Magic Technologies, Inc.

Headway Technologies, Incorporated

Marvell International Limited

Palo Alto Networks, Inc.

Juniper Networks, Inc.

Advanced Numicro Systems, Inc.

Wei Cao

Witold Kula

Cheng Tzong Horng

Chyu-Jiuh Torng

Liubo Hong

Tom Zhong

Dao-Chen Zheng

Ru-Ying Tong

Rongfu Xiao

Chyu Jiuh Torng

Wai-Ming Johnson Kan

Terry Kin Ting Ko

Chyu-Jinh Torng

Shu Li

Chen Liu

William Redington Hewlett, Ii

Chen-Jung Chien

Yifei Zhang

Hongying Sheng

Shangde Zhou

Ruying Tong

Po-Kang Wang

Yee-Chung Fu

Guomin Mao

Jun Chen

Xia Zhu

Xuejun Wu

Terry Ko

A URL categorization query is received. The URL categorization query includes at least one URL. The URL is used to determine a set of data distribution keys. A distributed key-value data store is queried using at least one data distribution key included in the determined set of data distribution keys. Categorization information is returned. The returned URL categorization information can be used to enforce policies.

Communicating URL categorization information

A URL and a categorization associated with the URL are received. The received URL comprises a hostname portion and a path portion. The hostname portion of the received URL comprises a plurality of segments separated by one or more delimiters. A key associated with the received URL is determined. An operation is performed on a database using the determined key. Examples of such operations include inserting the categorization into the database, changing a value associated with the key in the database, removing a key-value pair from the database, and querying the database.

Key-value storage for URL categorization

A URL categorization query is received. The URL categorization query includes at least one URL. The URL is used to determine a set of keys. A database is queried using the determined set of keys. Categorization information is returned. The returned URL categorization information can be used to enforce policies.

A URL and a categorization associated with the URL are received. A key associated with the received URL is determined. An operation is performed on a database using the determined key. Examples of such operations include inserting the categorization into the database, changing a value associated with the key in the database, removing a key-value pair from the database, and querying the database.

Systems, methods, and apparatus are provided for iteratively decoding a codeword. Once a codeword is received, the codeword is processed to generate an incremental hard decision value and a log likelihood ratio amplitude value. These values are generated by processing the codeword using a soft output Viterbi algorithm. A faulty symbol in the codeword is identified. A complete hard decision value is generated using the incremental hard decision value. The LLR amplitude value and complete hard decision value corresponding to the identified faulty symbol are selectively provided to a decoder and the decoder uses these values to decode the codeword.

Implementation of LLR biasing method in non-binary iterative decoding

In general, the invention is directed to techniques for scheduling resource access within an intermediate network device. For example, as described herein, a device receives packets for a plurality of sessions that include application-layer data for the sessions. The device determines a weight for each of the plurality of sessions and, during periods of resource congestion, selects one or more sessions for additional resource allocation based on the respective weights of the sessions. The device allocates additional memory resources to selected sessions to enable further buffering of application-layer data such that the device may apply the service to multiple sessions concurrently despite the resource congestion.

Network services resource management

A MTJ in an MRAM array is disclosed with a composite free layer having a lower crystalline layer contacting a tunnel barrier and an upper amorphous layer for improved bit switching performance. According to one embodiment, the amorphous layer has a NiFeM/NiFeMconfiguration where Mand Mare Mg, Hf, Zr, Nb, or Ta, and M1 is unequal to M2. The crystalline layer is Fe, Ni, or FeB with a thickness of at least 6 Angstroms that affords a high magnetoresistive ratio. The Mand Melements in the NiFeMand NiFeMlayers each have a content of 5 to 30 atomic %. The NiFeM/NiFeMconfiguration substantially reduces bit line switching current and number of shorted bits. In an alternative embodiment, the crystalline layer may be a Fe/NiFe bilayer. Annealing at 300&#xb0; C. to 360&#xb0; C. provides a high magnetoresistive ratio of about 150%.

Magnetic tunnel junction for MRAM applications

A MTJ in an MRAM array is disclosed with a composite free layer having a lower crystalline layer contacting a tunnel barrier and an upper amorphous NiFeX layer for improved bit switching performance. The crystalline layer is Fe, Ni, or FeB with a thickness of at least 6 Angstroms which affords a high magnetoresistive ratio. The X element in the NiFeX layer is Mg, Hf, Zr, Nb, or Ta with a content of 5 to 30 atomic %. NiFeX thickness is preferably between 20 to 40 Angstroms to substantially reduce bit line switching current and number of shorted bits. In an alternative embodiment, the crystalline layer may be a Fe/NiFe bilayer. Optionally, the amorphous layer may have a NiFeM/NiFeMconfiguration where Mand Mare Mg, Hf, Zr, Nb, or Ta, and M1 is unequal to M2. Annealing at 300&#xb0; C. to 360&#xb0; C. provides a high magnetoresistive ratio of about 150%.

A magnetic tunneling junction (MTJ) in an MRAM array is disclosed with a composite free layer having a FL1/FL2/FL3 configuration where FL1 and FL2 are crystalline magnetic layers and FL3 is an amorphous NiFeX layer for improved bit switching performance. FL1 layer is CoFe which affords a high magnetoresistive (MR) ratio when forming an interface with a MgO tunnel barrier. FL2 is Fe to improve switching performance. NiFeX thickness where X is Hf is preferably between 20 to 40 Angstroms to substantially reduce bit line switching current and number of shorted bits. Annealing at 330&#xb0; C. to 360&#xb0; C. provides a high MR ratio of 190%. Furthermore, low Hc and Hk are simultaneously achieved with improved bit switching performance and fewer shorts without compromising other MTJ properties such as MR ratio. As a result of high MR ratio and lower bit-to-bit resistance variation, higher reading margin is realized.

Composite free layer within magnetic tunnel junction for MRAM applications

A multi-layered bottom electrode for an MTJ device on a silicon nitride substrate is described. It comprises a bilayer of alpha tantalum on ruthenium which in turn lies on a nickel chrome layer over a second tantalum layer.

Bottom electrode for MRAM device

A MTJ in an MRAM array is disclosed with a composite free layer having a lower crystalline layer contacting a tunnel barrier and an upper amorphous NiFeX layer for improved bit switching performance. The crystalline layer is Fe, Ni, or FEB with a thickness of at least 6 Angstroms which affords a high magnetoresistive ratio. The X element in the NiFeX layer is Mg, Hf, Zr, Nb, or Ta with a content of 5 to 30 atomic % NiFeX thickness is preferably between 20 to 40 Angstroms to substantially reduce bit line switching current and number of shorted bits. In an alternative embodiments, the crystalline layer may be a Fe/NiFe bilayer. Optionally, the amorphous layer may have a NiFeM/NiFeMconfiguration where Mand Mare Mg, Hf, Zr, Nb, or Ta, and M1 is unequal to M2. Annealing at 300&#xb0; C. to 360&#xb0; C. provides a high magnetoresistive ratio of about 150%.

Reading margin is improved in a MTJ designed for MRAM applications by employing a pinned layer with an AP2/Ru/AP1 configuration wherein the AP1 layer is a CoFeB/CoFe composite and by forming a MgO tunnel barrier adjacent to the CoFe AP1 layer by a sequence that involves depositing and oxidizing a first Mg layer with a radical oxidation (ROX) process, depositing and oxidizing a second Mg layer with a ROX method, and depositing a third Mg layer on the oxidized second Mg layer. The third Mg layer becomes oxidized during a subsequent anneal. MTJ performance may be further improved by selecting a composite free layer having a Fe/NiFeHf or CoFe/Fe/NiFeHf configuration where the NiFeHf layer adjoins a capping layer in a bottom spin valve configuration. As a result, read margin is optimized simultaneously with improved MR ratio, a reduction in bit line switching current, and a lower number of shorted bits.

Two methods of fabricating a MEMS scanning mirror having a tunable resonance frequency are described. The resonance frequency of the mirror is set to a particular value by mass removal from the backside of the mirror during fabrication.

Process for MEMS scanning mirror with mass remove from mirror backside

A system including a first clock module, a second clock module and an adjustment module. The first clock module is configured to generate a first clock signal having a first frequency. The second clock module is configured to, based on the first clock signal, generate a second clock signal for an integrated circuit. An amount of current drawn by the integrated circuit is based on a second frequency of the second clock signal. An adjustment module is configured to receive an enable signal indicating whether the integrated circuit is being powered ON. In response to the enable signal indicating the integrated circuit is being powered ON, the adjustment module (i) determines a predetermined frequency, and (ii) generate a control signal based on the first clock signal. The control signal adjusts the second frequency of the second clock signal to be between the predetermined frequency and the first frequency.

Growing community of inventors

San Jose, CA, United States of America

Wei Cao