Hewlett-packard Development Company, L.P.

Hewlett Packard Enterprise Development LP

Other

Robert J Brooks

Blaine Douglas Gaither

Gregg Bernard Lesartre

Kevin Thomas McCarthy

Andrew Ray Wheeler

Benjamin Daniel Osecky

Alexander J Neudeck

David Fisk

Kathryn A Evertson

Brent Buchanan

Karl Joseph Bois

Robert J Blakely

Todd Byrne

Harold G Roberts

Dale P Devore

Benjamin D Osecky

In at least some examples, a computing node includes a processor and a local memory coupled to the processor. The computing node also includes a reflective memory bridge coupled to the processor. The reflective memory bridge maps to an incoming region of the local memory assigned to at least one external computing node and maps to an outgoing region of the local memory assigned to at least one external computing node.

Reflective memory bridge for external computing nodes

A state-retaining logic cell may include a plurality of inverters, an output node non-volatile (NVM) storage cell, and an input node NVM storage cell. The plurality of inverters may include a feed-forward inverter and a feed-back inverter disposed in a back-to-back arrangement. The output node NVM storage cell may include first and second terminals, where the first terminal is connected adjacent an output node of the feed-forward and the feed-back inverters, and the second terminal is connected to a programming rail. The input node NVM storage cell may include first and second terminals, where the first terminal is connected adjacent an input node of the feed-forward and the feed-back inverters, and the second terminal is connected to the programming rail.

State-retaining logic cell

A CPU architecture is proposed which flexibly allocates chip resources among threads. Execution units (microcores) are arranged in a ring. Instruction fetch units (front-ends) deposit instructions sequentially into storage elements within the microcores. Multiple front-ends can each feed segments of the ring; each such segment is a 'smart queue'. If, due to a sustained higher execution rate, a thread catches up to the next thread ahead of it, the slower thread steps aside and lets the faster thread play through. Other circumstances may lead to a thread consuming more than its usual share of resources, possibly even all of the microcores, for a time. The architecture has no instruction set dependencies; it is applicable to existing instruction set architectures and will speed up execution of them significantly as compared to conventional architectures.

CPU archtecture with highly flexible allocation of execution resources to threads

Techniques for updating data in a reflective memory region of a first memory device are described herein. In one example, a method for updating data in a reflective memory region of a first memory device includes receiving an indication that data is to be flushed from a cache device to the first memory device. The method also includes detecting a memory address corresponding to the data is within the reflective memory region of the first memory device and sending data from the cache device to the first memory device with a flush operation. Additionally, the method includes determining that the data received by the first memory device is modified data. Furthermore, the method includes sending the modified data to a second memory device in a second computing system.

Implementing coherency with reflective memory

A nonvolatile memory device is writable to a high resistance state and a low resistance state. The nonvolatile memory device may be heated to at least a threshold temperature, based on application of an alternating current (AC) signal, and may be written based on application of a voltage bias.

Writable device based on alternating current

According to an example, a method for storage device reading may include receiving an input signal indicative of a period of oscillation of a ring oscillator coupled to a storage device of a plurality of storage devices, and measuring the period of oscillation of the ring oscillator by a time-to-digital circuit. The method for storage device reading may further include determining a value of data stored in the storage device based on the measurement.

Apparatus and method for reading a storage device with a ring oscillator and a time-to-digital circuit

According to an example, a method for storage device write pulse control may include writing a storage device to a first polarity by driving a row address line (RAL) and a column address line (CAL) to an intermediate voltage level RCA for a cycle A. The RAL may be driven to a voltage level RB for a cycle B pulse duration, and the CAL may be maintained at RCA for the cycle B pulse duration. The RAL may be driven to a voltage level RC for a cycle C pulse duration, and the CAL may be driven to a voltage level CC for the cycle C pulse duration. The RAL may be driven to RCA, and the CAL may be driven to a voltage level CD for a cycle D pulse duration. The RAL may be maintained at RCA, and the CAL may be driven to RCA.

Storage device write pulse control

According to an example, multi-mode storage may include operating a first array including a first memory and a second array including a second memory in one or more modes of operation. The first memory may be a relatively denser memory compared to the second memory and the second memory may be a relatively faster memory compared to the first memory. The modes of operation may include a first mode of operation where the first array functions as the relatively denser memory compared to the second memory and the second array functions as the relatively faster memory compared to the first memory, a second mode of operation where the second array is operated as an automatic cache of a portion of a dataset, and a third mode of operation where a cache-tag functionality used to support the second mode of operation is instead used to provide a CAM.

Apparatus and method for multi-mode storage

Methods and apparatus are provided for use with data storage elements. A ring oscillator is coupled to a selected element within an array such that a feedback loop is defined. A period at oscillation for the ring oscillator is compared to a reference value. A data value stored within the selected element is determined accordingly. Stored data values remain essentially unaltered when accessed and read by way of the ring oscillator. Memory arrays having memristor or other storage elements can be used according to the present teachings.

Storage element reading using ring oscillator

System and methods of synchronizing reference frequencies are disclosed. In an exemplary implementation, a method may comprise providing separate reference frequencies for each of a plurality of operational components. The method may also comprise connecting the separate reference frequencies to one another in a modular, fault-tolerant circuit topology. The method may also comprise synchronizing the separate reference frequencies so that each of the operational components operate at the same frequency.

Systems and methods of synchronizing reference frequencies

In processing an instruction request, the invention determines whether the request is speculative or not based upon a bit field within the instruction. If the request is speculative, bus congestion and/or target memory is assessed for conditions and a decision is made, based on the conditions, as to whether or not to process the request. To facilitate the invention, certain bit fields within the instruction are encoded to identify the request as speculative or not. Additional bit fields may define a priority of a speculative request to influence the decision to process as based on the conditions. CPU architectures incorporating prefetch logic may be modified to recognize instructions encoded with speculation and priority identification fields to implement the invention in existing systems. Other logic, e.g., bus controllers and switches, may similarly process speculative requests to enhance system performance.

Method of optimization of CPU and chipset performance by support of optional reads by CPU and chipset

A method and apparatus for emulating hardware bus lock in a multi-architecture computer system includes a fault handler that acquires a semaphore reserved for bus lock and a semaphore that limits access to a page table. The fault handler includes an emulation module that sets a mode bit to prevent the bus lock and allows re-execution of the instruction that caused a request for a hardware bus lock. Using this method, the fault handler ensures a minimum disruption to operation of the computer system by restricting access to the least amount of computer system resources.

Transparent software emulation as an alternative to hardware bus lock

The builder's square aids in cutting sheet material, such as full sheets of plywood, and in particular reduces the time required for cutting. The square comprises a short leg and a long leg connected in an L-shape. The short leg is of greater height than the long leg so that it can bear against one edge of the sheet of plywood to be cut. There is a clamping member at the distal end of the long leg so that the square can be clamped to the plywood. The builder's square also provides improved accuracy and safety for the user.

Builder's square

A pair of cross-coupled inverters that hold a digital state are powered by supplies that also function as row select and column bit lines. A method of reading and writing the digital state of an individual cell, row of cells, or column of cells by manipulating these row-supply or column-supply lines. Reads and writes may be performed on either a row or column basis. A method for reading and logically OR'ing or AND'ing an entire row or column of cells. A method for querying on a row or column basis to function as a content addressable memory (CAM).

Growing community of inventors

Fort Collins, CO, United States of America

Robert J Brooks