Storage Workload Offloading via Virtual Machine Migration

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing storage systems face challenges in efficiently expanding capacity and migrating storage workloads without disrupting connections between hosts and storage subsystems, particularly in handling data reduction tasks like compression and deduplication, which consume significant computing resources and are difficult to scale or migrate.

Innovation Solution

The solution involves creating a virtual machine on a low-usage server to offload storage workloads by configuring virtual logical units and Fibre Channel ports, copying processing control tables, and switching storage I/O paths, allowing for the distribution of storage functions across multiple hosts to optimize resource usage and reduce workload.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If specialized hardware is provided for data reduction processing, then data reduction workload is improved, but device complexity increases and capacity expansion becomes difficult

Engineering Contradiction:

Improvedata reduction workloadVSAvoiddevice complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent replaces specialized hardware (mechanical system) with software-based data reduction processing on general-purpose servers. The storage controller virtual machine executes compression and deduplication algorithms in software, eliminating the need for dedicated hardware appliances while maintaining data reduction functionality.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent enables general-purpose servers to perform multiple functions including storage control, data reduction, and virtualization. The storage controller VM can be migrated between different servers, allowing the same hardware platform to serve multiple roles and enabling flexible resource allocation across the storage infrastructure.

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Power

If specialized hardware is used for storage processing, then processing capability is improved, but adaptability decreases and migration becomes difficult

Engineering Contradiction:

Improveprocessing capabilityVSAvoidmigration capability

Core Design Contradiction:

PowerVSAdaptability or versatility

Solution Approach 1:

The patent replaces specialized hardware with software-based storage control functions that can be virtualized and migrated. The storage controller VM encapsulates all processing capabilities in software form, allowing it to be moved between different physical servers without hardware dependency.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent extracts the storage control functionality from dedicated hardware and places it into a virtual machine that can be independently managed and migrated. The storage controller VM is separated from the physical hardware platform, allowing flexible deployment and migration across different server environments.

Inventive Principle:

Principle #2Taking out (Extraction)

3Productivity

If storage workload is concentrated on dedicated storage subsystem, then processing efficiency is improved, but resource utilization decreases

Engineering Contradiction:

Improveprocessing efficiencyVSAvoidresource utilization

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent merges storage control functions with general-purpose server resources. By consolidating storage controllers and data reduction processing onto shared server infrastructure, the system achieves better resource utilization while maintaining processing efficiency through virtualization and workload distribution.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent enables servers to serve dual purposes as both compute resources and storage control platforms. The same hardware resources can dynamically allocate capacity between virtual machine workloads and storage control functions, optimizing overall resource utilization across the infrastructure.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS9164880B2Method and apparatus for offloading storage workload

Publication Date: 2015.10.20 HITACHI VANTARA LTD

AI summary

Exemplary embodiments provide a technique to offload storage workload. In one aspect, a computer comprises: a memory; and a controller operable to manage a relationship among port information of an initiator port, information of a logical volume storing data from the initiator port, and port information of a target port to be used for storing data from the initiator port to the logical volume, and to cause another computer to process a storage function of a storage system including the logical volume and the target port by creating a virtual machine for executing the storage function and by configuring the relationship on said another computer, said another computer sending the data to the logical volume of the storage system after executing the storage function. In specific embodiments, by executing the storage function on said another computer, the workload of executing the storage function on the storage system is eliminated.