Emc Corporation

Emc IP Holding Company LLC

Robert J Pellowski

Hitesh Trivedi

Helen S Raizen

Harold Martin Sandstrom

Edith Epstein

Tao Tao

Deepak M Gaikwad

David Stephen Reiner

Eric I West

Jimmy K Seto

Harrison Roberts

An improved technique for storing I/O metrics includes assigning metric values to data buckets held in kernel memory. Each data bucket covers a particular range of values of a respective metric and is configured as a counter, whose count is incremented each time the multipathing driver obtains or computes a metric value that falls within the range of the data bucket. Bucket counts can be read by an external program to obtain aggregate information about I/O metrics over time. The aggregate information can be fed back to the multipathing driver to enable improved selections of paths for conveying data to and from a storage array.

Acquisition and kernel memory storage of I/O metrics

A method is disclosed for processing a kernel driver control command at a client machine. The method includes (a) receiving the kernel driver control command at the client machine via a user interface, (b) selecting a target server from a set of servers based on the kernel driver control command, (c) selecting a protocol path to use to reach the target server based on the availability of the target server via a set of possible protocol paths, (d) reformatting the kernel driver control command into a protocol format associated with the selected protocol path, and (e) sending the reformatted kernel driver control command to the target server via the selected protocol path for execution by a kernel driver running on the target server. A corresponding apparatus and computer program product for performing the method are also provided.

Dynamically selectable transport for kernel driver management

An improved technique for managing access to performance data of a kernel driver includes acquiring performance data in the kernel driver over identified intervals of time marked by the kernel driver itself and transferring the performance data to a persistence manager outside the kernel, where client applications can access the performance data and/or post-processed versions thereof without disturbing the operation of the kernel driver.

Efficient management of kernel driver performance data

A host computer accesses a storage volume via multiple paths and maintains path mode information identifying either an active or standby operating mode. An active-mode path is generally usable, and a standby-mode path is usable under exceptional conditions such as non-availability of an active-mode path. A reason identifier identifies a first operating condition causing the path to be placed in the standby mode. Upon a path-activating event for a standby-mode path, if there is a second operating condition indicating that the path should be maintained in the standby operating mode, then the path is maintained in the standby operating mode and the path mode information is set to identify the second operating condition, and otherwise the operating mode is set to active. A path-activating event can be a manual action by a user or automatic action such as expiration of an aging timer to re-attempt use of a standby path.

Management of path operating mode transitions in multipathing host computer

An improved technique for handling events in a multipathing driver employs an event queue and a queue manager that run in the kernel of a computing system. The queue manager receives events raised by the multipathing driver, as well as events raised by software constructs, such as application programs. Records of events are added to the event queue in the order the queue manager receives them. Event records may be consumed chronologically by external software. Preferably, the event queue is sufficiently large to store all events arising out of most predicted fault scenarios. Also, the queue manager is sophisticated and can perform certain diagnostic and analysis tasks without the aid of external software.

Ordered kernel queue for multipathing events

A processing platform comprises at least one server, computer or other processing device having a processor coupled to a memory, and implements a plurality of modules for adaptive optimization across an information technology (IT) infrastructure. The modules include a collector configured to gather information from the infrastructure, an analyzer coupled to the collector and configured to analyze the information gathered by the collector, a policy module specifying a plurality of policy sets, and a controller that is coupled to the collector, the analyzer and the policy module. The controller is configured to adjust one or more parameters of the infrastructure via corresponding control points. Associated with the analyzer is a situational analysis framework configured to periodically select and deploy for use by the controller a particular one of the specified plurality of policy sets responsive to changing operating conditions of the infrastructure. The infrastructure may comprise a virtual data center (VDC) or other type of virtual infrastructure.

Growing community of inventors

Ledyard, CT, United States of America

Robert J Pellowski