IDIO: Orchestrating Inbound Network Data on Server Processors
IDIO: Orchestrating Inbound Network Data on Server Processors
复制标题
IDIO：在服务器处理器上编排入站网络数据
DOI：
10.1109/lca.2020.3044923
复制
发表时间：
2021
影响因子：
2.3
通讯作者：
Kim, Nam Sung
中科院分区：
文献类型：
--
作者：
Alian, Mohammad;Shin, Jongmin;Kang, Ki-Dong;Wang, Ren;Daglis, Alexandros;Kim, Daehoon;Kim, Nam Sung
互联网中心的网络带宽需求每12到15个月翻一番。为了满足这一需求，高带宽网络接口卡（每个都能每秒传输100比特的数据）正在进入下一代网络中心的服务器。服务器端点上这种前所未有的数据传输速率提出了新的挑战，因为内存层次结构中的入站网络流量放置决策对端到端性能有直接影响。现代服务器级英特尔处理器利用DDIO技术将所有入站网络数据导入末级缓存（LLC），而不管网络流量的性质如何。从性能和能效的角度来看，这种静态数据放置策略都不是最佳的。在这项工作中，我们设计IDIO，一个框架，不像DDIO，动态地决定在哪里放置入站网络流量在服务器的多级内存hierarchy.IDIO动态监视系统的行为，并区分不同的流量类，以确定和定期重新评估每个流的最佳位置：LLC，中级（L2）缓存或DRAM。我们的研究结果表明IDIO增加了服务器的最大可持续负载高达33.3%，在各种网络功能。
Network bandwidth demand in datacenters is doubling every 12 to 15 months. In response to this demand, high-bandwidth network interface cards, each capable of transferring 100s of Gigabits of data per second, are making inroads into the servers of next-generation datacenters. Such unprecedented data delivery rates on server endpoints raise new challenges, as inbound network traffic placement decisions within the memory hierarchy have a direct impact on end-to-end performance. Modern server-class Intel processors leverage DDIO technology to steer all inbound network data into the last-level cache (LLC), regardless of the network traffic’s nature. This static data placement policy is suboptimal, both from a performance and an energy efficiency standpoint. In this work, we designIDIO, a framework that—unlike DDIO—dynamically decides where to place inbound network traffic within a server’s multi-level memory hierarchy.IDIOdynamically monitors system behavior and distinguishes between different traffic classes to determine and periodically re-evaluate the best placement location for each flow: LLC, mid-level (L2) cache or DRAM. Our results show thatIDIOincreases a server’s maximum sustainable load by up to33.3% across various network functions.
减少共置虚拟网络功能的末级缓存干扰
DOI：
--
发表时间：
2019
期刊：
International Conference on Computer Communications and Networks
影响因子：
--
作者：
Raphael Durner;Christian Sieber;W. Kellerer
通讯作者：
W. Kellerer