International Business Machines Corporation

Microsoft Technology Licensing, LLC

Lucent Technologies Inc.

Alcatel-lucent USA Inc.

Broadlight Ltd.

Joseph Naor

Yigal Bejerano

Srikanth Kandula

Liane Lewin-Eytan

Ishai Menache

Sriram S Rao

Avrilia Floratou

Ashvin Agrawal

Christopher William Douglas

Virajith Jalaparti

Mainak Ghosh

Gidon Gershinsky

David Ivancovsky

Miriam Allalouf

Mark Anthony Shawn Smith

Alexander Sprintson

Nicole S Immorlica

Raphael J Rom

Methods, systems, apparatuses, and computer program products are provided for prefetching data. A workload analyzer may identify job characteristics for a plurality of previously executed jobs in a workload executing on a cluster of one or more compute resources. For each job, identified job characteristics may include identification of an input dataset and an input bandwidth characteristic for the input dataset. A future workload predictor may identify future jobs expected to execute on the cluster based at least on the identified job characteristics. A cache assignment determiner may determine a cache assignment that identifies a prefetch dataset for at least one of the future jobs. A network bandwidth allocator may determine a network bandwidth assignment for the prefetch dataset. A plan instructor may instruct a compute resource of the cluster to load data to a cache local to the cluster according to the cache assignment and the network bandwidth assignment.

Cache and I/O management for analytics over disaggregated stores

A method of locating traffic reduction entities in a communication network, optionally where at least one reduction entity is located within each network path which allows traffic flow between endpoints and central stations. The method comprises assigning a plurality of allocation cost values each estimating cost of locating of one of a plurality of traffic reduction entities on one of a plurality of intermediate nodes of a communication network, calculating a plurality of excess load cost values each calculated according to excess load cost between two adjacent intermediate nodes of the plurality of intermediate nodes, identifying a node set selected from the plurality of intermediate nodes using a target function based on the plurality of cost values and the plurality of excess load cost values, and outputting instructions for locating the plurality of traffic reduction entities on the communication network according to the node set.

Locating traffic reduction utilities in communication networks

A method of applying a quality-aware volume reduction to communication networks is provided herein. The method includes the following stages: deriving, from the network: a bandwidth capacity limit of the links of the network, and an initial flow rate for each flow transmitted via nodes through the network; obtaining from each one of the nodes, a gain value which corresponds to a quantitative data demand requested by the node and given in terms of data quality; and calculating, for each one of the nodes, a data reduction ratio, that achieves: (i) a maximal overall revenue associated with the network, wherein the revenue for a node is calculated with respect to the quality of the data that arrives to the node, and (ii) an optimal bandwidth allocation over the links with respect to the gains and the data quality at each node, and to the limited bandwidth capacity of the links.

Applying quality aware volume reduction to communication networks

A method comprises allocating a respective restoration path to protect each of a plurality of primary paths in a network. Each primary path has a plurality of nodes and at least two subpaths connecting the nodes of the primary path. Each restoration path includes at least one link. Each link protects at least one of the subpaths of the primary path. Each of the subpaths of the primary path is protected. None of the links in the restoration path is included in the corresponding primary path. At least one of the links is shared among at least two restoration paths corresponding to at least two of the primary paths.

Method and system for shared backup allocation in networks based on partial information

An upstream bandwidth allocation scheduler is provided for use within a Line Termination (LT) unit in a point to multi-point data communications network in order to service multiple Network Termination (NT) units. The scheduler first allocates bandwidth among the several categories of data handled by the network (e.g., voice, IP, and ATM data), and then allocates bandwidth among the several users for each category within the network. A function &#x192; is defined, as part of a 'min-max weighted fairness' scheme, in order to allocate surplus bandwidth among both the categories and users of the network. The scheduler of the present invention, while allocating upstream bandwidth, also supports various quality of service (QoS) categories to subscribers of the network.

Method of providing QoS and bandwidth allocation in a point to multi-point network

Rounded solutions of a linear program are used to partition or group cells, such as cells in a wireless network, in order to provide efficient approximations of network updating and paging costs.

Methods and devices for grouping cells

Growing community of inventors

Haifa, Israel

Joseph Naor