Google Inc.

X Development LLC

Altera Corporation

Pts Corporation

Gossett and Gunter, Inc.

Michial Allen Gunter

Carroll Philip Gossett

Yuan Yuan

Kevin Yu

Charles Henry Leichner, Iv

Tammo Spalink

C Philip Gossett

David James Carmichael

Clifford L Biffle

Emmanouil Koukoumidis

Reiner Alwyn Pope

Xiaofang Jiang

David B Kita

Hal Ronald Varian

Denis Baylor

Oliver W Shih

Charles Ross

Herman Henry Schmit

Paul Kingsley Rodman

Lukasz Lew

Pavel Krajcevski

Brian Foley

Caroll Philip Gossett

Xiaofan Jiang

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for obtaining a first schedule, for a first hardware block of an integrated circuit device, where the first schedule identifies a first set of operations to be performed by the first hardware block. Obtaining a second schedule for a second hardware block of the integrated circuit device, where the second schedule identifies a second set of operations to be performed by the second hardware block and where operations of the second schedule are coordinated with operations of the first schedule such that the first schedule triggers the first hardware block to send data to the second block at a first pre-scheduled value of a counter, and the second schedule triggers the second hardware block to accept the data at an input at a second pre-scheduled value of the counter that is after the first pre-scheduled value. Performing, by the first hardware block, the first set of operations according to the first schedule, and performing, by the second hardware block, the second set of operations according to the second schedule.

Explicit scheduling of on-chip operations

A method includes: receiving control data at a first data selector of a plurality of data selectors, in which the control data comprises (i) a configuration registry address specifying a location in a configuration state registry and (ii) configuration data specifying a circuit configuration state of a circuit element of a computational circuit; transferring the control data, from the first data selector, to an entry in a trigger table registry; responsive to a first trigger event occurring, transferring the configuration data to the location in the configuration state registry specified by the configuration registry address; and updating a state of the circuit element based on the configuration data.

Control of deterministic machine learning systems using trigger tables and configuration state registries

An application specific integrated circuit (ASIC) chip includes: a systolic array of cells; and multiple controllable bus lines configured to convey data among the systolic array of cells, in which the systolic array of cells is arranged in multiple tiles, each tile of the multiple tiles including 1) a corresponding subarray of cells of the systolic array of cells, 2) a corresponding subset of controllable bus lines of the multiple controllable bus lines, and 3) memory coupled to the subarray of cells.

Application specific integrated circuit accelerators

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining, for each pair of adjacent chips in a plurality of chips connected in a series-ring arrangement of a semiconductor device, a corresponding loop latency for round trip data transmissions between the pair of chips. Identifying, from among the loop latencies, a maximum loop latency. Determining a ring latency for a data transmission originating from a chip of the plurality chips to be transmitted around the series-ring arrangement and back to the chip. Comparing half of the maximum loop latency to one N-th of the ring latency, where N is the number of chips in the plurality of chips, and storing the greater value as an inter-chip latency of the semiconductor device, the inter-chip latency representing an operational characteristic of the semiconductor device.

Synchronization in multi-chip systems

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for sharding dataflow graphs for a device having multiple synchronous tiles. One of the methods includes receiving a representation of a dataflow graph comprising a plurality of nodes that each represent respective matrix operations to be performed by a device having a plurality synchronous tiles. Candidate allocations of respective portions of the dataflow graph to each tile of the plurality of synchronous tiles are evaluated according to one or more resource constraints of the device. One of the candidate allocations is selected based on evaluating each candidate allocation.

Sharding for synchronous processors

A method of configuring an integrated circuit including multiple hardware tiles, includes: establishing a data forwarding path through the multiple hardware tiles by configuring each hardware tile, except for a last hardware tile, of the multiple hardware tiles to be in a data forwarding state, in which configuring each hardware tile, except for the last hardware tile, to be in a forwarding state includes installing a respective forwarding state counter specifying a corresponding predefined length of time that the hardware tile is in the data forwarding state; supplying, along the data forwarding path, each hardware tile of the plurality of hardware tiles with a respective program data packet comprising program data for the hardware tile; and installing, for each hardware tile of the multiple hardware tiles, the respective program data.

Initializing on-chip operations

A tile including circuitry for use with machine learning models, the tile including: a first computational array of cells, in which the computational array of cells is a sub-array of a larger second computational array of cells; local memory coupled to the first computational array of cells; and multiple controllable bus lines, in which a first subset of the multiple controllable bus lines include multiple general purpose controllable bus lines couplable to the local memory.

Aspects of the disclosure include methods, systems, and apparatus, including computer-readable storage media for multi-mode integrated circuits with balanced energy consumption. A method includes determining, by one or more processors and based at least on a maximum energy threshold for planned multi-mode system having one or more processing units, a respective number of operations that can be performed per clock cycle by the processing units for each operating mode. The system is configured to consume the same amount of energy per clock cycle in each operating mode, but perform more operations in operating modes corresponding to operations performed on smaller bit-width operands.

Multi-mode integrated circuits with balanced energy consumption

An application specific integrated circuit (ASIC) chip includes: a systolic array of cells; and multiple controllable bus lines configured to convey data among the systolic array of cells, in which the systolic array of cells is arranged in multiple tiles, each tile of the multiple tiles including 1) a corresponding sub array of cells of the systolic array of cells, 2) a corresponding subset of controllable bus lines of the multiple controllable bus lines, and 3) memory coupled to the subarray of cells.

Growing community of inventors

Oakland, CA, United States of America

Michial Allen Gunter