Semi-Persistent Tunnel Protocol for SAN Data Replication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data replication protocols in storage area networks (SANs) face inefficiencies due to high overhead, latency, and difficulty in maintaining in-order delivery and error detection, especially in multi-stream data transfers and environments with variable latency, leading to performance bottlenecks and resource underutilization.
Innovation Solution
A semi-persistent tunnel protocol is established between storage controllers, using pre-allocated resource bundles and Message Function Calls (MFCs) for efficient data transfer, allowing for in-order delivery and quick adaptation to link failures or congestion, with self-renewing resources and reduced frame traffic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If SCSI protocol is used for data transfer in SAN, then data transfer between storage controllers can be established, but high overhead and latency occur due to initiator-controlled data flow and frequent exchange setup/teardown
Solution Approach 1:
The patent pre-allocates communication resources (buffers, exchange identifiers, and tunnel parameters) before data transfer begins. This preliminary setup eliminates the need for frequent resource allocation during data transfer, reducing latency while maintaining data integrity through pre-configured error detection and correction mechanisms.
Solution Approach 2:
The patent establishes semi-persistent tunnels that remain active during data transfer operations, allowing continuous data flow without repeated exchange setup and teardown. This continuity eliminates the overhead of frequent protocol negotiation while maintaining reliable data transfer through sustained connection state.
2Ease of operation
If initiator controls all data flow in SCSI protocol, then command and response management is centralized, but target controller resources are underutilized creating performance bottlenecks
Solution Approach 1:
The patent enables the target controller to autonomously manage its own resources, including self-allocating buffers and self-managing exchange identifiers within the semi-persistent tunnel framework. This self-service capability allows the target to actively participate in data transfer management, utilizing its resources efficiently while maintaining coordinated command control.
Solution Approach 2:
The patent transforms the static initiator-controlled data flow into a dynamic peer-to-peer interaction model where both controllers can initiate and respond to data transfer operations. This dynamic approach allows flexible resource utilization while maintaining operational simplicity through standardized protocol interfaces.
3Adaptability or versatility
If frequent exchange setup and breakdown is performed in SCSI over Fibre Channel, then data transfer operations can be initiated, but significant overhead is incurred increasing resource requirements and latency
Solution Approach 1:
The patent pre-configures tunnel parameters, buffer allocations, and exchange identifier mappings before data transfer operations. This preliminary configuration establishes a stable framework that supports multiple data transfer operations without requiring repeated setup, reducing complexity while maintaining operational flexibility through parameter reusability.
Solution Approach 2:
The patent creates semi-persistent tunnels that can handle multiple data transfer operations, different data streams, and various command types within a single established connection framework. This universal approach allows one tunnel structure to serve multiple functions, reducing the need for separate configurations and simplifying resource management.
4Reliability
If traditional SCSI protocol is used for multi-stream data transfer, then data replication can be performed, but performance degrades due to protocol overhead and inability to efficiently handle variable latency
Solution Approach 1:
The patent pre-establishes tunnel parameters and resource allocations specifically optimized for data replication operations, including pre-configured buffer sizes and exchange identifier mappings that accommodate replication workloads. This preliminary setup enables high-speed replication while maintaining data accuracy through pre-validating replication parameters.
Solution Approach 2:
The patent maintains semi-persistent tunnels throughout replication operations, allowing continuous data flow between source and target storage controllers. This continuity eliminates protocol overhead that would otherwise slow replication, while maintained data integrity through sustained error detection and correction mechanisms active throughout the replication process.
Data Source
AI summary
A system for communicating between two devices in a network in which a semi-persistent tunnel is established between the two devices in advance of data communication. The semi-persistent tunnel includes resources that are pre-allocated in a first device at a first end of the communication link by a second device at the second end of the communication link. The first and second devices implement a plurality of processes for handling data transfer operations. Preferably, the semi-persistent tunnel also includes resources that are pre-allocated in a device at the second end of the communication link by the device at the first end of the communication link to allow bi-directional communication. Data transfer operations transmitted through the tunnel include an identification of specific resources of the pre-allocated resources that are to handle the data transfer operation. Data transfer operations also include a vector identifying a particular process among the plurality of processes that implement the data transfer operation.