Tenstorrent Inc.

Altera Corporation

Ivan Hamer

Ljubisa Bajic

Milos Trajkovic

Deshanand Pratap Singh

Terry Borer

Syed Gilani

Valavan Manohararajah

Stephen D Brown

Gabriel Quan

Paul McHardy

Tim Michael Vanderhoek

Kevin Chan

Lejla Bajic

Ian Chesal

Chris Sanford

Przemek Guzy

Steven Caranci

Karl Schabas

Alexander Grbic

Jimmy Kuo

Rebecca Katzin

Aleksandar Cejkov

Zvonko Vranesic

A processing core for the efficient execution of a directed graph is disclosed. The processing core includes a memory and a first and a second data tile stored in the memory. The first and second data tiles include a first and a second set of data elements stored contiguously in the memory. The processing core also includes metadata relationally stored with the first data tile in the memory. The processing core also includes an execution engine, a control unit, and an instruction. Execution of the instruction uses the execution engine, a first data element in the first set of data elements, and a second data element in the second set of data elements. The control unit conditions execution of the instruction using the metadata. A standard execution of the instruction generates a standard output. A conditional execution of the instruction operation generates a conditionally executed output.

Processing core with meta data actuated conditional graph execution

Processing cores with the ability to suppress operations based on a contribution estimate for those operations for purposes of increasing the overall performance of the core are disclosed. Associated methods that can be conducted by such processing cores are also disclosed. One such method includes generating a reference value for a composite computation. A complete execution of the composite computation generates a precise output and requires execution of a set of component computations. The method also includes generating a component computation approximation. The method also includes evaluating the component computation approximation with the reference value. The method also includes executing a partial execution of the composite computation using the component computation approximation to produce an estimated output. The method also includes suppressing the component computation, while executing the partial execution, based on the evaluation of the component computation approximation with the reference value.

Processing core with operation suppression based on contribution estimate

A processing core and associated methods for the efficient execution of a directed graph are disclosed. A disclosed processing core comprises a memory and a first data tile stored in the memory. The first data tile includes a first set of data elements and metadata stored in association with the first set of data elements. The processing core also comprises a second data tile stored in the memory. The second data tile includes a second set of data elements. The processing core also comprises an arithmetic logic unit configured to conduct an arithmetic logic operation using data from the first set of data elements and the second set of data elements. The processing core also comprises a control unit configured to evaluate the metadata and control the arithmetic logic unit to conditionally execute the arithmetic logic operation based on the evaluation of the metadata.

Processing core with metadata actuated conditional graph execution

Processing cores with the ability to suppress operations based on a contribution estimate for those operation for purposes of increasing the overall performance of the core are disclosed. Associated methods that can be conducted by such processing cores are also disclosed. One such method includes generating a reference value for a composite computation. A complete execution of the composite computation generates a precise output and requires execution of a set of component computations. The method also includes generating a component computation approximation. The method also includes evaluating the component computation approximation with the reference value. The method also includes executing a partial execution of the composite computation using the component computation approximation to produce an estimated output. The method also includes suppressing the component computation, while executing the partial execution, based on the evaluation of the component computation approximation with the reference value.

A method for designing a system on a target device utilizing field programmable gate arrays (FPGAs) includes identifying a group of components associated with a critical signal in the system. A first copy and a second copy of the group of components are generated where the first copy is driven by a first signal at a first state and the second copy is driven by a second signal at a second state. The system is configured to select an output of one of the first copy and the second copy in response to the critical signal.

Method and apparatus for performing layout-driven optimizations on field programmable gate arrays

A method for designing a system on a target device utilizing field programmable gate arrays is disclosed. A design is synthesized for the system. Components in the design are mapped onto resources on the target device. Placement locations are determined for the components on the target device. Components to replicate are identified in response to criticality determined from the placement locations.

Method and apparatus for performing logic replication in field programmable gate arrays

A method for positioning components of a system onto a target device utilizing programmable logic devices (PLDs) is disclosed. A first location on the target device for a first logic region having a first component is determined. Determined properties of the first logic region are preserved. The first logic region is integrated with a second logic region having a second component in view of the determined properties.

Method and apparatus for designing systems using logic regions

Growing community of inventors

Toronto, Canada

Ivan Hamer