Systems and methods for managing digital data in a fault tolerant matrix

Abstract

Aspects of the present disclosure relate to systems and methods for automatic management of digital data volumes logically maintained in a dynamically scalable fault tolerant matrix. The data volumes may be distributed across a cluster of connected server nodes included in a cloud computing architecture. A processing device in communication with the matrix ensure that read/write request may be serviced by the matrix to access the digital data maintained within the data volumes may be continuously accessed, regardless of data volume failure that are missing, offline, or in a failed state.

What is claimed is: 
     
         1 . A system for generating a fault tolerant matrix comprising:
 a cluster of computing nodes distributed throughout a communications network, respective computing nodes of the cluster of computing nodes maintaining a plurality of data volumes logically arranged in a matrix;   a processing device in operable communication with at least one computing node of the cluster of computing nodes to:
 receive a request requesting access to data stored in a particular data volume of the plurality of data volumes arranged in the matrix; and 
 determine whether the matrix is degraded; and 
 when the data volume is degraded, execute erasure coding algorithms to 
   regenerate the data, based on a row of the matrix in which the data volume is maintained.   
     
     
         2 . The system of  claim 1 , wherein to execute erasure coding algorithms based on a row comprises:
 when the request is a read request, determining whether at least two of the data volumes of the plurality of volumes have failed in a single row of the matrix;   when two of the data volumes of the plurality of data volumes have failed in a single row of the matrix, execute erasure coding algorithms to regenerate the data, based on portions of data included in all data volumes of the column of the matrix in which the particular data volume is maintained.   
     
     
         3 . The system of  claim 1 , wherein to execute erasure coding algorithms based on a row comprises:
 when the request is a read request, determining whether one of the data volumes of the plurality of volumes has failed in a single row of the matrix;   when one of the data volumes of the plurality of data volumes has failed in a single row, execute erasure coding algorithms to regenerate the data, based on portions of data included in all data volumes of the row of the matrix in which the particular data volume is maintained.   
     
     
         4 . The system of  claim 1 , wherein to execute erasure coding algorithms based on a row comprises:
 when the request is a write request:
 execute erasure coding algorithms to calculate new parity data, based on portions of data included in all data volumes of the row of the matrix in which the particular data volume is maintained; 
 execute erasure coding algorithms to calculate new parity data, based on portions of data included in all data volumes of the column of the matrix in which the particular data volume is maintained; and 
   
     
     
         5 . The system of  claim 1 , wherein the at least one processor is further configured to:
 determine whether a data volume of the plurality of data volumes has failed; and   when the data volume has failed, generate an error.   
     
     
         6 . The system of  claim 1 , further comprising increasing the size of the matrix, and thereby reducing a number of parity drives required to be maintained in the matrix. 
     
     
         7 . A system for generating a fault tolerant matrix comprising:
 a cluster of computing nodes distributed throughout a communications network, respective computing nodes of the cluster of computing nodes maintaining a plurality of data volumes logically arranged in a matrix;   a processing device in operable communication with at least one computing node of the cluster of computing nodes to:
 receive a request requesting access to data stored in a particular data volume of the plurality of data volumes arranged in the matrix; and 
 determine whether the matrix is degraded; and 
 when the data volume is degraded, execute erasure coding algorithms to 
   regenerate the data, based on a row of the matrix in which the data volume is maintained.   
     
     
         8 . The system of  claim 1 , wherein to execute erasure coding algorithms based on a row comprises:
 when the request is a read request, determining whether at least two of the data volumes of the plurality of volumes have failed in a single row of the matrix;   when two of the data volumes of the plurality of data volumes have failed in a single row of the matrix, execute erasure coding algorithms to regenerate the data, based on portions of data included in all data volumes of the column of the matrix in which the particular data volume is maintained.   
     
     
         9 . The system of  claim 1 , wherein to execute erasure coding algorithms based on a row comprises:
 when the request is a read request, determining whether one of the data volumes of the plurality of volumes has failed in a single row of the matrix;   when one of the data volumes of the plurality of data volumes has failed in a single row, execute erasure coding algorithms to regenerate the data, based on portions of data included in all data volumes of the row of the matrix in which the particular data volume is maintained.   
     
     
         10 . The system of  claim 1 , wherein to execute erasure coding algorithms based on a row comprises:
 when the request is a write request:
 execute erasure coding algorithms to calculate new parity data, based on portions of data included in all data volumes of the row of the matrix in which the particular data volume is maintained; 
 execute erasure coding algorithms to calculate new parity data, based on portions of data included in all data volumes of the column of the matrix in which the particular data volume is maintained; and 
   
     
     
         11 . The system of  claim 1 , wherein the at least one processor is further configured to:
 determine whether a data volume of the plurality of data volumes has failed; and   when the data volume has failed, generate an error.   
     
     
         12 . The system of  claim 1 , further comprising increasing the size of the matrix, and thereby reducing a number of parity drives required to be maintained in the matrix. 
     
     
         13 . A system for generating a fault tolerant matrix comprising:
 a cluster of computing nodes distributed throughout a communications network, respective computing nodes of the cluster of computing nodes maintaining a plurality of data volumes logically arranged in a matrix;   a processing device in operable communication with at least one computing node of the cluster of computing nodes to:
 receive a request requesting access to data stored in a particular data volume of the plurality of data volumes arranged in the matrix; and 
 determine whether the matrix is degraded; and 
 when the data volume is degraded, execute erasure coding algorithms to 
   regenerate the data, based on a row of the matrix in which the data volume is maintained.   
     
     
         14 . The system of  claim 1 , wherein to execute erasure coding algorithms based on a row comprises:
 when the request is a read request, determining whether at least two of the data volumes of the plurality of volumes have failed in a single row of the matrix;   when two of the data volumes of the plurality of data volumes have failed in a single row of the matrix, execute erasure coding algorithms to regenerate the data, based on portions of data included in all data volumes of the column of the matrix in which the particular data volume is maintained.   
     
     
         15 . The system of  claim 1 , wherein to execute erasure coding algorithms based on a row comprises:
 when the request is a read request, determining whether one of the data volumes of the plurality of volumes has failed in a single row of the matrix;   when one of the data volumes of the plurality of data volumes has failed in a single row, execute erasure coding algorithms to regenerate the data, based on portions of data included in all data volumes of the row of the matrix in which the particular data volume is maintained.   
     
     
         16 . The system of  claim 1 , wherein to execute erasure coding algorithms based on a row comprises:
 when the request is a write request:
 execute erasure coding algorithms to calculate new parity data, based on portions of data included in all data volumes of the row of the matrix in which the particular data volume is maintained; 
 execute erasure coding algorithms to calculate new parity data, based on portions of data included in all data volumes of the column of the matrix in which the particular data volume is maintained; and 
   
     
     
         17 . The system of  claim 1 , wherein the at least one processor is further configured to:
 determine whether a data volume of the plurality of data volumes has failed; and   when the data volume has failed, generate an error.   
     
     
         18 . The system of  claim 1 , further comprising increasing the size of the matrix, and thereby reducing a number of parity drives required to be maintained in the matrix. 
     
     
         19 . A system for generating a fault tolerant matrix comprising:
 a cluster of computing nodes maintaining a plurality of data volumes logically arranged in a matrix;   a processing device included within each computing node of the cluster of computing nodes to:
 receive a request requesting access to data stored in a particular data volume of the plurality of data volumes arranged in the matrix; and 
 determine whether the matrix is degraded; and 
 when the data volume is degraded, execute erasure coding algorithms to 
   regenerate the data, based on a row of the matrix in which the data volume is maintained.