Memory systems including examples of calculating hamming distances for neural network and data center applications

Abstract

Examples of systems and method described herein provide for the processing of image codes (e.g., a binary embedding) at a memory system including a Hamming processing unit. Such images codes may generated by various endpoint computing devices, such as Internet of Things (IoT) computing devices, Such devices can generate a Hamming processing request, having an image code of the image, to compare that representation of the image to other images (e.g., in an image dataset) to identify a match or a set of neural network results. Advantageously, examples described herein may be used in neural networks to facilitate the processing of datasets, so as to increase the rate and amount of processing of such datasets. For example, comparisons of image codes can be performed “closer” to the memory devices, e.g., at a processing unit having memory devices.

What is claimed is: 
     
         1 . A method comprising:
 responsive to request to calculate a Hamming distance among a plurality of images and a target image associated with an image code received from a host computing device, providing at least one memory access request to a plurality of memory devices to access information associated with the plurality of images stored at the plurality of memory devices; and   calculating a respective Hamming distance of a plurality of Hamming distances by comparing an image of the plurality of images to the image code.   
     
     
         2 . The method of  claim 1 , further comprising receiving the request to calculate the Hamming distance obtaining via a host bus. 
     
     
         3 . The method of  claim 2 , further comprising receiving the request to calculate the Hamming distance obtaining at Hamming control logic. 
     
     
         4 . The method of  claim 2 , further comprising receiving the request to calculate the Hamming distance obtaining at a processor. 
     
     
         5 . The method of  claim 1 , further comprising generating the image code based on the target image using a hash. 
     
     
         6 . The method of  claim 1 , further comprising receiving the image from an Internet of Things (IoT) computing device. 
     
     
         7 . The method of  claim 6 , wherein the IoT computing device comprises at least one of a camera, a smartphone device, or an image capture device. 
     
     
         8 . The method of  claim 1 , wherein the Hamming processing request is generated based on a neural network request to obtain image processing results using the plurality of Hamming distances. 
     
     
         9 . The method of  claim 1 , further comprising accessing, from the plurality of memory devices, a respective image code for an image of the plurality of images as the information associated with the plurality of images. 
     
     
         10 . The method of  claim 1 , further comprising hashing the at least one image of the plurality of images to generate at least one hashed image code for that respective image to calculate the respective Hamming distance of the plurality of Hamming distances. 
     
     
         11 . An apparatus comprising:
 Hamming control logic circuitry configured to, responsive to request to calculate a Hamming distance among a plurality of images and a target image associated with an image code received from a host computing device, provide at least one memory access request to a plurality of memory devices to access information associated with the plurality of images stored at the plurality of memory devices, wherein the Hamming control logic circuitry is further configured to calculate a respective Hamming distance of a plurality of Hamming distances by comparing an image of the plurality of images to the image code to.   
     
     
         12 . The apparatus of  claim 11 , wherein the Hamming control logic circuitry is further configured to receive the request to calculate the Hamming distance obtaining via a host bus. 
     
     
         13 . The apparatus of  claim 11 , wherein the Hamming control logic circuitry is further configured to receive the Hamming processing request via a PCIe bus. 
     
     
         14 . The apparatus of  claim 11 , further comprising a processor including the hamming control logic circuitry. 
     
     
         15 . The apparatus of  claim 11 , wherein the image code is generated based on the target image using a hash. 
     
     
         16 . The apparatus of  claim 11 , wherein the image is received from an Internet of Things (IoT) computing device. 
     
     
         17 . The apparatus of  claim 11 , wherein the IoT computing device comprises at least one of a camera, a smartphone device, or an image capture device. 
     
     
         18 . The apparatus of  claim 11 , wherein the Hamming processing request is generated at the host computing device based on a neural network request to obtain image processing results using the plurality of Hamming distances. 
     
     
         19 . The apparatus of  claim 11 , wherein the Hamming control logic circuitry is further configured to access, from the plurality of memory devices, a respective image code for an image of the plurality of images as the information associated with the plurality of images. 
     
     
         20 . The apparatus of  claim 11 , wherein the Hamming control logic circuitry is further configured to hash the at least one image of the plurality of images to generate at least one hashed image code for that respective image to calculate the respective Hamming distance of the plurality of Hamming distances.