Zoned Namespace SSD Cache Server Storage Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In content delivery networks (CDNs), increasing the number or type of distribution servers leads to a decrease in cache hit rate, making it difficult to maintain high performance while keeping costs low.
Innovation Solution
Implementing a cache server with a zoned namespace (ZNS) solid state drive (SSD) that assigns multiple zones for data storage, allowing sequential writing and erasure on a zone-by-zone basis, reducing the need for over-provisioning capacity and minimizing garbage collection, thereby increasing storage capacity at a lower cost.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the storage capacity of the cache server is increased to improve cache hit rate, then the cache hit rate is improved, but the cost increases
Solution Approach 1:
The storage device is divided into multiple zones, each capable of independent sequential writing and erasure operations. This segmentation allows the cache server to efficiently manage storage capacity by treating each zone as an independent unit, reducing the need for over-provisioning while maintaining high cache hit rates through optimized zone-based data placement and eviction strategies.
2Quantity of substance
If traditional SSD storage is used, then data can be stored, but over-provisioning capacity is required and garbage collection is needed, increasing cost and complexity
Solution Approach 1:
The storage device is divided into multiple zones, each capable of independent sequential writing and erasure operations. This segmentation allows the cache server to efficiently manage storage capacity by treating each zone as an independent unit, reducing the need for over-provisioning while maintaining high cache hit rates through optimized zone-based data placement and eviction strategies.
Solution Approach 2:
The invention extracts and eliminates the need for garbage collection mechanisms by using zone-based sequential erasure. Instead of requiring complex garbage collection to manage worn blocks, the system performs targeted erasure of entire zones, simplifying the storage management architecture while maintaining full storage capacity.
3Productivity
If the number of distribution servers is increased to improve content delivery, then content delivery capability is improved, but cache hit rate decreases
Solution Approach 1:
The storage device is divided into multiple zones, each capable of independent sequential writing and erasure operations. This segmentation allows the cache server to efficiently manage storage capacity by treating each zone as an independent unit, reducing the need for over-provisioning while maintaining high cache hit rates through optimized zone-based data placement and eviction strategies.
Data Source
AI summary
An information processing apparatus includes a network interface, a storage device, and a processor. The processor is configured to assign a plurality of zones in the storage device. Each of the zones is a contiguous physical address range of the storage device that is mapped to a contiguous logical address range. The processor is configured to generate zone management information for each of the plurality of zones, store content received from the origin server via the network interface, in one of writable zones and update a writable address of the zone management information for the one of the writable zones. The processor is configured to operate to transmit the received content, and control the storage device to delete data stored therein in units of a zone upon a predetermined cache clearing criteria being met.