DISTRIBUTED STORAGE NETWORK INCLUDING MEMORY DIVERSITY
First Claim
1. A method comprising:
- generating a plurality of data slices in a distributed storage processing unit;
determining metadata associated with the plurality of data slices, the metadata including information that can be used to determine storage diversity preferences;
determining characteristics of a plurality of storage units included in a distributed storage network;
selecting particular storage units of the plurality of storage units based on a favorable match between the storage diversity preferences and the characteristics of the particular storage units; and
delivering the plurality of data slices to the particular storage units.
4 Assignments
0 Petitions
Accused Products
Abstract
A distributed storage processing unit can generate data slices and determine metadata for each of the data slices. The metadata includes information that can be used to determine storage diversity preferences, which can include requirements that data slices generated from a common data segment each be stored in memories of the same (or different) type and model, memories with the same (or different) failure rates, memories having fast read (or write) characteristics, and so on. Decisions about which memory units to use for storing data slices can be made based on how closely the characteristics of the memories match the storage diversity preferences. The decision can be made at a distributed storage processing unit that generates the data slices, at a distributed storage unit receiving the data slices for storage, or elsewhere.
105 Citations
21 Claims
-
1. A method comprising:
-
generating a plurality of data slices in a distributed storage processing unit; determining metadata associated with the plurality of data slices, the metadata including information that can be used to determine storage diversity preferences; determining characteristics of a plurality of storage units included in a distributed storage network; selecting particular storage units of the plurality of storage units based on a favorable match between the storage diversity preferences and the characteristics of the particular storage units; and delivering the plurality of data slices to the particular storage units. - View Dependent Claims (2, 3, 4, 5, 6)
-
-
7. A method comprising:
-
receiving a data slice to be stored in one of a plurality of memories at a dispersed storage unit; obtaining metadata associated with the plurality data slice, the metadata including an indication of storage diversity preferences; determining characteristics of the plurality of memories; selecting particular memories of the plurality of memories based on a favorable match between the storage diversity preferences and the characteristics of the particular memories; and storing the data slices to the particular memory. - View Dependent Claims (8, 9, 10, 11)
-
-
12. A dispersed storage processing unit comprising:
-
a processor to generate a plurality of data slices, each of the plurality of data slices including metadata; the processor further to select candidate storage units based on a correlation between the metadata and characteristics of the candidate storage units; and a communications interface to deliver the plurality of data slices to the candidate storage units. - View Dependent Claims (13, 14, 15, 16)
-
-
17. A dispersed storage unit comprising:
-
a communications interface to receive a data slice for storage, the data slice including metadata; a plurality of memory units; a processor to select at least one of the memory units to store the data slice, the selection based on a correlation between the metadata and characteristics of the plurality of memory units; the processor further to direct the data slices to the at least of the memory units to be stored. - View Dependent Claims (18, 19, 20, 21)
-
Specification