Virtual Machine Migration Cache Copying

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional storage area networks face challenges in improving communication and data access efficiency due to the latency associated with hard disk drives and the random nature of read and write operations, particularly in handling high-priority storage requests in a timely manner.

Innovation Solution

The implementation of priority queuing and message passing mechanisms within storage networks, utilizing flash memory technology to segregate and prioritize storage requests and acknowledgments, and enabling in-band communication over Fibre Channel or Ethernet protocols to optimize storage operations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If traditional hard disk drives are used for storage operations, then storage capacity is achieved, but access latency increases due to rotational media dependencies

Engineering Contradiction:
Improveaccess latencyVSAvoiddata access speed
Core Design Contradiction:
Loss of timeVSSpeed

Solution Approach 1:

The patent segments storage operations into priority-based queues, separating high-priority requests from standard requests. This allows critical storage operations to be processed independently without being blocked by lower-priority operations on rotational media, thereby reducing access latency for time-sensitive data.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a cache layer as an intermediary between the processor and hard disk drives. This cache stores frequently accessed data, allowing the processor to retrieve data quickly without waiting for rotational media to rotate to the correct position, thus reducing access latency while maintaining storage capacity.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Speed

If sequential data placement is used on hard drives, then rotational latency is minimized, but random data access performance deteriorates

Engineering Contradiction:
Improvesequential read speedVSAvoidrandom access latency
Core Design Contradiction:
SpeedVSLoss of time

Solution Approach 1:

The patent implements dynamic data placement strategies that adapt to access patterns. Frequently accessed random data is placed in cache or on sequential sectors, while less frequently accessed data can be placed anywhere on the rotational media. This dynamic approach optimizes for both sequential and random access performance depending on actual usage patterns.

Inventive Principle:
Principle #15Dynamics

3Adaptability or versatility

If standard Ethernet communication is used in storage networks, then network compatibility is maintained, but storage network performance decreases

Engineering Contradiction:
Improvenetwork compatibilityVSAvoidstorage network throughput
Core Design Contradiction:
Adaptability or versatilityVSProductivity

Solution Approach 1:

The patent makes the storage network interface universal by supporting multiple communication protocols including both traditional Ethernet and Fibre Channel. This allows the same hardware infrastructure to serve both compatibility requirements and high-performance requirements, maintaining broad adaptability while enabling optimized storage operations when Fibre Channel is available.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS10042768B1Virtual machine migration
Publication Date: 2018.08.07 AVAGO TECHNOLOGIES INTERNATIONAL SALES PTE LTD
  • US10042768B1 patent drawing
  • US10042768B1 patent drawing
  • US10042768B1 patent drawing

AI summary

A method and system are provided for a storage area network including a plurality of virtual servers. In one example, the system sends a storage area network message to a second virtual server to prepare to perform one or more functions associated with the storage area network. The system copies a first cache table associated with a first cache memory and a first virtual server to a second cache table associated with a second cache memory and the second virtual server. The system uses a processor to execute one or more storage requests in the second virtual server, by using the first cache memory to support the performance of the one or more functions by the second virtual server, and without completely filling cache data into the second cache memory.