Dynamic Set Stealing to Improve Cache Performance
Dynamic Set Stealing to Improve Cache Performance
复制标题
DOI：
10.1109/sbac-pad55451.2022.00017
复制
发表时间：
2022-11
期刊：
影响因子：
--
通讯作者：
Brady Testa;Samira Mirbagher Ajorpaz;Daniel A. Jiménez
中科院分区：
文献类型：
--
作者：
Brady Testa;Samira Mirbagher Ajorpaz;Daniel A. Jiménez
在末级缓存(LLC)中，替换策略取决于工作负载特性。使这项政策适应当前的工作量一直是一个活跃的研究领域。以前的工作包括集合决斗，例如DIP[40]，它使用静态替换策略和基于机器学习的模型，如Glider[45]或多视角重用预测[19]。两者都比最近最少使用(LRU)策略有所改进，但也有可能进一步改进。DIP面临着资源浪费的问题，因为相互竞争的主要政策集的规模是固定的。每种机器学习方法都使用一组固定的功能，这些功能是离线选择的，并不是对所有工作负载都是最佳的。我们引入了集合窃取与感知器表(SSPT)，这是一种新的资源管理策略，允许结合许多替换策略的优点来最大化性能并消除浪费的处理器资源。在一组81个SPEC基准工作负载上，该策略在基准LRU策略上实现了9.45%的几何平均加速比，而Glider的加速比为9.28%，多视角重用预测的加速比为7.62%(假设2 MB LLC)。在一组90个大数据(Gap和XS)工作负载上，我们在LRU上实现了9.70%的几何平均加速比，相比之下，Glider的加速比为8.79%，多视角重用预测的加速比为7.86%(假设LLC为2 MB)。
In the last-level cache (LLC), replacement policy is dependent on workload characteristics. Adapting the policy to the current workload has been an active area of research. Previous works includes set dueling exemplified by DIP [40] which uses static replacement policies and machine learning based models such as Glider [45] or Multiperspective Reuse Prediction [19]. Both provide improvement over the least-recently-used (LRU) policy, but additional improvement is possible. DIP suffers from wasted resources as leading sets of the competing policies are fixed in size. Machine learning approaches each use a fixed set of features that were selected offline that are not optimal for all workloads. We introduce Set Stealing with Perceptron Tables (SSPT), a novel resource management policy that allows combining the strengths of many replacement policies to maximize performance and eliminate wasted processor resources. This policy achieves a 9.45% geometric mean speedup over a baseline LRU policy on a set of 81 SPEC benchmark work-loads, compared to Glider's 9.28% and Multiperspective Reuse Prediction at 7.62% assuming a 2 MB LLC. We achieve a 9.70% geometric mean speedup over LRU on a set of 90 big data (GAPS and XS) workloads, compared to Glider's 8.79% and Multiperspective Reuse Prediction at 7.86% assuming a 2 MB LLC.