Garibaldi: A Pairwise Instruction-Data Management for Enhancing Shared Last-Level Cache Performance in Server Workloads

TL;DR

Garibaldi introduces a pairwise instruction-data cache management scheme that improves server workload performance by better preserving hot instructions and prefetching data, reducing LLC misses and frontend bottlenecks.

Contribution

This work presents a novel pairwise instruction-data management scheme, Garibaldi, that enhances LLC performance by coupling instruction and data access hotness and applying selective protection and prefetching.

Findings

Achieves 13.2% CPU performance improvement on baseline LLC

Achieves 6.1% CPU performance improvement on Mockingjay LLC

Effectively reduces instruction cache misses in server workloads

Abstract

Modern CPUs suffer from the frontend bottleneck because the instruction footprint of server workloads exceeds the private cache capacity. Prior works have examined the CPU components or private cache to improve the instruction hit rate. The large footprint leads to significant cache misses not only in the core and faster-level cache but also in the last-level cache (LLC). We observe that even with an advanced branch predictor and instruction prefetching techniques, a considerable amount of instruction accesses descend to the LLC. However, state-of-the-art LLC designs with elaborate data management overlook handling the instruction misses that precede corresponding data accesses. Specifically, when an instruction requiring numerous data accesses is missed, the frontend of a CPU should wait for the instruction fetch, regardless of how much data are present in the LLC. To preserve hot instructions in the LLC, we propose Garibaldi, a novel pairwise instruction-data management scheme. Garibaldi tracks the hotness of instruction accesses by coupling it with that of data accesses and adopts management techniques. On the one hand, this scheme includes a selective protection mechanism that prevents the cache evictions of high-cost instruction cachelines. On the other hand, in the case of unprotected instruction line misses, Garibaldi conservatively issues prefetch requests of the paired data lines while handling those misses. In our experiments, we evaluate Garibaldi with 16 server workloads on a 40-core machine. We also implement Garibaldi on top of a modern LLC design, including Mockingjay. Garibaldi improves 13.2% and 6.1% of CPU performance on baseline LLC design and Mockingjay, respectively.