Lsi Corporation

Arindam Banerjee

Pradeep Radhakrishna Venkatesha

John Thomas

Girish Kumar Bk

Satish Sangapu

Eric Bonnell

William Lomelino

Jayaraj Rajappan

A method and controller device for upgrading firmware in a virtualized storage environment having a virtual machine manager, guest virtual machines and a storage device. The method includes downloading a new firmware solution bundle to a first logical area of the storage device, and installing the new firmware containing the virtual machine manager and guest virtual machines. The installation includes moving the solution bundle to a scratch area carved out of a P-cache area in the storage device, extracting the new firmware, copying the new firmware to the first logical area, marking the first logical area as the Active area, and marking the second logical area as the Staging area. The method also includes rebooting the virtualized storage environment with the installed new firmware, committing to the new firmware if the installation is successful, and rolling back the firmware version from the new firmware to the current firmware if the installation is not successful.

Method and system for firmware upgrade of a storage subsystem hosted in a storage virtualization environment

Disclosed is a system and method that resolves a mismatch between software versions executing on redundant controllers. A mismatch between a first software version executing on a first redundant controller and a second software version executing on a second redundant controller is identified. By comparing software version identifiers associated with the first software version, the second software version, and a stored system software identifier, a preferred software version is selected. The preferred software version is copied from a controller running the preferred software version to the other controllers.

Code synchronization

A method and controller device for supplying battery power to a virtualized storage environment having a storage controller with a virtual machine manager and a second virtual machine. In response to a battery engaged event, the first virtual machine manager enables the image of the second virtual machine to be shared with a new instance of the second virtual machine so that the image does not have to be loaded therein. The first virtual machine manager then creates the new virtual machine. The old virtual machine shuts down non-necessary hardware devices and sets necessary hardware devices to low power mode. During this time, the new virtual machine executes a backup specific start-of-day (SOD) initialization sequence. The method also synchronizes the new and old virtual machines. The method also initiates a cache memory backup operation upon synchronization of the new and old virtual machines and then shuts down the old virtual machine.

Method and system for reducing power loss to backup IO start time of a storage device in a storage virtualization environment

Partner mirroring is provided with tray loss protection in an N node storage cluster architecture. A master proxy receives and records broadcasts of nodes in a cluster and selects mirror partners in a round robin fashion, so that even numbered nodes are mirrored with other even numbered nodes and odd numbered nodes are mirrored with other odd numbered nodes. In an N node storage cluster architecture which includes a cluster of dual controllers, tray loss protection is provided using such an odd numbered and even numbered mirror pairing process.

Automatic discovery of cache mirror partners in an N-node cluster

Disclosed are a system and method to preserve and recover unwritten data present in data cache of a disk subsystem across power outages. In one embodiment, a method of a controller is described. The method includes applying a write-back technique between a host server and a data store, accessing a dirty data in a cache memory during a power outage. The method may apply an algorithm for efficiently offloading the dirty data to a non-volatile storage device during the power outage. In addition the method may apply the algorithm to efficiently transfer the dirty data from the non-volatile storage device to the data store when power is restored.

System and method to preserve and recover unwritten data present in data cache of a disk subsystem across power outages

A system includes a data storage device, a controller coupled with the data storage device, a backup device coupled with the controller for backing up a modified portion of data and volatile memory metadata stored by the controller, and a backup power source for powering the controller. The controller includes a pre-specified region of volatile memory for storing backup device metadata for managing a modified portion of data, the metadata comprising one or more intents corresponding to modified data written back to the data storage device. The controller is configured to invalidate the one or more intents. During a restore operation, the controller is configured to store the backup device metadata in the pre-specified region of volatile memory when a charge on the backup power source is at least a minimum threshold charge and to store the updated backup device metadata in the backup device during an interruption of power.

Growing community of inventors

Bengaluru, India

Arindam Banerjee