Artificial Intelligence-Enabled Search for a Storage System

Abstract

The present disclosure describes apparatuses and methods for artificial intelligence-enabled search of a storage system. In some aspects, a metadata manager of a storage system receives a label of an object that an AI engine detects in data stored in the storage system. The metadata manager creates, in a relational section of a metadata database, an entry for the detected object with an identifier of the label of the detected object and an address of a node corresponding to the detected object. The metadata manager also creates, in a navigational portion of the metadata database and with the address of the detected object, the node that includes a reference to a relative node of another object and a weight of a relationship between the node and the relative node. By so doing, the metadata database may be searched based on weighted relationships between various nodes, thereby enabling contextual or implicit search of data in the storage system.

What is claimed is: 
     
         1 . A method for artificial intelligence-enabled search of a storage system, comprising:
 processing, with an artificial intelligence engine implementing an artificial intelligence model, data of a storage system to provide a set of respective labels for objects detected in the data by the artificial intelligence engine;   constructing, based on the set of respective labels, a metadata database comprising a weighted graph of nodes, each of the nodes corresponding to one of the objects detected in the data, the nodes connected to others of the nodes via respective weighted links, each of the respective weighted links having a weight value indicative of a contextual relationship between the object to which the node corresponds and another of the objects to which one of the linked nodes corresponds; and   enabling search for the objects detected in the data based on one or more tables of the metadata database that comprise, for each of the detected objects:
 an identifier associated with the respective label for the object; 
 an address reference for the node in the weighted graph that corresponds to the object; and 
 one or more of the weight values that are indicative of the contextual relationship between the object to which the node corresponds and one or more other objects in the data that correspond to other linked nodes connected to the node of the object via the respective weighted links of the weighted graph. 
   
     
     
         2 . The method of  claim 1 , wherein the artificial intelligence model is a first artificial intelligence model, the set of respective labels is a first set of respective labels, and the method further comprises:
 processing, with the artificial intelligence engine implementing a second artificial intelligence model, the data of the storage system to provide a second set of respective labels for the objects detected in the data; and   altering, based on the second set of respective labels, one of:
 at least some of the weighted links that connect the nodes in the weighted graph of the metadata database; or 
 at least some of the weight values of the respective weighted links between the nodes that are indicative of the contextual relationships between the respective objects to which the nodes correspond. 
   
     
     
         3 . The method of  claim 2 , wherein altering at least some of the weighted links that connect the nodes in the weighted graph comprises unlinking, based on the second set of respective labels for the objects, at least two of the nodes that correspond to respective objects that no longer have a contextual relationship. 
     
     
         4 . The method of  claim 2 , wherein altering at least some of the weight values of the weighted links comprises redefining one or more of the weight values of the weighted links that indicate the contextual relationship between the objects to which the nodes correspond. 
     
     
         5 . The method of  claim 1 , further comprising:
 receiving, from the artificial intelligence engine implementing the artificial intelligence model, an indication with at least one of the respective labels that indicates a confidence level of the artificial intelligence engine in detecting at least one of the objects in the data.   
     
     
         6 . The method of  claim 1 , wherein the one or more tables of the metadata database further comprise, for each of the detected objects, an address or path to a location in the data of the storage system at which the object was detected. 
     
     
         7 . The method of  claim 1 , further comprising constructing the one or more tables of the metadata database that comprise the identifiers associated with the respective labels for the objects, the address references for the nodes that correspond to the objects, and the one or more weight values indicative of the contextual relationships between the object detected in the data. 
     
     
         8 . The method of  claim 1 , wherein enabling search for the detected objects based on the one or more tables of the metadata database is effective to enable contextual or implicit search of the data stored by the storage system. 
     
     
         9 . The method of  claim 1 , wherein:
 the storage system comprises at least one of a solid-state drive (SSD), a hard disk drive (HDD), or an aggregate array of storage media drives; and   the method is implemented by a storage controller of the SSD, a storage controller of the HDD, or a controller that provides a host interface for the aggregate array of storage media drives.   
     
     
         10 . An apparatus comprising:
 an interface to receive data from a host;   storage media configured to store the data received from the host;   a controller configured to enable access to the data stored on the storage media;   a metadata manager configured to:
 process, with an artificial intelligence engine associated with the apparatus, the data stored to the storage media to provide a set of respective labels for objects detected in the data by the artificial intelligence engine; 
 construct, based on the set of respective labels, a metadata database comprising a weighted graph of nodes, each of the nodes corresponding to one of the objects detected in the data, the nodes connected to others of the nodes via a respective weighted link, each of the respective weighted links having a weight value indicative of a contextual relationship between the object to which the node corresponds and another of the objects to which one of the linked nodes corresponds; and 
 enable search for objects detected in the data based on one or more tables of the metadata database that include, for each of the detected objects:
 an identifier associated with the respective label for the object; 
 an address reference for the node in the weighted graph that corresponds to the object; and 
 one or more of the weight values that are indicative of the contextual relationship between the object to which the node corresponds and one or more other objects in the data that correspond to other linked nodes connected to the node of the object via the respective weighted links of the weighted graph. 
 
   
     
     
         11 . The apparatus of  claim 10 , wherein the artificial intelligence model is a first artificial intelligence model, the set of respective labels is a first set of respective labels, and the metadata manager is further configured to:
 process, with the artificial intelligence engine implementing a second artificial intelligence model, the data stored to the storage media to provide a second set of respective labels for the objects detected in the data; and   alter, based on the second set of respective labels, one of:
 at least some of the weighted links that connect the nodes in the weighted graph of the metadata database; or 
 at least some of the weight values of the respective weighted links between the nodes that are indicative of the contextual relationships between the respective objects to which the nodes correspond. 
   
     
     
         12 . The apparatus of  claim 11 , wherein to alter at least some of the weighted links that connect the nodes in the weighted graph, the metadata manager is further configured to unlink, based on the second set of respective labels for the objects, at least two of the nodes that correspond to respective objects that no longer have a contextual relationship. 
     
     
         13 . The apparatus of  claim 11 , wherein to alter at least some of the weight values of the weighted links, the metadata manager is further configured to redefine one or more of the weight values of the weighted links that indicate the contextual relationship between the objects to which the nodes correspond. 
     
     
         14 . The apparatus of  claim 11 , wherein the metadata manager is further configured to construct the one or more tables of the metadata database that comprise the identifiers associated with the respective labels for the objects, the address references for the nodes that correspond to the objects, and the one or more weight values indicative of the contextual relationships between the object detected in the data. 
     
     
         15 . The apparatus of  claim 10 , wherein the one or more tables of the metadata database further comprise, for each of the detected objects, an address or path to a location in the data of the storage media at which the object was detected. 
     
     
         16 . A System-on-Chip (SoC) comprising:
 an interface to storage media of a storage system;   an interface to a host from which data is received for writing to the storage media;   a hardware-based processor;   a memory storing processor-executable instructions that, responsive to execution by the hardware-based processor, implement a metadata manager to:
 process, with an artificial intelligence engine associated with the SoC, the data stored to the storage media to provide a set of respective labels for objects detected in the data by the artificial intelligence engine; 
 construct, based on the set of respective labels, a metadata database comprising a weighted graph of nodes, each of the nodes corresponding to one of the objects detected in the data, the nodes connected to others of the nodes via a respective weighted link, each of the respective weighted links having a weight value indicative of a contextual relationship between the object to which the node corresponds and another of the objects to which one of the linked nodes corresponds; and 
 enable search for objects detected in the data based on one or more tables of the metadata database that include, for each of the detected objects:
 an identifier associated with the respective label for the object; 
 an address reference for the node in the weighted graph that corresponds to the object; and 
 one or more of the weight values that are indicative of the contextual relationship between the object to which the node corresponds and one or more other objects in the data that correspond to other linked nodes connected to the node of the object via the respective weighted links of the weighted graph. 
 
   
     
     
         17 . The SoC of  claim 16 , wherein the artificial intelligence model is a first artificial intelligence model, the set of respective labels is a first set of respective labels, and the metadata manager is further implemented to:
 process, with the artificial intelligence engine implementing a second artificial intelligence model, the data stored to the storage media to provide a second set of respective labels for the objects detected in the data; and   alter, based on the second set of respective labels, one of:
 at least some of the weighted links that connect the nodes in the weighted graph of the metadata database; or 
 at least some of the weight values of the respective weighted links between the nodes that are indicative of the contextual relationships between the respective objects to which the nodes correspond. 
   
     
     
         18 . The SoC of  claim 16 , wherein to alter at least some of the weighted links that connect the nodes in the weighted graph, the metadata manager is further implemented to unlink, based on the second set of respective labels for the objects, at least two of the nodes that correspond to respective objects that no longer have a contextual relationship. 
     
     
         19 . The SoC of  claim 16 , wherein the one or more tables of the metadata database further comprise, for each of the detected objects, an address or path to a location in the data of the storage system at which the object was detected. 
     
     
         20 . The SoC of  claim 16 , wherein the artificial intelligence engine is embodied at least in part as hardware of the SoC, executable code that is stored on the memory of the SoC, or executable code stored to another memory that is accessible by the hardware-based processor of the SoC.