Virtual Storage System Boot Phase Storage Type Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In cloud environments, virtual storage systems lack the ability to distinguish between different types of storage devices during the boot phase, preventing the creation of hybrid aggregates necessary for effective data tiering, as information about storage device properties is not available via hypervisor APIs.

Innovation Solution

A mechanism is implemented to detect and persist the types of backing storage for cloud volumes by using an API to obtain information during the boot phase and storing it in a cache, allowing the virtual storage system to determine and initialize the type of storage devices, enabling data tiering by creating hybrid aggregates with different performant tiers.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If the virtual storage system uses hypervisor APIs to obtain storage device information during boot phase, then the system can access storage devices, but the storage device type information is not available preventing data tiering

Engineering Contradiction:
Improvestorage device type informationVSAvoidboot phase operation
Core Design Contradiction:
Loss of informationVSEase of operation

Solution Approach 1:

The patent applies preliminary action by obtaining storage device type information during the boot phase before the virtual storage system fully operationalizes. The system proactively queries the hypervisor API early in the boot process to retrieve backing storage type information (e.g., SSD, HDD, NVMe) and caches this information for subsequent use, ensuring data tiering capabilities are established before normal operation begins.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent uses a cache memory as an intermediary to store and preserve storage device type information that would otherwise be lost between boot phases. This intermediary component bridges the gap between the hypervisor API information availability and the virtual storage system's need for this information during operational phases, enabling data tiering functionality.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If the virtual storage system creates hybrid aggregates for data tiering, then performance is optimized by moving cold data to cheaper storage, but the system complexity increases

Engineering Contradiction:
Improvedata tiering performanceVSAvoidhybrid aggregate structure
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent applies segmentation by dividing the storage system into distinct performance tiers based on storage device types. It creates hybrid aggregates that segment storage resources into performance tiers (e.g., hot data tier using SSDs, cold data tier using HDDs), allowing independent management and optimization of each tier while maintaining overall system productivity through automated data placement.

Inventive Principle:
Principle #1Segmentation

3Measurement precision

If the system obtains storage information via API during boot phase, then accurate storage device type detection is achieved, but additional processing time is required

Engineering Contradiction:
Improvestorage device type detection accuracyVSAvoidboot phase duration
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs the storage device type detection as a preliminary action during the boot phase, querying the hypervisor API early when the virtualization layer is initializing. By obtaining and caching this information upfront, the system achieves precise storage device type detection without requiring additional processing time during operational phases, as the information is readily available from the cache.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11842084B2Detecting and persisting across boots types of backing storage of cloud volumes available to a virtual storage system
Publication Date: 2023.12.12 NETAPP INC
  • US11842084B2 patent drawing
  • US11842084B2 patent drawing
  • US11842084B2 patent drawing

AI summary

Systems and methods for facilitating performance of data tiering by detecting and persisting types of backing storage of cloud volumes that are available for use by a virtual storage system are provided. In one example, during a boot phase of the virtual storage system within a compute instance of a cloud environment, information regarding a type of backing storage for each of multiple storage devices associated with the compute instance that is not available via hypervisor APIs is obtained via an API exposed by the cloud environment. The multiple storage devices may then be initialized to facilitate subsequent use of a subset thereof in connection with performing data tiering by persisting information indicative of the type of backing storage for the storage device within a reserved header region on each storage device.