VMFS Journal Space Reservation via Metadata Batching

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing Virtual Machine File System (VMFS) journaling file system faces performance degradation due to severe write I/O operations and limited journal space, leading to inefficient use of journal space and suboptimal transaction parallelism and throughput.

Innovation Solution

The solution categorizes metadata updates into common and exclusive updates, allowing for optimized journal space reservation by batching logical transactions into physical transactions, where common updates require a single journal space reservation and exclusive updates require separate reservations, thereby improving the efficiency of journal space usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If journal space is reserved for each metadata update in VMFS, then data integrity and consistency are maintained, but write I/O performance degrades due to severe performance penalties from multiple subsystem involvement

Engineering Contradiction:
Improvedata integrityVSAvoidwrite I/O performance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent combines multiple individual metadata update operations into a single batch transaction. Instead of processing each metadata update separately with individual journal reservations, the system aggregates multiple updates and processes them together in one transaction, reducing the number of subsystem involvement events and improving write I/O performance while maintaining data integrity.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent performs preliminary actions by pre-reserving journal space for batched transactions before actual write operations occur. This allows the system to prepare and organize multiple metadata updates in advance, then execute them as a unified operation, reducing the performance penalty associated with repeated subsystem involvement during actual write operations.

Inventive Principle:
Principle #10Preliminary action

2Quantity of substance

If journal space is limited in VMFS, then storage efficiency improves, but transaction parallelism and throughput are reduced

Engineering Contradiction:
Improvejournal spaceVSAvoidtransaction throughput
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

The patent merges multiple logical transactions into fewer physical transactions, allowing more transactions to be processed within the limited journal space. By batching metadata updates together, the system increases transaction parallelism and throughput without requiring additional journal space, as multiple operations share the same journal reservation.

Inventive Principle:
Principle #5Merging (Combining)

3Reliability

If multiple subsystems are involved in VMFS metadata updates, then data consistency is ensured, but overall system performance degrades

Engineering Contradiction:
Improvedata consistencyVSAvoidsystem performance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent reduces subsystem involvement by combining multiple metadata update operations into a single batch transaction. This approach maintains data consistency through unified transaction processing while minimizing the frequency of subsystem engagement, thereby improving overall system performance.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS12050926B2Journal space reservations for virtual disks in a virtualized computing system
Publication Date: 2024.07.30 VMWARE INC
  • US12050926B2 patent drawing
  • US12050926B2 patent drawing
  • US12050926B2 patent drawing

AI summary

An example method of handling, at a hypervisor on a host in a virtualized computing system, a write input/output (IO) operation to a file on a storage device having a virtual machine file system (VMFS) is described. The method includes: generating logical transactions for the write IO operation having updates to metadata of the VMFS for the file; estimating, for the logical transactions, common space reservations for those of the updates to common fields in the metadata for the file; estimating, for the logical transactions, exclusive space reservations for those of the updates to exclusive fields in the metadata for the file; batching the logical transactions into a physical transaction, which includes a single reservation of space in a journal of the VMFS based on the common space reservations and a reservations of space in the journal for each of the exclusive space reservations, respectively.