Chained Nonvolatile Storage Server Load Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hybrid drives with SSD and HDD combinations face bottlenecks when SSDs become saturated, leading to increased latency and cascading issues due to lack of collaboration among multiple hybrid drives, resulting in inefficient load balancing and premature service disruption.
Innovation Solution
Implementing a storage server that chains nonvolatile storage devices to reroute write requests and maintain a cache index for data blocks, allowing diversion to alternate devices and maintaining data coherence across multiple devices, thereby preventing bottlenecks and ensuring high availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multiple hybrid drives are used to increase storage capacity, then storage capacity is improved, but load balancing deteriorates because drives operate independently without collaboration
Solution Approach 1:
The patent merges multiple independent hybrid drives into a collaborative storage system where drives share information about their SSD saturation status. The storage manager consolidates the state of all hybrid drives to make coordinated load balancing decisions, transforming independent operations into a unified system that prevents cascading bottlenecks.
Solution Approach 2:
The system implements feedback mechanisms where the storage manager continuously monitors SSD saturation levels across multiple hybrid drives and uses this information to dynamically reroute storage activity. This feedback loop enables proactive load balancing by redirecting writes before bottlenecks develop, rather than reacting after saturation occurs.
2Productivity
If storage activity is rerouted to remote federated replicas when SSD becomes saturated, then load balancing is attempted, but latency increases due to high latency networking
Solution Approach 1:
The storage manager performs preliminary assessment of local hybrid drive availability before initiating remote rerouting. By monitoring SSD saturation proactively and identifying alternative local drives in advance, the system can redirect storage activity to nearby resources without incurring remote networking latency, only using federated replicas as a last resort.
3Reliability
If hybrid drive is taken out of service when SSD becomes saturated, then data integrity is protected, but availability deteriorates and remedial activity increases
Solution Approach 1:
The system dynamically adjusts the service status of hybrid drives based on real-time SSD saturation levels and alternative drive availability. Rather than statically taking drives offline when saturation occurs, the storage manager continuously evaluates system state and can bring drives back online when conditions improve, optimizing both data integrity protection and availability.
4Reliability
If remedial activity such as resilvering is performed when drive is taken out of service, then data integrity is restored, but bandwidth consumption increases and time is lost
Solution Approach 1:
The storage manager performs preliminary rerouting to alternative local hybrid drives before drive failure or saturation occurs, preventing the need for subsequent resilvering operations. By maintaining redundant local copies and proactively managing load distribution, the system avoids the bandwidth-intensive remedial activity that would otherwise be required to restore data integrity.
Data Source
AI summary
Techniques herein are for chaining nonvolatile storage devices to achieve high availability. A method involves a storage server receiving a write request to store data blocks in a first nonvolatile memory device. The storage server comprises a plurality of nonvolatile memory devices that cache data blocks stored on primary storage. The plurality of nonvolatile memory devices comprises the first nonvolatile memory device. The storage server maintains a cache index of data blocks that reside in the plurality of nonvolatile memory devices. Based on one or more criteria, the storage server reroutes the write request to a second nonvolatile memory device of the plurality of nonvolatile memory devices and stores an identifier of the second nonvolatile memory device in the cache index.