A SmartNIC-Accelerated Monitoring Platform for In-band Network Telemetry
A SmartNIC-Accelerated Monitoring Platform for In-band Network Telemetry
复制标题
DOI：
10.1109/lanman49260.2020.9153279
复制
发表时间：
2020-07
期刊：
影响因子：
--
通讯作者：
Yixiao Feng;Sourav Panda;Sameer G. Kulkarni;K. Ramakrishnan;N. Duffield
中科院分区：
文献类型：
--
作者：
Yixiao Feng;Sourav Panda;Sameer G. Kulkarni;K. Ramakrishnan;N. Duffield
带内网络遥测技术（INT）的最新发展通过收集数据平面上的信息来对网络元素的性能和负载进行细粒度监控。INT使流量源能够在数据包中嵌入遥测指令，避免单独探测或不频繁的基于管理的监控。INT汇聚节点通过检索由不同INT信息源附加的INT元数据指令来跟踪和收集指标。然而，跟踪到达接收器的数据包中的INT状态是计算密集型的（需要对每个数据包进行复杂的操作），并且对标准P4匹配-操作数据包处理管道保持线路速率具有挑战性。我们提出了一个网络遥测平台，其中INT接收器在监控主机的SmartNIC上使用不同的（基于c的）算法实现，补充了P4数据包处理管道。该设计加速了数据包处理，并比单独的P4匹配操作处理更有效地处理复杂的int相关操作。当P4管道解析INT头时，一个通用的Micro-C算法执行复杂的INT任务（例如聚合、事件检测、通知等）。我们证明，与仅使用p4的实现相比，INT处理的分区显著降低了处理开销，提供了准确、及时和几乎无丢失的事件通知。
Recent developments in In-band Network Telemetry (INT) provide granular monitoring of performance and load on network elements by collecting information in the data plane. INT enables traffic sources to embed telemetry instructions in data packets, avoiding separate probing or infrequent management-based monitoring. INT sink nodes track and collect metrics by retrieving INT metadata instructions appended by different sources of INT information. However, tracking the INT state in packets arriving at the sink is both compute intensive (requiring complex operations on each packet), and challenging for the standard P4 match-action packet processing pipeline to maintain line-rate. We propose a network telemetry platform in which the INT sink is implemented using distinct (C-based) algorithms on a SmartNIC in the monitoring host, complementing the P4 packet processing pipeline. This design accelerates packet processing and handles complex INT-related operations more efficiently than P4 match-action processing alone. While the P4 pipeline parses INT headers, a general-purpose Micro-C algorithms performs complex INT tasks (e.g. aggregation, event-detection, notification, etc.). We demonstrate that partitioning of INT processing significantly reduces processing overhead vs. a P4-on1y implementation, providing accurate, timely and almost loss-free event notification.