Secondary Journal Volume Mapping for Low-Cache Remote Copy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing remote copy techniques face performance issues and resource inefficiencies when implemented in configurations with unbalanced resources between primary and secondary storage systems, particularly in public cloud environments with limited cache memory, leading to increased RPO and host IO performance degradation.
Innovation Solution
A storage system design that associates journal data with a capacity pool and logical addresses, eliminating the need for duplicate caching and reducing the number of components involved in the copy process, thereby improving copy performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If journal data is written to both JVOL and PVOL using cache memory, then data consistency is maintained, but cache memory capacity is consumed twice with the same data
Solution Approach 1:
The patent merges the storage locations by allowing journal data to be written directly to the PVOL without duplicating it in the JVOL. The PVOL serves dual purposes as both the primary volume and the journal storage location, eliminating redundant cache memory consumption while maintaining data consistency through the unified storage approach.
Solution Approach 2:
The PVOL is given multi-functionality by serving as both the primary data storage volume and the journal volume. This universal usage of the PVOL for both data storage and journaling purposes eliminates the need for separate JVOL cache memory allocation, thereby reducing overall cache memory consumption while maintaining reliability.
2Reliability
If journal data is restored from JVOL to SVOL with copy processing, then data is replicated to secondary storage, but cache memory capacity is consumed twice by JVOL and SVOL
Solution Approach 1:
The patent merges the journal storage function into the PVOL, eliminating the separate JVOL. When data is written to the PVOL, it is simultaneously available for replication to the SVOL without requiring duplicate caching. This merging approach reduces cache memory consumption while maintaining reliable data replication to the secondary storage system.
Solution Approach 2:
The patent extracts the journaling function from the separate JVOL and integrates it directly into the PVOL operation. By taking out the redundant journal storage step and performing journaling inline with the primary volume writes, the system reduces cache memory consumption during both writing and restoration operations while maintaining data replication reliability.
3Quantity of substance
If SDS with small cache memory is used for secondary storage, then cost is optimized, but journal data is easily destaged increasing temporary storage requirements
Solution Approach 1:
By merging journal storage into the PVOL, the patent eliminates the need for large cache memory on the SDS side. The journal data is already present in the PVOL cache on the primary side, so replication to the SVOL can proceed with minimal additional cache requirements on the cost-optimized SDS secondary storage system, maintaining copy performance without requiring expensive large-cache hardware.
4Reliability
If journal data stays in JVOL exceeding capacity, then remote copy state is stopped, but host IO performance deteriorates
Solution Approach 1:
The patent merges journal storage into the PVOL, eliminating the separate JVOL that can exceed capacity. Since journal data is written directly to the PVOL alongside primary data, there is no separate journal volume to fill up and cause remote copy state interruptions. This eliminates the trade-off between maintaining remote copy reliability and preserving host IO performance.
Data Source
AI summary
In a secondary storage system, a processor of the secondary storage system receives a journal including journal data that is write target data for a primary volume from a primary storage system and metadata, associates a logical address of a secondary journal volume that manages the journal data included in the journal with a predetermined area of a pool in the secondary storage system, and associates an address of the predetermined area associated with the journal data of the pool with a logical address of a secondary volume in which the write target data indicated by the metadata included in the journal is stored.