Persistent Memory Storage Stack RDMA Replication Latency

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Hard drives suffer from high latency and are prone to failure, leading to data corruption or loss, making them unsuitable for certain storage situations.

Innovation Solution

Implementing a system that uses persistent-memory devices for data storage, both locally and remotely, with synchronous data replication via RDMA to reduce latency and ensure data availability, utilizing a storage stack that performs write operations independently of the central processing unit to maintain transparency and high availability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If hard drives are used for data storage, then data can be retained for long periods, but latency is high and data loss can occur

Engineering Contradiction:

Improvedata retention reliabilityVSAvoidstorage latency

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system segments storage functionality across multiple persistent memory devices distributed across different computing devices. Local persistent memory provides fast access while remote persistent memory devices provide redundancy and extended capacity, dividing the storage system into independent segments that can operate autonomously

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a storage stack and memory controller as intermediary components between the CPU and persistent memory devices. The storage stack handles data operations independently, using RDMA technology to enable direct memory access between devices without CPU involvement, thus reducing latency while maintaining reliability

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If data is replicated remotely across multiple devices, then data availability and redundancy are improved, but system complexity increases

Engineering Contradiction:

Improvedata availabilityVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The storage stack implements self-service mechanisms by autonomously managing data replication, synchronization, and recovery operations. The system automatically detects failures and initiates data reconstruction from redundant copies without requiring manual intervention or complex centralized control

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent merges storage management functions with the existing memory controller and storage stack infrastructure. By integrating persistent memory management into the current storage hierarchy and using standard RDMA protocols, the system achieves remote replication without adding significant architectural complexity

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS10313471B2Persistent-memory management

Publication Date: 2019.06.04 RED HAT INC

AI summary

Data can be stored in a persistent-memory device, rather than a hard drive, of a computing device. A copy of the data can also be stored in another persistent-memory device of a remote computing device. For example, a central processing unit (of the computing device) can perform a first write operation to cause a file to be stored in the persistent-memory device. A memory controller can perform a second write operation to cause another memory controller of the remote computing device to store a copy of the file in the other persistent-memory device of the remote computing device.