Migrating Client Applications to Storage Processors

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current data storage systems face inefficiencies in resource utilization, particularly for data-intensive applications, where high data transfer costs and latency occur due to the need for frequent communication between hosts and storage systems, leading to increased overhead and reduced performance.

Innovation Solution

A method is introduced to migrate client applications from hosts to data storage systems for temporary execution, utilizing dedicated resources such as processors and memory, allowing direct access to data and reducing data transfer costs, with dynamic allocation based on workload and resource availability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If client applications execute on host processors, then application execution flexibility is maintained, but data transfer overhead and latency increase due to frequent communication between hosts and storage systems

Engineering Contradiction:
Improvedata transfer overheadVSAvoidsystem architecture complexity
Core Design Contradiction:
Loss of energyVSDevice complexity

Solution Approach 1:

The patent merges the client application execution environment with the data storage system by migrating applications from external hosts to the storage system's processor. This consolidation eliminates the need for frequent data transfers between hosts and storage systems, reducing data transfer overhead while managing complexity through integrated resource allocation mechanisms.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces a new dimension to traditional storage architecture by enabling application execution directly on the storage system's processor rather than requiring external host processors. This dimensional shift from external execution to embedded execution fundamentally changes the system architecture, reducing communication overhead while requiring sophisticated resource management to handle the increased complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Productivity

If data is frequently transferred between hosts and storage systems, then application operations can be performed, but latency increases and performance decreases

Engineering Contradiction:
Improveapplication operation speedVSAvoiddata transfer latency
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-migrating client applications to the storage system's processor before they need to access data. This advance positioning of applications eliminates the need for repeated data transfers during operation, significantly reducing latency and improving application operation speed while the resource allocation manager ensures smooth transitions.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The resource allocation manager acts as an intermediary that coordinates application migration and resource allocation. It mediates between the storage system's processing resources and the migrated applications, optimizing data access patterns and eliminating unnecessary data transfers, thereby reducing latency and improving operational speed.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Loss of energy

If storage system resources are allocated to migrated applications, then data transfer overhead is reduced, but resource availability for storage operations may be impacted

Engineering Contradiction:
Improvedata transfer overheadVSAvoidresource availability for storage operations
Core Design Contradiction:
Loss of energyVSReliability

Solution Approach 1:

The patent implements dynamic resource allocation where the resource allocation manager continuously monitors and adjusts resource distribution between migrated applications and storage operations. This dynamic approach ensures that sufficient resources are always available for storage operations while providing adequate resources to migrated applications, thereby reducing data transfer overhead without compromising storage operation reliability.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The resource allocation manager changes resource allocation parameters dynamically based on system conditions, application needs, and storage operation requirements. By adjusting these parameters in real-time, the system reduces data transfer overhead for migrated applications while maintaining adequate resource availability for storage operations, balancing both competing requirements.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS8825964B1Adaptive integration of cloud data services with a data storage system
Publication Date: 2014.09.02 EMC IP HLDG CO LLC
  • US8825964B1 patent drawing
  • US8825964B1 patent drawing
  • US8825964B1 patent drawing

AI summary

Described are techniques for performing processing in a data storage system. A client application executing on a host is identified as a candidate for migration to the data storage system. First data used by the client application is stored on one or more physical storage devices of the data storage system. The client application is migrated to the data storage system for temporary execution on the data storage system. The client application is executed on the data storage system for a time period using a first portion of resources of the data storage system allocated for exclusive use by migrated client applications executing on the data storage system.