MIND: In-Network Memory Management for Disaggregated Data Centers
MIND: In-Network Memory Management for Disaggregated Data Centers
复制标题
DOI：
10.1145/3477132.3483561
复制
发表时间：
2021-07
期刊：
影响因子：
--
通讯作者：
Seung-seob Lee;Yanpeng Yu;Yupeng Tang;Anurag Khandelwal;Lin Zhong;A. Bhattacharjee
中科院分区：
文献类型：
--
作者：
Seung-seob Lee;Yanpeng Yu;Yupeng Tang;Anurag Khandelwal;Lin Zhong;A. Bhattacharjee
内存分解通过将内存和计算物理分离到网络连接的资源“刀片”中，保证了数据中心的透明弹性、高资源利用率和硬件异构性。然而，现有设计以牺牲资源弹性为代价来实现性能，将内存共享限制在单个计算刀片上，以避免网络上昂贵的内存一致性流量。在这项工作中，我们展示了新兴的可编程网络交换机可以通过在网络结构中放置内存管理逻辑来为分解架构实现高效的共享内存抽象。我们发现，在网络中集中内存管理可以实现网络内缓存一致性协议的带宽和延迟高效实现，而可编程交换机 ASIC 则可以线速支持其他内存管理逻辑。我们通过 MIND1（一种用于机架规模分解的网络内存管理单元）实现了这些见解。 MIND 可实现透明的资源弹性，同时与实际工作负载的先前内存分解建议的性能相匹配。
Memory disaggregation promises transparent elasticity, high resource utilization and hardware heterogeneity in data centers by physically separating memory and compute into network-attached resource "blades". However, existing designs achieve performance at the cost of resource elasticity, restricting memory sharing to a single compute blade to avoid costly memory coherence traffic over the network. In this work, we show that emerging programmable network switches can enable an efficient shared memory abstraction for disaggregated architectures by placing memory management logic in the network fabric. We find that centralizing memory management in the network permits bandwidth and latency-efficient realization of in-network cache coherence protocols, while programmable switch ASICs support other memory management logic at line-rate. We realize these insights into MIND1, an in-network memory management unit for rack-scale disaggregation. MIND enables transparent resource elasticity while matching the performance of prior memory disaggregation proposals for real-world workloads.