Last-level Cache Replacement Policy Research Articles

LDF (less dirty first): dirtiness‐aware cache replacement policy for PCM main memory

A new last-level cache replacement policy for systems with a phase-change memory (PCM) main memory is presented. The proposed policy aims at reducing the write traffic to PCM by considering the fine-grained dirtiness of cache blocks when making a replacement decision. Experimental results show that the proposed policy reduces the write traffic to the PCM by 26 and 17% on average and up to 52 and 33% compared to not recently used and re-reference interval prediction, respectively.

Improving writeback efficiency with decoupled last-write prediction

In modern DDRx memory systems, memory write requests compete with read requests for available memory resources, significantly increasing the average read request service time. Caches are used to mitigate long memory read latency that limits system performance. Dirty blocks in the last-level cache (LLC) that will not be written again before they are evicted will eventually be written back to memory. We refer to these blocks as last-write blocks. In this paper, we propose an LLC writeback technique that improves DRAM efficiency by scheduling predicted last-write blocks early. We propose a low overhead last-write predictor for the LLC. The predicted last-write blocks are made available to the memory controller for scheduling. This technique effectively re-distributes the memory requests and expands writes scheduling opportunities, allowing writes to be serviced efficiently by DRAM. The technique is flexible enough to be applied to any LLC replacement policy. Our evaluation with multi-programmed workloads shows that the technique significantly improves performance by 6.5%-11.4% on average over the traditional writeback technique in an eight-core processor with various DRAM configurations running memory intensive benchmarks.