Spandex: A Flexible Interface for Efficient Heterogeneous Coherence

Spandex: A Flexible Interface for Efficient Heterogeneous Coherence
复制标题

氨纶:实现高效异质相干的灵活接口

DOI:
--
复制
发表时间:
2018
期刊:
International Symposium on Computer Architecture
影响因子:
--
通讯作者:
S. Adve
S. Adve
中科院分区:
--
文献类型:
--
作者:
Johnathan Alsop;Matthew D. Sinclair;S. Adve

文献摘要

被引文献

相似文献

最近的异质体系结构趋向于更严格的集成,并且共享的内存很大程度上是由于这种转变启用了有效的通信和可编程性。由MESI等硬件连贯协议支持,而其他人则喜欢轻巧的软件连贯协议或使用具有不同状态和通信粒度的Scratchpads(例如,现代解决方案)往往会扩展现有的中间式CPU COHERESS,通常是通过添加层次构造协议,通过中间共享的缓存。尽管在功能上,这些策略缺乏灵活性,并且通常会遭受性能限制,使其对于某些新兴的加速器和工作负载而言,我们需要有效地集成现有设备的灵活界面 - 我们的内存结构的侵入性更改。由于潜水员的一致性和内存的要求,每个设备能够以适合其特定访问属性的方式进行通信。相对于基于MESI的层次解决方案,氨纶平均而言,氨纶将执行时间降低16%(最大29%)和27%(最大58%)。
Recent heterogeneous architectures have trended toward tighter integration and shared memory largely due to the efficient communication and programmability enabled by this shift. However, such integration is complex, because accelerators have widely disparate methods for accessing and keeping data coherent. Some processors use caches %that are backed by hardware coherence protocols like MESI, while others prefer lightweight software coherence protocols or use specialized memories like scratchpads with differing state and communication granularities. Modern solutions tend to build interfaces that extend existing MESI-style CPU coherence protocols, often by adding hierarchical indirection through intermediate shared caches. Although functionally correct, these strategies lack flexibility and generally suffer from performance limitations that make them sub-optimal for some emerging accelerators and workloads. Instead, we need a flexible interface that can efficiently integrate existing and future devices – without requiring intrusive changes to their memory structure. We introduce Spandex, an improved coherence interface based on the simple and scalable DeNovo coherence protocol. Spandex (which takes its name from the flexible material commonly used in one-size-fits-all textiles) directly interfaces devices with diverse coherence properties and memory demands, enabling each device to communicate in a manner appropriate for its specific access properties. We demonstrate the importance of this flexibility by comparing this strategy against a more conventional MESI-based hierarchical solution for a diverse range of heterogeneous applications. On average for the applications studied, Spandex reduces execution time by 16% (max 29%) and network traffic by 27% (max 58%) relative to the MESI-based hierarchical solution.