Virtual Device Sparing in Memory Ranks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current reliability, availability, and serviceability (RAS) features in memory systems, such as rank sparing and double device data correction, are not resource-efficient due to high power consumption and reduced bandwidth, and require wasteful resource reservation for failed memory devices.

Innovation Solution

Virtual device sparing techniques that reserve only a fraction of a rank's capacity to cover for a failed device, utilizing error correction to copy data into a virtual spare device within the same rank, reducing power and bandwidth usage until a failure occurs, and allowing for efficient resource utilization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If rank sparing is used to reserve a whole rank of memory for failed devices, then system reliability is improved, but resource efficiency deteriorates due to wasteful reservation of entire rank capacity

Engineering Contradiction:
Improvesystem reliabilityVSAvoidmemory capacity utilization
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent segments the memory address space into device-level units rather than reserving entire ranks. Each memory device has its own dedicated address space that can be independently reserved and activated only when needed, replacing the coarse-grained rank-level reservation with fine-grained device-level reservation.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements dynamic reservation where the spare device is activated only when a failure is detected. The system transitions from a static reserved state to an active correction state by mapping failed device addresses to the spare device's address space, enabling on-demand resource allocation based on actual failure conditions.

Inventive Principle:
Principle #15Dynamics

2Reliability

If double device data correction with lockstepping is used, then data reliability is improved, but power consumption increases and bandwidth is reduced

Engineering Contradiction:
Improvedata reliabilityVSAvoidpower consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent uses copying to transfer data from multiple functional devices to a spare device's address space. When a failure is detected, the system copies data from the affected devices to the reserved spare device, which then serves as the active replacement. This eliminates the need for continuous lockstepping operations while maintaining data integrity.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent performs preliminary action by pre-reserving address spaces for spare devices and pre-positioning correction capabilities. The system prepares the spare device's address space in advance but only activates it when needed, avoiding continuous operational overhead while maintaining readiness for failure correction.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS9201748B2Virtual device sparing
Publication Date: 2015.12.01 INTEL CORP
  • US9201748B2 patent drawing
  • US9201748B2 patent drawing
  • US9201748B2 patent drawing

AI summary

Systems and techniques for virtual device sharing. A failure of one of a plurality of memory devices corresponding to a first rank in a memory system is detected. The memory system has a plurality of ranks, each rank having a plurality of memory devices used to store a cache line. A portion of the cache line corresponding to the failed memory device is stored in a memory device in a second rank in the memory system and the remaining portion of the cache line in the first rank of the memory system.