Track Format Code in Cache Control Block for Storage Copy Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In storage environments with copy relationships between source and target storages, there is a need for improved techniques to process host read/write requests efficiently, as existing methods incur significant latency due to the need to rebuild track format metadata and context switching, especially when data is not cached.
Innovation Solution
Incorporating a track format code in the cache control block for target tracks, which allows for fast access to track format metadata from a track format table, reducing the need to read metadata from storage and minimizing latency by using the track format code to process subsequent requests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If track format metadata is rebuilt from cache data for target tracks in copy relationships, then data consistency is maintained, but processing latency increases significantly
Solution Approach 1:
The patent pre-stores track format metadata in association with cache control blocks before it is needed. When data is staged from source storage to target cache, the corresponding track format metadata is retrieved and stored in the cache control block in advance, so that when a read request arrives, the metadata is already available and does not need to be rebuilt, thus eliminating the latency penalty while maintaining data consistency
Solution Approach 2:
The patent introduces an intermediary structure (cache control block) that holds both the staged data and its associated track format metadata together. This intermediary allows the system to access metadata directly from the cache control block without needing to rebuild it from data, acting as a mediator between the staged data and the metadata retrieval process, thereby reducing processing time while ensuring consistency
2Productivity
If track format metadata is stored in cache control blocks with track format codes, then subsequent request processing speed improves, but cache structure complexity increases
Solution Approach 1:
The patent merges track format metadata with the existing cache control block structure by storing the metadata in the metadata portion of the cache control block and maintaining an association through track format codes. This combination allows the system to access both the staged data and its metadata from a single cache control block without requiring separate storage structures, thereby improving processing speed while limiting the increase in overall cache structure complexity
Solution Approach 2:
The patent copies track format metadata from source storage into the cache control block when data is staged, creating a duplicate copy that resides locally in the cache. This copied metadata can then be accessed immediately without returning to source storage, improving processing speed for subsequent requests while the copy is maintained within the existing cache control block framework
Data Source
AI summary
Provided area computer program product, system, and method for maintaining track format metadata for target tracks in a target storage in a copy relationship with source tracks in a source storage. Upon receiving a request to a requested target track in the target storage, the source track for the requested target track is staged from the source storage to a cache to be used as the requested target track in response to determining that the copy relationship information indicates that a source track needs to be copied to the requested target track. A determination is made of track format metadata for the requested target track, comprising the staged source track, indicating a format and layout of data in the requested target track and a track format code identifying the track format metadata. The track format code is included in a cache control block for the requested target track.