International Business Machines Corporation

Google Inc.

Infinera Corporation

Anurag Sharma

Matthew Johnston

David Allan Pease

Rodolfo Enrique Alvizu Gomez

Maohua Lu

Mihail Corneliu Constantinescu

Abdullah Gharaibeh

Amit P Singh

Rene Marcel Schmogrow

Wenjie Jiang

Jon Earl Graham

Christina Vlachou

Jinjie He

Min Wen

Frank Bernhard Schmuck

Renu Tewari

Pin Zhou

Manoj Premanand Naik

Robert Michael Rees

Mark Andrew Smith

Steven R Welch

Wayne Curtis Hineman

Sanjeev Radhakrishnan

Veera W Deenadhayalan

Sigurdur Asgeirsson

Balaji Balasubramanian

Johann Tomas Sigurdsson

Sri Mohana Satya Srinivas Singamsetty

Christopher Liou

Tomas Gunnarsson

Wayne Gray

Amit Joshi

Michael Anthony Richmond

Finnur Thorarinsson

Sunil Namdeo Shilimkar

Ananta Iyengar

Robert Nygaard

Ray Bennett

Robert Withrow

Srinivasan Seetharaman

Tohru Hasagawa

Brian Biskeborn

Mahidhar Rajala

Allocating network resources to one or more signals that are to be conveyed over the network by calculating a transport capacity for a sublink of the network based on a spectral efficiency of at least one subpath included in the sublink, and allocating the sublink to at least one signal based on the calculated transport capacity.

Spectrum-aware cross-layer optimization

The subject matter described herein provides systems and techniques for a network planning and optimization tool that may allow for network capacity planning using key network failures for an arbitrary pair of network topology and demands. Performing network capacity planning with key network failures, instead of using other techniques, may avoid over-building the topology of a network. In particular, key network failures may be generated from the probabilistic failures, and the impact of these failures on a network may be computed. Expected flow availability SLO or a function thereof may be computed, using this information, and used by the tool to design a robust network. With an embedded flow availability calculation and updated risk framework, the capacitated cross-layer network topologies output by the tool may meet network demands/flows with their respective SLO type at the lowest cost.

Availability SLO-aware network optimization

Methods, systems, and apparatus, including computer-readable storage media, optimizing interior gateway protocol (IGP) metrics using reinforcement learning (RL) for a network domain. The system can receive a topology (G) of a network domain, a set of flows (F), and an objective function. The system can optimize, using reinforcement learning, the objective function based on the received topology and the one or more flows F. The system can determine updated IGP metrics based on the optimization of the objective function. The IGP metrics for the metric domain may be updated with the updated IGP metrics.

Assigning routing paths based on interior gateway protocol metric optimization

Determining an upgrade path from a starting topology to a target topology of a network is computationally intense and does not guarantee a steadily increasing usable capacity of the network at each stage within the upgrade path. The disclosed technology allows for a sequence of stages related to network upgrades to be generated. The technology ensures that networks can be upgraded in a sequential manner, where each step in the sequence does not violate service level objectives related to the network, ensures operational continuity of the network by users of the network, and ensures that the available network resources increase as the sequential upgrades are rolled out. The pathway determined is determined in a computationally efficient manner.

Sequenced capacity deployment for WAN and datacenter networks

Data deduplication for data storage tapes includes intercepting tape control commands for a single data storage tape. The intercepted tape control commands are modified for adding processing logic and parameters for placement of deduplicated file data on the single data storage tape. Deduplication metadata is written to a metadata portion of the single data storage tape. The deduplicated file data is written to a data portion of the single data storage tape based on the placement to increase read throughput for a deduplicated set of individual files and to reduce an average number of per-file gaps on the single data storage tape without re-duplicating deduplicated data for meeting optimization of individual file accesses.

Optimized data placement for individual file accesses on deduplication-enabled sequential storage systems

A method and apparatus for the efficient dynamic allocation of additional bandwidth in an optical virtual private network (OVPN) are described herein. A software defined network (SDN) controller may receive a request from an end-user of an OVPN for active bandwidth. Then, the SDN controller may determine whether the OVPN requires additional bandwidth to satisfy the request. As of the request, the additional bandwidth may be pre-deployed, but not yet activated for use. In an example, a path computation element (PCE) of the SDN controller may make this determination. Further, the SDN controller, on a condition that additional bandwidth is required, may determine whether additional bandwidth may be activated based on a policy of the end-user. As a result, the SDN controller, on a condition that additional bandwidth may be activated, may transmit an activation message to one or more line cards of the OVPN to activate the additional bandwidth.

Dynamic allocation of network bandwidth

Data deduplication for data storage tapes includes intercepting tape control commands for a single data storage tape. The intercepted tape control commands are modified by adding processing logic and parameters for placement of deduplicated file data on the single data storage tape. Deduplication metadata is written to a metadata portion of the single data storage tape. The deduplicated file data is written to a data portion of the single data storage tape based on the placement to increase read throughput for a deduplicated set of individual files and to reduce an average number of per-file gaps on the single data storage tape without re-duplicating deduplicated data for meeting optimization of individual file accesses.

Embodiments of the invention relate to receiving a write request that includes a write data and an address of a target block in tertiary storage. In response to the write request, a write-miss is detected at a cache located in persistent storage. Based on detecting the write-miss, the write data and associated metadata are written to a fast write storage location and the write request is marked as complete. In addition, the target block is retrieved from the address in the tertiary storage and stored in the cache. Contents of the fast write storage location are merged with the contents of the target block in the cache.

Fast write and management of persistent cache in a system that includes tertiary storage

Data deduplication for data storage tapes comprises determining the read throughput of a deduplicated set of individual files on a single data storage tape, and determining a placement of deduplicated file data on a single data storage tape to reduce an average number of per-file gaps on the tape. Deduplicated file data is placed on the single data storage tape based on said placement to increase an average read throughput for a deduplicated set of individual files.

A method for accessing data stored in a distributed caching storage system containing a home cluster and a secondary cluster is provided. A first copy of a file is stored on the home cluster and a second copy of the file is stored on the secondary cluster. The second copy of the file is associated with an inode data structure having a consistency attribute. An input/output request is received directed to the file and indicates that file is in an inconsistent state by updating the inode's consistency attribute. The first copy and the second copy of the file is updated according to the received input/output request and it is determined whether the first copy and the second copy were updated successfully. The maintaining of the inode's consistency attribute is indicative of the inconsistent state of the file.

Clustered and highly-available wide-area write-through file system cache

Embodiments of the invention relate to providing file system extended attribute support in an operating system. A request from an application to access extended attribute data in a multi-platform file system using an interface specified by the operating system is received. It is determined whether the specified interface supports an exchange of full content of the extended attribute data between the application and the multi-platform file system. Based on determining that the specified interface does not support an exchange of full content, the request is processed by an alternate interface. Based on the access being a read, the full content of the extended attribute data is retrieved and returned to the application using the alternate interface. Based on the access being a write, the full content of the extended attribute data is received from the application using the alternate interface and written to the multi-platform file system.

File system extended attribute support in an operating system with restricted extended attributes

The present invention provides a method and system of specifying and enforcing at least one run-time policy for at least one computer process executing on a computer system, where the computer system includes a computer operating system. In an exemplary embodiment, the method and system include (1) relating the policy with an executable file of the process, (2) associating the policy with a running instance of the process, and (3) enforcing the policy on the running instance.

Specifying and enforcing run-time policies for application processes being executed on a computer

Growing community of inventors

Mountain View, CA, United States of America

Anurag Sharma