mobile device can be identified based on the token, and the event message can be conveyed to the computing device. The event messages may include a severity level, and the preferences may include a severity threshold and a message detail level.

Push notifications of system events in a restricted network

Methods, computing systems and computer program products implement embodiments of the present invention that include defining a plurality of failure domains for sets of storage devices in a storage facility, and defining, using the failure domains, one or more limitations for distributing data on the storage devices. Upon identifying a data distribution configuration for a software defined storage system that is compliant with the one or more limitations, the identified data distribution configuration can be presented to a user. The failure domains may include physical failure domains, logical failure domains, or a combination of physical and logical failure domains, and the limitations may include mandatory limitations or a combination of mandatory and non-mandatory limitations. In embodiments including non-mandatory limitations, the data distribution configuration may not comply with all the non-mandatory limitations, and any non-mandatory limitations that are not met can be flagged and presented to the user.

User-defined failure domains for software-defined storage systems

Provided are a computer program product, system, and method for generating a configuration page for rendering in a graphical user interface (GUI) for managing provisioning of storage resources in a storage system. A determination is made of allocation commitment certainties for virtual storage ranges to allocate from the storage system based on physical capacity in the storage system and an estimated data reduction rate. Each of the allocation commitment certainties indicates a likelihood that a corresponding of the virtual storage ranges will be available for allocation in the storage system. Visual representation of a plurality of user selectable storage spaces indicating the allocation commitment certainties for the virtual storage ranges in the user selectable storage spaces are rendered in the configuration page. A user selection graphic that when selected provisions a selected of the user selectable storage spaces in the storage system is rendered in the configuration page.

Generating a configuration page for rendering in a graphical user interface (GUI) for managing provisioning of storage resources in a storage system

Provided are a computer program product, computer system, and method for smart selection of a storage module to be excluded when a connection between two storage modules is broken. An indication is received from a first storage module that a connection between the first storage module and a second storage module is broken. In response to determining that the second storage module is accessible, values of exclusion criteria for the first storage module are determined and summed to identify a first exclusion total. Then, values of exclusion criteria for the second storage module are determined and summed to identify a second exclusion total. In response to determining that the first exclusion total exceeds the second exclusion total, the second storage node is excluded from the cluster. In response to determining that the second exclusion total exceeds the first exclusion total, the first storage node is excluded from the cluster.

Smart selection of a storage module to be excluded

Methods, computing systems and computer program products implement embodiments of the present invention that include detecting a loss of communication with a given storage node among multiple storage nodes in a distributed computing system. Upon detecting the loss of communication, a log including updates to the data stored in the given storage node is recorded and, the recorded updates can be applied to the given storage node upon communication with the given storage node being reestablished. In some embodiments, the distributed storage system may be configured as a software defined storage system where the storage nodes can be implemented as either virtual machines or software containers. In additional embodiments, upon detecting the loss of communication, a redistribution of the mirrored data among remaining storage nodes is initiated upon detecting the loss of communication, and the redistribution is rolled back upon reestablishing the communication.

Rebuild rollback support in distributed SDS systems

Methods, computing systems and computer program products implement embodiments of the present invention that include configuring multiple servers coupled to a network as a software defined storage (SDS) grid. A first given server receives, via the network, an input/output (I/O) request from a host computer, and determines a location of data associated with the I/O request. In some embodiments, each of the servers maintains a local grid data map that store locations for all data managed by the SDS grid. Upon identifying, in its respective local grid data map, that a second given server is configured to process the I/O request, the first given server forwards the I/O request to the second given server for processing, and upon receiving a result of the I/O request from the second given server, the first given server conveys the result of the I/O request to the host computer.

Clientless software defined grid

Methods, computing systems and computer program products implement embodiments of the present invention that include assigning multiple storage resources to a reserve pool, and defining one or more storage policy parameters that include allocation policies and deallocation policy parameters. A software defined storage (SDS) controller is initialized by allocating, from the reserve pool, one or more of the storage resources to the SDS controller, and a current state of the SDS controller is monitored. In a first embodiment, upon detecting that the state of the SDS controller meets a given allocation policy parameter, an additional storage resource from the reserve pool is allocated from the reserve pool to the SDS controller. In a second embodiment, upon detecting that the state of the SDS controller meets a given deallocation policy parameter, a given the SDS controller is deallocated from the SDS controller and released back to the storage pool.

Policy driven storage hardware allocation

Provided are a computer program product, system, and method for distributed global data vaulting in a grid of server computers. At each of the server computers, it is determined that there is a system shutdown. For each data element at that server computer, a list of one or more vault devices at other server computers is retrieved from a vault distribution table, and the data element is dumped to the one or more vault devices at the other server computers.

Distributed global data vaulting mechanism for grid based storage

Exemplary method, system, and computer program product embodiments for efficient state tracking for clusters are provided. In one embodiment, by way of example only, in a distributed shared memory architecture, an asynchronous calculation of deltas and the views is performed while concurrently receiving client requests and concurrently tracking the client requests times. The results of the asynchronous calculation may be applied to each of the client requests that are competing for data of the same concurrency during a certain period with currently executing client requests. A latency is bound for the client requests by a time necessitated for the asynchronous calculation of at least two of the deltas where a first state snapshot is atomically taken while simultaneously calculating the at least two of the deltas.

Efficient state tracking for clusters

Provided are a computer program product, computer system, and method for utilization of a thin provision grid storage array. Storage devices are divided into a plurality of slices, wherein each of the slices has a same number of partitions. Volumes in a pool are configured, wherein the pool is associated with physical capacity of a slice, wherein a volume size is unlimited, and wherein a cumulative size of all volumes in the pool is unlimited. While data is being written to the volumes in the pool, partitions of a first slice of the plurality of slices are allocated to the volumes in the pool and, in response to determining that the first slice needs additional physical capacity, allocating a partition of a second slice to the first slice by updating a mapping table.

Growing community of inventors

Haifa, Israel

Michael Keller