Dell Products Inc.

Emc Corporation

Emc IP Holding Company LLC

Paul Bradley

Xuedong Jiang

Nigel Hislop

Hanna Yehuda

Anoop George Ninan

Boris Farizon

Samuil Shmuylovich

David Barta

Brian O'Halloran

John P Sheehy, Jr

John C Sexton

Management applications (hosts) implement management operations on a storage system by passing system calls (syscalls) to a management system. In some embodiments, all management traffic is passed from one or more instances of a management application to a management daemon executing in the context of the storage system operating system. The management daemon sorts syscalls received from the management applications and implements the syscalls using one or more array control devices. The array control devices are special target devices configured on external control hosts for sending management commands to the storage system operating system and receiving information from the storage system. In some embodiments, Non-Volatile Memory express (NVMe) devices to be used as array control devices, and enable TCP connections to be used as logical array control devices. Adding multiple protocol support in the existing management software infrastructure enables users to seamlessly migrate to the new data center technologies.

Multiple protocol array control device support in storage system management

A management daemon sorts syscalls received from the management applications into First In First Out (FIFO) queues and, when there is an available array control device, selects a syscall from one of the FIFO queues, and spawns a thread on the available array control device to service the selected syscall. The management system has a syscall scheduler configured to optimize handling of syscalls, by breaking long running syscalls into discrete operations, implementing a portion of the syscall, and then stopping operation of the syscall while saving the state of the partially executed syscall. The partially executed syscall is placed back into a queue for subsequent operation, thus freeing up the array control devise for use in connection with other syscalls. By optimizing syscall scheduling, and incremental processing of long running syscalls, it is possible to reduce the overall syscall processing time for a set of syscalls.

Method and apparatus for optimizing system call (syscall) processing

Embodiments of the present disclosure relate to dynamically adjusting storage device system configurations. Information from one or more storage devices is collected. Dynamic adjustment of each storage device's system configuration is enabled based on the information. The information can include performance data, telemetry data, system configuration data, component configuration data, change events/commands, and/or alerts. Each storage device's components can be monitored with at least one daemon. Information patterns can be identified. The patterns can be matched to spikes in data traffic and/or storage device performance. The matched patterns and spikes can be correlated with each storage device's performance and errors. Storage device system configuration adjustment options can be determined. Rules for each option can be generated. Dynamic adjustment of each storage device's system configuration based on the generated one or more rules can be enabled.

Dynamic storage device system configuration adjustment

A SAN management agent computes a validator corresponding to a full data set from which a relative (change) data set is derived. The validator, such as a checksum, identifies the full data set to which the relative data set is applicable. The validator is transmitted with the relative data set, and the store process compares the validator with the previous validator stored with the full data set in a database. If the validators correspond (i.e. match), then the received relative data set is applicable to the full data set stored at the database, and the relative data set may be processed as a change set. The remote agent therefore computes the validator on a full data set at a remote node, and transmits the validator with the change set. Subsequent relative data sets may then compare the current validator of the DB data set with the validator from which the relative set was derived.

Methods and apparatus for persisting management information changes

A status monitoring and reporting mechanism which distinguishes more significant events and prominently displays events pertaining to nodes, interconnections, and other network entities which may present problems, and to suppresses benign messages which may not require immediate attention, provides an 'at a glance' view of overall system health. By processing and displaying aggregate status events from a plurality of agent components to highlight the most severe and critical events from among a large volume of general status events allows an operator to quickly diagnose potential problems before detrimental results occur, and helps maintain a general efficient system health by isolating less efficient areas to allow for timely remedial action prior to failure. Various agents disseminated in a storage area network determine and report status events back to a central server. The server receives, aggregates, and processes the status events to display a status array indicative of overall system health.

Growing community of inventors

Macroom, Ireland

Paul Bradley