Storage Device Access Mediation via Dynamic Budget and De-randomization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in managing storage device access among multiple entities, leading to bandwidth contention, thrashing, and inefficient I/O operations, particularly in mechanical HDDs, where higher priority entities may block others and random access costs are high.
Innovation Solution
A method that identifies and prioritizes low-cost I/O operations based on seek time and budget credits, ensuring fair access by dispatching I/O operations to entities with sufficient budget, and using de-randomization to optimize sequential and sparse I/O patterns, thereby reducing seek times and improving HDD throughput.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If higher priority entities are given preferential access to storage device, then response time for high priority entities is improved, but other entities are blocked from accessing the storage device
Solution Approach 1:
The patent implements dynamic priority adjustment where entity priorities are not fixed but change based on current system state, I/O queue conditions, and budget credit availability. The storage device mediator continuously adapts which entity receives preferential access, switching between entities based on real-time conditions rather than static priority assignments.
Solution Approach 2:
The system changes the parameter of entity priority dynamically. Instead of fixed priority levels, the mediator adjusts priority parameters based on current I/O queue depth, entity budget credits, and storage device utilization, allowing the system to optimize between response time and access availability by modifying priority parameters in response to changing conditions.
2Adaptability or versatility
If round-robin or random scheme is used to allow entities to access storage device, then all entities can access storage device, but thrashing occurs with large number of random I/O accesses
Solution Approach 1:
The storage device mediator performs preliminary actions by pre-scheduling and pre-ordering I/O operations based on predicted low-cost access patterns. Instead of reacting to random requests, the mediator proactively reorders I/O operations to create sequential or sparse access patterns before they are executed, reducing seek times and improving productivity.
Solution Approach 2:
The patent introduces a storage device mediator as an intermediary layer between entities and the storage device. This mediator reorders and schedules I/O operations to optimize access patterns, transforming random entity requests into sequential or sparse patterns that reduce thrashing and improve I/O efficiency while maintaining access availability for all entities.
3Productivity
If storage device bandwidth is shared among multiple entities, then resource utilization is improved, but I/O operation cost increases due to seek time
Solution Approach 1:
The patent merges multiple I/O operations from different entities into optimized sequential or sparse access patterns. By combining and reordering operations that access adjacent or nearby storage locations, the system reduces the total number of seek operations while maintaining bandwidth sharing, thereby reducing I/O operation costs without sacrificing resource utilization.
Solution Approach 2:
The storage device mediator acts as an intermediary that reorders I/O operations to minimize seek time. It analyzes the spatial and temporal characteristics of I/O requests from multiple entities and resequences them to create efficient access patterns, reducing the time loss from random seeking while preserving the benefits of shared bandwidth utilization.
4Adaptability or versatility
If I/O operations are dispatched without budget consideration, then all entities can access storage device freely, but bandwidth contention occurs among entities
Solution Approach 1:
The system implements feedback mechanisms where the storage device mediator monitors entity budget credit consumption and storage device utilization in real-time. Based on this feedback, the mediator dynamically adjusts I/O dispatch decisions, preventing entities with exhausted budgets from further access and prioritizing entities that can efficiently utilize available bandwidth, thus resolving bandwidth contention while maintaining fair access.
Solution Approach 2:
The patent applies dynamic budget management where entity access rights are not static but change based on consumed bandwidth and remaining budget credits. The mediator continuously adjusts which entities can access the storage device and at what rate, creating a dynamic system that prevents bandwidth contention by adapting access permissions to current utilization conditions.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system is configured to use a de-randomizer and budget data structure to economize I/O operations for a shared storage device while still allowing access to the device to a number of different entities. Embodiments can identify a comparatively low cost next operation as compared to other I/O operations, including a cost for seek time, for a first entity to dispatch to the storage device when the first entity has sufficient budget to have the I/O operation performed on its behalf and to identify an I/O operation for a second entity to dispatch to the storage device when there is insufficient budget for the first entity.