17899531. MEMORY DEVICE WITH EMBEDDED DEEP LEARNING ACCELERATOR IN MULTI-CLIENT ENVIRONMENT simplified abstract (Micron Technology, Inc.)
MEMORY DEVICE WITH EMBEDDED DEEP LEARNING ACCELERATOR IN MULTI-CLIENT ENVIRONMENT
Organization Name
Inventor(s)
Dmitri Yudanov of Rancho Cordova CA (US)
Troy Allen Manning of Meridian ID (US)
MEMORY DEVICE WITH EMBEDDED DEEP LEARNING ACCELERATOR IN MULTI-CLIENT ENVIRONMENT - A simplified explanation of the abstract
This abstract first appeared for US patent application 17899531 titled 'MEMORY DEVICE WITH EMBEDDED DEEP LEARNING ACCELERATOR IN MULTI-CLIENT ENVIRONMENT
Simplified Explanation
- Memory device stores data from host device - Memory device includes memory interface to host device - Processing devices perform computations for neural network using data from memories - Output of neural network stored in memory - Controller manages memory access to avoid conflicts - Communication with host device using DRAM bus protocol, including sending neural network output
Potential Applications
- Artificial intelligence - Machine learning - Data processing
Problems Solved
- Efficient memory access - Avoiding conflicts in memory usage - Streamlining communication between devices
Benefits
- Faster data processing - Improved memory management - Enhanced performance in neural network computations
Original Abstract Submitted
Systems, methods, and apparatus related to memory devices. In one approach, a memory device includes memories that store data received from a host device. The memory device includes a memory interface to the host device. The memory device further includes one or more processing devices to perform, using a portion of the data stored in one or more of the memories, computations for a neural network. An output of the neural network is stored in one of the memories. The memory device has a controller that controls memory access by the host device and the processing devices to avoid a conflict. The memory device communicates with the host device over the memory interface using a DRAM bus protocol. This communication includes sending the output of the neural network to the host device.