International Business Machines Corporation

Alan M Frazier

Bart Richard Cilfone

Patrick Aaron Tamborski

Sanjaya Kumar

Jason K Resch

Scott Michael Horan

Mark Duane Seaborn

Shibhani Rai

Andrew D Baptist

Wesley B Leggette

Manish Motwani

Bart Richard Clifone

A computing device includes an interface configured to interface and communicate with a communication system, a memory that stores operational instructions, and processing circuitry operably coupled to the interface and to the memory that is configured to execute the operational instructions to perform various operations. The computing device monitors a current state of a distributed storage and task processing network (DSTN) management unit by periodically collecting metadata that is compared to a previously recognized similar state. The computing device facilitates returning a resolution to the DSTN management unit to modify the current state using corrective actions and the DSTN management unit is configured to automatically execute the resolution or to record the resolution. Recorded resolutions are later passed to operators assigned to the DSN.

Automatic intervention of global coordinator

A distributed storage network (DSN) processes storage unit maintenance tasks on multiple tiers within the DSN. A master storage unit coordinates pending maintenance tasks when a DSN management unit, originally processing the pending maintenance tasks, changes its status to offline. The method includes the master storage unit aggregating pending maintenance tasks from corresponding DSN storage units into an ordered list of maintenance tasks, facilitating, based on the ordered list of maintenance tasks, coordination of a next maintenance task with a corresponding storage unit and directing execution of the next maintenance task by the corresponding storage unit.

Multi-tier coordination of destructive actions

A multi-system dispersed storage network (DSN) comprises a first DSN system that includes a first plurality of storage units, a first plurality of dispersed storage (DS) processing units and a first set of management units and a second DSN system that includes a second plurality of storage units, a second plurality of DS processing units and a second set of management units. The first set of management units manages a first set of configuration data of the first DSN system. A first management unit of the second set of management units, when, in a first mode, manages at least a first configuration data of the first set of configuration data, in a second mode, manages at least a first configuration data of the second set of configuration data, and in a third mode, manages at least the first configuration data of the first and second sets of configuration data.

Isolation of management data for security and operational advantages

A method for coordinating management of operation risks in a distributed storage network (DSN) that includes multiple distributed computing systems including DSN memories begins with a global coordinating unit receiving messages including metadata from managing units associated with the multiple distributed computing systems. The method continues with the global coordinating unit determining, based on the metadata, that storage units in one or more of the distributed computing systems are executing or planning to execute an operation that could result in data loss or data outage. The method continues with the global coordinating unit transmitting an alert, including a command to halt the operation at the affected distributed computing systems.

Global coordination of in-progress operation risks for multiple distributed storage network memories

A method begins by determining, by a local management unit of a first locally operated dispersed storage network (DSN) memory of a DSN, a DSN memory issue, where the DSN includes a plurality of locally operated DSN memories and at least one global DSN management unit, and where the plurality of locally operated DSN memories includes the first locally operated DSN memory. The method continues by determining whether the DSN issue is a global DSN issue or a local DSN issue. When the DSN issue is the global DSN issue, the method continues by determining whether the local management unit is configured as a DSN global management unit. When the local management unit is not configured as the global DSN management unit, the method continues by generating DSN operational issue information based on the DSN memory issue and sending the DSN operational issue information to the global DSN management unit.

Catastrophic data loss prevention by global coordinator

A method for execution by one or more processing modules of one or more computing devices of a dispersed storage network (DSN), the method begins by identifying, for data stored within a DSN memory, one or more encryption keys used to encrypt data stored within the DSN memory. The method continues by identifying, for data stored within a portion of the DSN memory requiring sanitization, a master key of the one or more encryption keys that encrypts all of the data stored within the portion to be sanitized. The method continues by determining, if the master key is not used to encrypt data stored outside of the portion to be sanitized. The method continues, if the master key is not used to encrypt data stored outside of the portion to be sanitized, by sanitizing the data stored within a portion of the DSN memory by erasing the master key.

Using a master encryption key to sanitize a dispersed storage network memory

A method for a dispersed storage network (DSN), the method begins by creating a plurality of DSN storage sites including a subset of a plurality of DSN storage units, retrieving current site configuration for each of the plurality of DSN storage units and electing a local master management unit for each of the DSN storage sites. The method continues by maintaining aggregated views of each the DSN storage sites via each of the local master management units, collecting, via the local master management unit, local state information from the subset of a plurality of DSN storage units, calculating, via the local master management unit, tabulated state information and distributing the tabulated state information across all storage units within the DSN storage site, thus reaching consistency. The method includes updating the DSN management unit with the tabulated state information.

Site-based coordination of distributed storage network memory

A method includes delegating, by a managing unit of a dispersed storage network (DSN), authority to configure a sub-portion of at least a portion of the DSN managed by the managing unit to a designated entity according to one or more configuration constraints. The managing unit manages parent configuration data that contains a compilation of configuration changes made to the at least a portion of the DSN. The method further includes managing, by the designated entity, child configuration data that includes a compilation of configuration changes made to the sub-portion of the at least a portion of the DSN. The method further includes, when resolution of the child configuration data and the parent configuration data is triggered, determining whether the child configuration data is formatted in a manner consistent with formatting of the parent configuration data and, if so, incorporating the child configuration data into the parent configuration data.

Delegating dispersed storage network configuration capabilities while preserving ownership constraints

A storage unit (SU) in a dispersed storage network (DSN) coordinates with affiliated dispersed storage units (SUs) to designate a leader SU among the plurality of SUs and when the SU is designated the leader, receives management information that is associated with the affiliated SUs from at least some of the affiliated SUs. The SU processes the management information from the at least some of the affiliated SUs to determine whether at least one of the affiliated SUs is offline; and based on a determination that the at least one of the SUs of the affiliated SUs is offline, transmits the management information for the affiliated SUs to one or more administrators associated with the DSN.

Peer to peer health monitoring of dispersed storage units in a distributed storage network

A method for a dispersed storage network (DSN) includes electing a local master management unit for each DSN storage site. The method continues by maintaining aggregated views of each the DSN storage sites via each of the local master management units, collecting, via the local master management unit, local state information from the subset of a plurality of DSN storage units, calculating, via the local master management unit, tabulated state information and distributing the tabulated state information across all storage units within the DSN storage site, thus reaching consistency. The method includes generating, based on the tabulated state information, alerts for one or more targets within the DSN storage site and distribution of the alerts to the one or more targets within the DSN storage site. The method further includes updating the DSN management unit with the tabulated state information and alert information.

Site-centric alerting in a distributed storage system

A method begins by obtaining dispersed storage network (DSN) operational issue information concerning a DSN memory issue in a first locally operated DSN memory of a plurality of locally operated DSN memories. The method continues with determining whether the DSN memory issue is regarding a local DSN issue or a global DSN issue. When the DSN memory issue is the global DSN issue, the method continues with determining a corrective action in response to the DSN memory issue and determining whether the global DSN management unit is configured for automatic correction of the DSN memory issue. When the global DSN management unit is configured for automatic correction of the DSN memory issue in the first locally operated DSN memory of the plurality of locally operated DSN memories, the method continues with executing the corrective action in the first locally operated DSN memory.

A method begins with one or more processing modules of a dispersed storage network (DSN) identifying a set of encoded data slices generated from a data segment based on an error encoding dispersal function. The method continues with the one or more processing modules identifying a set of storage units (SUs) that distributedly store the set of EDSs, and determining a first plurality of performance values associated with the set of SUs, followed by the one or more processing modules receiving from the set of SUs a second plurality of performance values generated by the set of SUs. The one or more processing modules continue by comparing the first plurality of performance parameter values to first desired performance values, followed by comparing the second plurality of performance parameters values to second desired performance parameter values; and based on an unfavorable comparison of the first plurality of parameter values and the second plurality of performance parameters values to the respective desired values, identifying any SUs that are below a threshold performance, and then rectifying the SUs that are below the threshold performance.

Fault detection and recovery in a distributed storage network

A method includes generating an entry for inclusion in a configuration queue regarding a configuration submission regarding DSN managed by a managing unit that uses a manager's configuration database stored in a first database. The method further includes updating a first version of the configuration queue stored in first database and a second version of the configuration queue stored in a manager's vault to include the entry. When the manager's configuration database requires restoring, the method further includes copying a backup version of the manager's configuration database to the first database to create an initially restored manager's configuration database and establishing an association between the managing unit and the initially restored manager's configuration database. When associated, the method further includes accessing the second version, identifying configuration changes not included in the initially restored manager's configuration database, and updating the initially restored manager's configuration database to include the identified configuration changes.

Growing community of inventors

Palatine, IL, United States of America

Alan M Frazier