Inventor

PENNER HARTMUT

US32 patents

⚠️ This page may combine multiple inventors who share the name “PENNER HARTMUT”. Patents are grouped by organization below to help distinguish them — per-person disambiguation is on the roadmap.

PENNER HARTMUT

IBM

Massively parallel neural inference computing elements

Providing remote, reliant and high performance PCI express device in cloud computing environments

Providing remote, reliant and high performance PCI express device in cloud computing environments

Method and system for generating and applying patches to a computer program concurrently with its execution

Method and system for applying patches to a computer program concurrently with its execution

Scheduler for mapping neural networks onto an array of neural cores in an inference processing unit

Runtime reconfigurable neural network processor core

Providing remote, reliant and high performance PCI express device in cloud computing environments

Extending existing storage devices in virtualized environments

Extending existing storage devices in virtualized environments

Data representation for dynamic precision in neural network cores

Networks for distributing parameters and data to neural network compute cores

Runtime reconfigurable neural network processor core

Instruction distribution in an array of neural network cores

Distributed, event-based computation using neuromorphic cores

Data distribution in an array of neural network cores

Massively parallel neural inference computing elements

Extending existing storage devices in virtualized environments

Hierarchical parallelism in a network of distributed neural network cores

Defect resistant designs for location-sensitive neural network processor arrays

Parallel computational architecture with reconfigurable core-level and vector-level parallelism

Extending existing storage devices in virtualized environments

Extending existing storage devices in virtualized environments

Compound instruction set architecture for a neural inference chip

Efficient and secure direct storage device sharing in virtualized environments

Efficient and secure direct storage device sharing in virtualized environments

Efficient and secure direct storage device sharing in virtualized environments

Efficient and secure direct storage device sharing in virtualized environments

Multi-agent instruction execution engine for neural inference processing

Selective multicast delivery on a bus-based interconnect

BACHER UTZ

PCI express multiplier device

FRANKE HUBERTUS

Method to embed a light-weight kernel in a full-weight kernel to provide a heterogeneous execution environment