Speculative Block IO Aggregation for SCM Wear Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing information handling systems face uneven wear leveling of storage class memories (SCMs) due to intensive block I/O operations, leading to reduced durability and frequent replacements, especially in virtualized environments, where hypervisor management is unaware of wear leveling requirements, resulting in software overhead and alerts.
Innovation Solution
An information handling system with a disk manager that classifies block I/O jobs as intensive writes, invokes a speculative I/O aggregator module to create a delta disk image at a secondary SCM, redirecting the I/O job to this image, thereby distributing the load and promoting even wear across all DIMMs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If intensive block I/O operations are performed on SCM DIMMs, then system performance is improved, but wear leveling becomes uneven and durability decreases
Solution Approach 1:
The system segments the I/O workload by identifying intensive write operations and routing them through a separate path using delta disk images, while normal I/O operations continue to access the original SCM DIMM directly. This segmentation prevents intensive writes from causing uneven wear on the primary storage medium.
Solution Approach 2:
A disk manager intermediary is introduced that monitors I/O operations and selectively redirects intensive write operations to delta disk images. This intermediary layer manages the wear distribution by deciding which operations should be redirected, protecting the primary SCM DIMM from excessive wear while maintaining system performance.
2Reliability
If wear leveling is managed through software monitoring and alerts, then reliability is improved, but software overhead increases
Solution Approach 1:
The system performs preliminary action by proactively redirecting intensive write operations to delta disk images before wear problems occur. The disk manager identifies and redirects these operations in advance, preventing wear issues rather than reacting to them through alerts and corrective actions.
Solution Approach 2:
The system extracts the wear-causing intensive write operations from the normal I/O path and handles them separately through delta disk images. This extraction removes the harmful factor (intensive writes) from the primary SCM DIMM while maintaining the necessary functionality.
3Reliability
If delta disk images are created for intensive writes, then wear distribution is improved, but device complexity increases
Solution Approach 1:
The system applies local quality by creating delta disk images only for specific intensive write operations rather than for all I/O operations. This selective approach distributes wear evenly only where needed (on intensive writes) while leaving normal operations to use the primary SCM DIMM, avoiding unnecessary complexity.
Data Source
AI summary
An information handling system may include at least one processor and a non-transitory memory coupled to the at least one processor. The information handling system may be configured to implement a disk manager configured to receive a command to perform a block input/output (I/O) job at a first physical storage resource. The disk manager may, based on characteristics of the block I/O job, classify the block I/O job as an intensive write. In response to the block I/O job being classified as an intensive write, the disk manager may invoke a speculative I/O aggregator module of the disk manager. The disk manager may create a delta disk image for the first physical storage resource at a second physical storage resource, and redirect the block I/O job to the delta disk image at the second physical storage resource.