Netapp, Inc.

Emc IP Holding Company LLC

Emc Corporation

Network Appliance, Inc.

Darren Sawyer

Swaminathan Ramany

Kaladhar Voruganti

Gokul Soundararajan

Lakshmi Narayanan Bairavasundaram

Vipul Mathur

Surendar Chandra

Manpreet Singh

Michael Cao

Raymond Luk

Dan Truong

Kesari Mishra

Alexander Sideropoulos

Richard E Honicky, Jr

The present invention addresses encryption systems and methods in the de-duplication of data in a multi-tenant environment. The system provides isolation between tenants' stored data and the storage system. Tenant keys are assigned to tenants. The storage system stores raw data objects backed up for the tenants and fingerprints, corresponding to the data objects, in a single use key encrypted format. Fingerprints are wrapped with a storage system key held by the storage system. A request is received to retrieve data backed up for a tenant. The request includes fingerprints corresponding to the data objects to retrieve, and a tenant key, the fingerprints being in the single use key encrypted format and wrapped with the tenant key. The received fingerprints are unwrapped using the tenant key to retrieve data objects corresponding to the received fingerprints. The data objects are transmitted to the tenant and the tenant key is removed.

Data encryption in a de-duplicating storage in a multi-tenant environment

Systems and methods are disclosed for scheduling a plurality of tasks for execution on one or more processors. An example method includes obtaining a counter value of a counter. The method also includes for each work queue of a plurality of work queues, identifying an execution period of the respective work queue and comparing a counter value to an execution period of the respective work queue. Each work queue includes a set of tasks and is defined by an execution period at which to run the respective set of queued tasks. The method further includes selecting, based on the comparing, a subset of the plurality of work queues. The method also includes scheduling a set of tasks of slower frequency queued in a selected work queue for execution on one or more processors before a set of tasks queued in a non-selected work queue. The work items may be scheduled in O(1) because the design inherently prioritizes the tasks based on the urgency of their completion, and may do so by resetting a work queue pointer.

Method and system for scheduling repetitive tasks in O(1)

The present invention addresses encryption systems and methods in the de-duplication of data in a multi-tenant environment. The system provides isolation between tenants' stored data and the storage system. The tenants' data is broken down into many smaller raw data items. Fingerprints are generated for the raw data and compared to fingerprints of raw data previously stored on the storage system. The raw data and fingerprint are encrypted with a single use key (SUK) by the storage system. The SUK encrypted fingerprint is wrapped with a storage system key and stored with other fingerprints. The SUK encrypted fingerprint is also returned to the tenants and wrapped with a tenant key. The use of tenant key wraps allows the tenant data to be protected and confidential to each tenant but allows the raw data to be shared by all tenants.

Collaborative management of shared resources is implemented by a storage server receiving, from a first resource manager, notification of a violation for a service provided by the storage server or device coupled to the storage server. The storage server further receives, from each of a plurality of resource managers, an estimated cost of taking a corrective action to mitigate the violation and selects a corrective action proposed by one of the plurality of resource managers based upon the estimated cost. The storage server directs the resource manager that proposed the selected corrective action to perform the selected corrective action.

Collaborative management of shared resources

Collaborative management of shared resources selects corrective action based on normalized cost

A method and an apparatus to manage data in a networked storage server have been disclosed. In one embodiment, the method includes providing a single logical layer between a number of logical subsystems in a storage server and a number of data management agents in the storage server, wherein the data management agents are operable to service requests from a number of client machines following different data management protocols, and exporting a set of application programming interfaces (APIs) from the single logical layer to interface with the data management agents.

Method and an apparatus to provide access to operational data in a storage server via multiple data management protocols

Available performance capacities of storage servers and storage devices in a storage infrastructure are determined using a plurality of metrics, wherein each of the storage devices is managed by one of the storage servers. Each aggregate is hosted by a storage server and includes multiple storage devices. A relationship between the plurality of metrics is analyzed. An aggregate from which to allocate storage capacity to a volume from a plurality of aggregates is selected based on the available performance capacities and the relationship between the plurality of metrics. The selection is performed without information about properties of the volume. Storage capacity of the selected aggregate is automatically allocated to the volume. After said allocation, the plurality of aggregates is approximately load balanced.

Load balanced storage provisioning

A method and a system for modeling file system operation streams of a computing environment. File system operations of the file system operation streams are workloads produced by the devices of the computing environment. For example, a device of the computing environment produces the workloads, which are requests for information from a storage system of the computing environment. The requests are captured and form traces. The traces are then used as input to a model of the computing environment. The model is iteratively trained by using the traces to learn the dynamic behavior of the computing environment. Specifically, the model is trained to identify sequences of file system operations that are legal. Legal sequences can then be generated in realistic workloads that simulate the workloads produced by the devices of the computing environment to a high degree of probability.

Growing community of inventors

San Carlos, CA, United States of America

Darren Sawyer