Zero-CPU Collection with Direct Telemetry Access
Zero-CPU Collection with Direct Telemetry Access
复制标题
DOI：
10.1145/3484266.3487366
复制
发表时间：
2021-10
期刊：
影响因子：
--
通讯作者：
Jonatan Langlet;Ran Ben Basat;Sivaramakrishnan Ramanathan;G. Oliaro;M. Mitzenmacher;Minlan Yu;G. Antichi
中科院分区：
文献类型：
--
作者：
Jonatan Langlet;Ran Ben Basat;Sivaramakrishnan Ramanathan;G. Oliaro;M. Mitzenmacher;Minlan Yu;G. Antichi
可编程开关正在推动细粒度测量的大幅增长。这给必须处理来自许多交换机的报告的遥测收集器带来了巨大的压力。过去的研究通过提高收集器的堆栈性能或限制从交换机发送的数据量来认识到这个问题。在本文中，我们采用了一种不同的激进方法：交换机负责直接将可查询的遥测数据插入收集器的内存中，绕过它们的CPU，从而提高它们的收集可扩展性。我们建议使用一种称为直接遥测访问的方法，其中交换机联合将遥测报告直接写入同一收集器的内存区域，无需协调。我们的解决方案 DART 是概率性的，用内存冗余和查询成功概率换取收集器上的 CPU 资源。我们使用 P4 交换机和 RDMA NIC 等商用硬件对 DART 进行原型设计，并表明我们可以通过合理的内存开销获得较高的查询成功率。例如，我们可以收集胖树拓扑上的 INT 路径跟踪信息，而无需收集器 CPU 的参与，同时实现 99.9% 的查询成功概率，并且每个流仅使用 300 字节。
Programmable switches are driving a massive increase in fine-grained measurements. This puts significant pressure on telemetry collectors that have to process reports from many switches. Past research acknowledged this problem by either improving collectors' stack performance or by limiting the amount of data sent from switches. In this paper, we take a different and radical approach: switches are responsible for directly inserting queryable telemetry data into the collectors' memory, bypassing their CPU, and thereby improving their collection scalability. We propose to use a method we call direct telemetry access, where switches jointly write telemetry reports directly into the same collector's memory region, without coordination. Our solution, DART, is probabilistic, trading memory redundancy and query success probability for CPU resources at collectors. We prototype DART using commodity hardware such as P4 switches and RDMA NICs and show that we get high query success rates with a reasonable memory overhead. For example, we can collect INT path tracing information on a fat tree topology without a collector's CPU involvement while achieving 99.9% query success probability and using just 300 bytes per flow.