我要评分
获取效率
正确性
完整性
易理解

Introduction

The Ucache smart read cache uses smart I/O prefetch to accurately identify hotspot requests, prefetch I/Os of the sequential pattern, interval pattern and more, and load I/Os to the read cache in advance. In addition, the read cache uses the LRU algorithm to evict cold data, improving the I/O hit ratio and read performance.

The following figure shows the working principle of the Ucache smart read cache.

  • When an application delivers I/O requests, the I/O features are input to the DAS engine.
  • The DAS engine identifies the I/O patterns, predicts I/Os, and outputs the pattern of I/Os to be prefetched.
  • The read cache uses high-speed media (such as NVMe drives) to cache data. It maintains the mapping between cached data and data in main storage, identifies hotspot data from cached data, and uses the LRU algorithm to evict cold data, improving the cache hit ratio.

The DAS engine is implemented in the Kunpeng Storage Acceleration Library (KSAL). For details about DAS interfaces, see section 7.5 in the Kunpeng BoostKit 24.0.RC2 SDS KSAL Developer Guide. The following describes the implementation and installation of the read cache library.