Zone Namespace Data Storage Controller Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data storage devices face performance degradation due to garbage collection, especially in high-performance data centers handling AI and big data, where varying data sizes from hosts require flexible writing methods without interference between multiple operating systems and application programs.
Innovation Solution
A data storage device with a controller that allocates zone namespaces (ZNS) physically and logically divided for each application program, allowing for flexible data writing by opening sub-buffer zones, segmenting write data, buffering, and migrating it to corresponding ZNS regions based on workload analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data is stored in traditionally allocated regions without ZNS, then the storage structure is simple, but performance degradation occurs due to garbage collection interference between multiple operating systems and application programs
Solution Approach 1:
The storage device is divided into multiple independent zone namespaces (ZNS), each allocated to specific operating systems or application programs. Each ZNS contains independently managed zones that can be accessed without interfering with other ZNS, thereby eliminating garbage collection interference between different workloads while maintaining manageable complexity through modular organization.
Solution Approach 2:
The patent introduces a new organizational dimension by implementing ZNS as a logical layer above the physical storage structure. This adds a namespace dimension that allows multiple independent data management domains to coexist on the same physical storage medium, enabling performance isolation without requiring separate physical storage devices for each workload.
2Adaptability or versatility
If the storage device opens all ZNS regions to handle varying data sizes flexibly, then adaptability to different workloads is improved, but power consumption increases due to unnecessary regions being active
Solution Approach 1:
The storage device dynamically opens and closes ZNS regions based on real-time workload requirements. When data is written to a particular ZNS, the corresponding region is opened for active operations. When no writes occur to a ZNS for a predetermined period, that region is automatically closed to reduce power consumption. This dynamic state management allows the system to maintain high adaptability to varying data sizes while minimizing energy usage by keeping only necessary regions active.
3Reliability
If garbage collection is performed frequently to maintain storage performance, then data retrieval efficiency is improved, but overall system productivity decreases due to the time consumed by internal write operations
Solution Approach 1:
The patent extracts the garbage collection function from the global storage management level and confines it to individual zone levels within each ZNS. Each zone performs its own garbage collection independently on its own data, preventing garbage collection operations in one ZNS from interfering with read/write operations in other ZNS. This extraction of the garbage collection function to the zone level maintains data retrieval efficiency within each zone while preserving overall system throughput by allowing parallel operations across multiple zones.
Data Source
AI summary
A data storage device includes a storage including a buffer zone and a data zone and a controller for exchanging data with the storage by allocating at least one zone namespace (ZNS) in the data zone, a ZNS being a data storage region that is physically and logically divided and allocated to each of application programs driven in a host. The controller opens one or more sub buffer zones in the buffer zone, divides write data from the host into one or more segments respectively corresponding to sizes of the one or more sub buffer zones, buffers each of the one or more segments in a corresponding one of the one or more sub buffer zones, opens a ZNS corresponding to a length of the write data in the data zone, and migrates the one or more segments buffered in the sub buffer zones to the opened ZNS.