Heat-Tiered Storage Controller Workload Interface
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional heat-based storage tiering policies in cloud computing environments do not account for computational workloads and access interfaces, leading to skewed data distribution and increased access latencies for end users.
Innovation Solution
Implementing a workload and interface-cognizant heat-tiered storage policy, where a storage controller updates the heat attributes of file system objects based on the accessing workload and interface, allowing for intelligent distribution of data among multiple tiers to optimize access latencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If conventional heat-based storage tiering policies are used, then data is distributed among storage tiers based on access frequency, but access latency increases for end users due to skewed data distribution caused by management process accesses
Solution Approach 1:
The patent segments the heat calculation by separating management process accesses from end user accesses. The storage controller identifies the interface type (management vs. end user) and applies different weighting factors accordingly, allowing independent optimization of data distribution for end user workloads while still accommodating management processes.
Solution Approach 2:
The patent applies different quality factors (weights) to different types of accesses based on their purpose. End user accesses receive a weight of 1.0 while management process accesses receive a weight of 0.0, creating localized quality differences in heat calculation that reflect the different priorities and requirements of each access type.
2Quantity of substance
If heat-based tiering is implemented without workload awareness, then storage capacity is optimized by placing cold data in lower tiers, but end user access performance deteriorates due to artificial heat inflation by management processes
Solution Approach 1:
The patent implements dynamic heat calculation that adapts to the accessing workload type. The storage controller dynamically adjusts the weight factor based on whether the access is from an end user or management process, allowing the heat metric to reflect true end user access patterns rather than being distorted by management operations.
Solution Approach 2:
The patent introduces feedback mechanisms where the storage controller monitors access patterns, identifies management vs. end user accesses, and adjusts heat calculation accordingly. This feedback loop ensures that data distribution decisions are based on actual end user behavior rather than being skewed by management process activities.
Data Source
AI summary
A data storage system has multiple tiers of data storage including an upper tier having a lower access latency and a lower tier having a higher access latency. A storage controller of the data storage system receives, via an interface, an access request of a workload for a target file system object, where the target file system object has an associated temperature. In response to the access request for the target file system object, the storage controller accesses the target file system object in the data storage and conditions update of the associated temperature of the target file system object based on at least one of a set including the interface and the workload. The storage controller distributes a collection of file system objects including the target file system object among the multiple tiers based on respective heats of file system objects in the collection.