Storage System Virtual Volume Endpoint Binding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional storage systems struggle to provide service level agreements (SLA) guarantees and data services on a per-virtual machine (VM) basis, leading to inefficiencies in resource management and service delivery due to the lack of scalability and granularity in virtualized environments with multiple VMs sharing storage resources.
Innovation Solution
The implementation of a storage system that exports virtual volumes on a per-workload basis, allowing for SLA guarantees and data services to be provided per virtual disk, using protocol endpoints for standard protocols like SCSI and NFS, enabling separate management and access for each VM's virtual disks and snapshots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple VMs share storage resources in a conventional storage system, then resource utilization is improved, but the ability to provide SLA guarantees and data services on a per-VM basis deteriorates
Solution Approach 1:
The storage system segments storage resources into virtual volumes that can be individually allocated to specific VMs. Each virtual volume acts as an independent storage unit with its own SLA guarantees and data services, enabling per-VM customization while maintaining shared infrastructure. This segmentation resolves the contradiction by allowing both high resource utilization through sharing and granular service control through individual volume management.
2Ease of operation
If storage systems provide centralized oversight and management, then operational efficiency is improved, but the granularity of per-workload service delivery deteriorates
Solution Approach 1:
The storage system implements a universal management interface that can handle both centralized oversight functions and granular per-workload service configurations through a unified architecture. The virtual volume abstraction layer provides a consistent management model that works at both the aggregate and individual levels, allowing administrators to manage storage centrally while automatically enforcing per-VM SLAs and services without increasing operational complexity.
3Quantity of substance
If conventional storage systems use LUN-based virtualization, then storage aggregation is improved, but the ability to isolate workloads and provide per-VM SLA guarantees deteriorates
Solution Approach 1:
The system transitions from traditional LUN-based virtualization to a hierarchical virtual volume model that adds a new dimension of abstraction. Virtual volumes are organized in layers that enable both aggregate storage management and fine-grained workload isolation. This dimensional change allows the system to maintain large aggregated storage pools while simultaneously providing per-VM SLA guarantees through the hierarchical structure, resolving the contradiction between capacity aggregation and reliability enforcement.
Data Source
Figure 1A
Figure 1B
Figure 2A
AI summary
The storage system exports logical storage volumes that are provisioned as storage objects. These storage objects are accessed on demand by connected computer systems using standard protocols, such as SCSI and NFS, through logical endpoints for the protocol traffic that are configured in the storage system. Prior to issuing input-output commands to a logical storage volume, the computer system sends a request to bind the logical storage volume to a protocol endpoint. In response a first identifier for the protocol endpoint and a second identifier for the logical storage volume is returned. Different second identifiers may be generated for different logical storage volumes even though the same protocol endpoint is being used. Therefore, a single protocol endpoint may serve as a gateway for multiple logical storage volumes.