Discrete-Time Modeling of NFV Accelerators that Exploit Batched Processing

Abstract

Network Functions Virtualization (NFV) is among the latest network revolutions, promising increased flexibility and avoiding network ossification. At the same time, all-software NFV implementations on commodity hardware raise performance issues when comparing to ASIC solutions. To address these issues, numerous software acceleration frameworks for packet processing have been proposed in the last few years. One central mechanism of many of these frameworks is the use of batching techniques , where packets are processed in groups as opposed to individually. This is required to provide high-speed capabilities by minimizing framework overhead, reducing interrupt pressure, and leveraging instruction-level cache hits. Several such system implementations have been proposed and experimentally benchmarked in the past. However, the scientific community has so far only to a limited extent attempted to model the system dynamics of modern NFV routers exploiting batching acceleration. In this article, we propose a simple, generic model for this type of batching-based systems that can be applied to predict all relevant key performance indicators. In particular, we extend our previous work and formulate the calculation of the queue size as well as waiting time distributions in addition to the batch size distribution and the packet loss probability. Furthermore, we introduce the waiting time distribution as a relevant QoS parameter and perform an in-depth parameter study, widening the set of investigated variables as well as the range of values. Finally, we contrast the model prediction with experimental results gathered in a high-speed testbed including an NFV router, showing that the model not only correctly captures system performance under simple conditions, but also in more realistic scenarios in which traffic is processed by a mixture of functions.

Performance Evaluation of Next-Generation Data Plane Architectures and their Components

“…Performance Monitoring [18] Performance Modeling [1,2,19] Control Plane Performance [10,14] Traffic Modeling [6,17,22,23] Network Security [15] Data Plane Abstraction [3,12,20] Methodology [21] Simulative Performance Analysis Performance Modeling. Subsequently, in Chapter 2 of this thesis, we apply our monitoring approach to obtain measurement values for a state-of-the-art software router.…”

“…[19] • Geißler, S., Lange, S., Linguaglossa, L., Rossi, D., Zinner, T., Hoßfeld, T.: "Discrete-Time Modeling of NFV Accelerators that Exploit Batched Processing, " in ACM Transactions on Modeling and Performance Evaluation of Computing Systems (ToMPECS), 2021. [1] 2.1 Background and Related Work…”

“…However, reliable and accurate measurement of network function specific KPIs, like packet processing times, is not only relevant in live environments and for monitoring purposes. Instead, the processing time can be used as an input parameter for theoretical models ahead of deployment when it comes to predicting performance under certain circumstances [19,1,2]. Additionally, the softwarization of networks is often accompanied by the application of software development paradigms during network function development.…”

“…In the following, the architecture, monitoring logic, and a proof-of-concept implementation of KOMon [18] (Kernel-based Online Monitoring) is discussed, before a case study using an exemplary VNF is conducted. The code for both the monitoring tool and the VNF used in the case study can be found on GitHub 1 . In the presented use case, we use the network stack of the Linux Kernel v4.11 [97].…”

“…In addition, we have shown that the offset included in the measurement is statistically independent of the magnitude of the processing times of the network function. This allows for KOMon to be used to determine the distribution of processing times as it is needed for theoretical performance models such as [19,1,2]. In addition, this offset can be eliminated from the values as it is possible to calibrate the system by performing measurements involving a network function of which the processing time is known.…”