PA-Flow: Gradual Packet Aggregation at Virtual Network I/O for Efficient Service Chaining
PA-Flow: Gradual Packet Aggregation at Virtual Network I/O for Efficient Service Chaining
复制标题
PA-Flow：虚拟网络 I/O 处的渐进式数据包聚合，以实现高效的服务链
DOI：
--
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
H. Matsuo
中科院分区：
文献类型：
--
作者：
Yuki Taguchi;Ryota Kawashima;Hiroki Nakayama;Tsunemasa Hayashi;H. Matsuo
虚拟网络功能（VNF）的逻辑链接使得能够实现各种网络服务的玩具阻塞式组合。VNF通常基于在商用服务器上运行的虚拟机（VM）。然而，由于虚拟网络I/O开销，这种基于VM的VNF具有较差的分组转发性能，因此核心网络功能的虚拟化可能降低整体网络性能。在本文中，我们提出了一种新的数据包聚合方法（PA流），聚合具有相同的下一跳VNF的数据包，没有进一步的延迟。与现有的数据包聚合方法，我们的方法是有效的虚拟网络I/O，以减少主机内的数据包处理成本。此外，聚集的分组在通过服务链时被进一步聚集，这大大减少了核心网络处的业务量。我们已经实现了一个PA-Flow模块，它在DPDK/vhost-user框架下工作在虚拟交换机和QEMU之间。对于2-VNF服务链，PA-Flow显示出比默认的基于DPDK/vhost用户的系统高170%的吞吐量。此外，由于数据包数量的减少，延迟也改善了约2 us。
A logical chaining of Virtual Network Functions (VNFs) enables toy-blocking style composition of various network services. VNFs are generally based on Virtual Machines (VMs) running on commodity servers. However, such VM-based VNFs have poor packet forwarding performance because of the virtual network I/O overhead, and therefore virtualization of the core network functions can degrade overall network performance. In this paper, we propose a novel packet aggregation method (PA-Flow) that aggregates packets having a same next-hop VNF with no further latency. Unlike existing packet aggregation approaches, our method is effective for the virtual network I/O to reduce packet processing cost inside the host. Moreover, aggregated packets are further aggregated as going through the service chain, which drastically reduces traffic amount at the core network. We have implemented a PA-Flow module that works between a virtual switch and QEMU under the DPDK/vhost-user framework. PA-Flow showed 170% higher throughput than that of the default DPDK/vhost-user based system for a 2-VNF service chain. In addition, latency was also improved by about 2 us because of the packet amount reduction.
DOI：
10.1145/2674005.2674996
发表时间：
2014-12
期刊：
Proceedings of the 10th ACM International on Conference on emerging Networking Experiments and Technologies
影响因子：
--
作者：
Luo Mai;Lukas Rupprecht;Abdul Alim;Paolo Costa;Matteo Migliavacca;P. Pietzuch;A. Wolf
通讯作者：
Luo Mai;Lukas Rupprecht;Abdul Alim;Paolo Costa;Matteo Migliavacca;P. Pietzuch;A. Wolf