NFS Client Same-Node Direct Access Bypass

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

When a Network File System (NFS) client and server are running on the same node, they can cause file cache resource conflicts and inefficiencies due to the need for network communication, leading to deadlocks and computational overhead from context switches.

Innovation Solution

The NFS client determines if it and the server are on the same node by comparing IP addresses, allowing it to bypass the RPC and networking layers, directly accessing the file system and avoiding cache allocation and context switches.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If NFS client and server communicate over the network using RPC protocol, then file access can be achieved across different nodes, but file cache resource conflicts and deadlocks occur when client and server are on the same node

Engineering Contradiction:

Improvefile access reliabilityVSAvoidcache management complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system dynamically changes the access parameter by detecting whether the NFS client and server are on the same node. When they are on the same node, the system switches from network RPC access to direct local file system access, changing the access path parameter to eliminate cache conflicts and deadlocks while maintaining file access reliability

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system implements dynamic access path selection by continuously monitoring the relative position of client and server. The access method is dynamically adjusted based on whether they are on the same node or different nodes, optimizing performance and avoiding conflicts adaptively

Inventive Principle:

Principle #15Dynamics

2Adaptability or versatility

If NFS protocol is used for file access, then network file sharing is enabled, but computational overhead from context switches and network communication increases

Engineering Contradiction:

Improvenetwork file sharing capabilityVSAvoidcomputational overhead

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The system changes the access parameter based on node location. When client and server are on the same node, it switches from network RPC protocol to direct local file system access, eliminating network communication overhead and context switches while preserving network file sharing capability when needed

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system extracts and removes the unnecessary network RPC layer and context switching overhead when client and server are on the same node, directly accessing the file system without going through the NFS protocol stack, thereby reducing computational overhead

Inventive Principle:

Principle #2Taking out (Extraction)

3Ease of operation

If NFS client mounts exported share, then file access is simplified for applications, but file cache space allocation conflicts arise when client and server are on the same node

Engineering Contradiction:

Improvefile access simplicityVSAvoidcache allocation reliability

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The system changes the access parameter by detecting same-node conditions. When client and server are on the same node, it switches from mounted share access with RPC to direct file system access, eliminating cache allocation conflicts while maintaining simplified file access through the virtual file system interface

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10257257B2Direct access to network file system exported share

Publication Date: 2019.04.09 HEWLETT PACKARD ENTERPRISE DEV LP

AI summary

Techniques for direct access of a Network File System (NFS) exported share are provided. In one aspect, a request to access a NFS exported share may be received at a NFS client. It may be determined that a NFS server exporting the share is running on the same node that is running the NFS client. The exported share may be directly accessed by the NFS client.