Reducing Memory Overhead Associated With Memory Protected By A Fault Protection Scheme

摘要

A memory request initiates access to a memory location in the memory. The memory location is evaluated to determine whether the memory location is located within a first portion of the memory or within a second portion of the memory. In response to determining that the memory location is located within the first portion and that the memory request is a read request, the memory location located within the first portion is accessed. In response to determining that the memory request is a write request and the memory location is located within the first portion, the memory location located within the first portion and a duplicate of the memory location is accessed. When the memory location is located within the second portion, the memory location located within the second portion using a redundant array of independent disks (RAID) memory mechanism, in response to the memory request being the write request.

主权项

1. A method to reduce write overhead to a memory in a fault protection scheme, the method comprising:
receiving a memory request that initiates access to a memory location in the memory, wherein the memory includes a first portion and a second portion, the first portion duplicated to provide a duplicated first portion; evaluating the memory location to determine whether the memory location is located within the first portion or within the second portion; in response to determining that the memory location is located within the first portion:
accessing the memory location located within the first portion, in response to the memory request being a read request, andaccessing the memory location located within the first portion and a duplicate of the memory location that is located within the duplicated first portion, in response to the memory request being a write request; and in response to determining that the memory location is located within the second portion:
accessing the memory location located within the second portion, in response to the memory request being the read request, andaccessing the memory location located within the second portion using a redundant array of independent disks (RAID) memory mechanism, in response to the memory request being the write request.