P8: P4 With Predictable Packet Processing Performance

被引:35
作者
Harkous, Hasanin [1 ,2 ]
Jarschel, Michael [2 ]
He, Mu [1 ]
Pries, Rastin [2 ]
Kellerer, Wolfgang [1 ]
机构
[1] Tech Univ Munich, LKN, D-80333 Munich, Germany
[2] Nokia Bell Labs, Standardizat Lab, D-81541 Munich, Germany
来源
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT | 2021年 / 18卷 / 03期
关键词
Pipelines; Performance evaluation; Program processors; Hardware; Switches; Complexity theory; programmable data plane; P4; software-defined networking; modeling; device benchmarking; SOFTWARE-DEFINED NETWORKS;
D O I
10.1109/TNSM.2020.3030102
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Data plane programmability brings network flexibility to a new level. However, it introduces the complexity of the data path's program as a new factor that influences packet forwarding latency and thus devices' performance. Accurate identification of the relation between data path complexity and packet forwarding latency enables the design and management of networks with predictable performance. In this article, we leverage the characteristics of P4 programming language to provide a method for estimating the packet forwarding latency as a function of the data path program. We analyze the impact of different P4 constructs on packet processing latency for three state-of-the-art P4 devices: Netronome SmartNIC, NetFPGA-SUME, and T4P4S DPDK-based software switch. Besides comparing the performance of these three targets, we use the derived results to propose a method for estimating the average packet latency, at compilation time, of arbitrary P4-based network functions implemented using the surveyed P4 constructs. The proposed method is finally validated using a set of realistic network functions, which shows that our method estimates the average packet latency with sub-microsecond precision.
引用
收藏
页码:2846 / 2859
页数:14
相关论文
共 50 条
[1]  
[Anonymous], 2013, Queue, DOI DOI 10.1145/2559899.2560327
[2]   An Accurate and Efficient Modeling Framework for the Performance Evaluation of DPDK-Based Virtual Switches [J].
Begin, Thomas ;
Baynat, Bruno ;
Gallardo, Guillaume Artero ;
Jardin, Vincent .
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT, 2018, 15 (04) :1407-1421
[3]   P4-To-VHDL: Automatic generation of high-speed input and output network blocks [J].
Benacek, Pavel ;
Pus, Viktor ;
Kubatova, Hana ;
Cejka, Tomas .
MICROPROCESSORS AND MICROSYSTEMS, 2018, 56 :22-33
[4]  
Bifulco R, 2018, IEEE INT CONF HIGH
[5]  
Blenk A., 2019, 2019 ACMIEEE S ARCHI, P1
[6]   Programming Protocol-Independent Packet Processors [J].
Bosshart, Pat ;
Daly, Dan ;
Gibb, Glen ;
Izzard, Martin ;
McKeown, Nick ;
Rexford, Jennifer ;
Schlesinger, Cole ;
Talayco, Dan ;
Vahdat, Amin ;
Varghese, George ;
Walker, David .
ACM SIGCOMM COMPUTER COMMUNICATION REVIEW, 2014, 44 (03) :87-95
[7]   Forwarding Metamorphosis: Fast Programmable Match-Action Processing in Hardware for SDN [J].
Bosshart, Pat ;
Gibb, Glen ;
Kim, Hun-Seok ;
Varghese, George ;
McKeown, Nick ;
Izzard, Martin ;
Mujica, Fernando ;
Horowitz, Mark .
ACM SIGCOMM COMPUTER COMMUNICATION REVIEW, 2013, 43 (04) :99-110
[8]  
Bressana P., 2020, P ACM SIGCOMM WORKSH, P1
[9]   NFV-Bench: A Dependability Benchmark for Network Function Virtualization Systems [J].
Cotroneo, Domenico ;
De Simone, Luigi ;
Natella, Roberto .
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT, 2017, 14 (04) :934-948
[10]  
D. Project, 2010, DPDK: Data plane development kit