High speed packet forwarding compiled from protocol independent data plane specifications

Laki, Sándor; Horpácsi, Dániel; Vörös, Péter; Kitlei, Róbert; Leskó, Dániel; Tejfel, Máté

doi:10.1145/2934872.2959080

Cited by 24 publications

(14 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It adopts DPDK poll-mode drivers to deliver packets, completely avoiding the overhead imposed by kernel stack and interrupt handling. t4p4s [21] is a platform-independent software switch specifically designed for P4 [33]. A compiler is implemented to generate switching code from P4 programs, and a hardware abstraction layer deals with platform-dependent details.…”

Section: A Software Switchesmentioning

confidence: 99%

“…HyperV [48] is a P4 dataplane hypervisor, and its DPDK target achieves comparable performance to PISCES. However, as detailed in [21], t4p4s outperforms it by two times running the baseline L2 forwarding application. We thus only consider t4p4s in our comparison.…”

Section: A Software Switchesmentioning

confidence: 99%

“…Our evaluation takes into account four switches designed with a self-contained architecture: VALE [13], VPP [65], t4p4s [21], and OvS-DPDK [18]. VALE is an L2 learning switch based on netmap, which can interconnect both physical NICs and virtual interfaces and forward packets at high speed.…”

Section: B Architecturementioning

confidence: 99%

“…The contribution of our work is twofold: (i) we describe our proposed experimental methodology for a fair evaluation and comparison of software switch performance; (ii) we apply our methodology to evaluate the performance of seven state-of-the-art software switches, namely Open vSwitch DPDK (OvS-DPDK) [18], FastClick [7], Berkeley Extensible Software Switch (BESS) [10], netmap suite [3], [13], [19], [20], Snabb [5], t4p4s [21], and FD.io VPP [22]. We start by analyzing the design space of these frameworks to build a basic understanding of their respective designs.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Comparing the performance of state-of-the-art software switches for NFV

Zhang

Linguaglossa

Gallo

et al. 2019

Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies

View full text Add to dashboard Cite

With the ultimate goal of replacing proprietary hardware appliances with Virtual Network Functions (VNFs) implemented in software, Network Function Virtualization (NFV) has been gaining popularity in the past few years. Software switches route traffic between VNFs and physical Network Interface Cards (NICs). It is of paramount importance to compare the performance of different switch designs and architectures. In this paper, we propose a methodology to compare fairly and comprehensively the performance of software switches. We first explore the design spaces of seven state-of-the-art software switches and then compare their performance under four representative test scenarios. Each scenario corresponds to a specific case of routing NFV traffic between NICs and/or VNFs. In our experiments, we evaluate the throughput and latency between VNFs in two of the most popular virtualization environments, namely virtual machines (VMs) and containers. Our experimental results show that no single software switch prevails in all scenarios. It is, therefore, crucial to choose the most suitable solution for the given use case. At the same time, the presented results and analysis provide a deeper insight into the design tradeoffs and identifies potential performance bottlenecks that could inspire new designs.

show abstract

Section: A Software Switchesmentioning

confidence: 99%

Section: A Software Switchesmentioning

confidence: 99%

Section: B Architecturementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Comparing the performance of state-of-the-art software switches for NFV

Zhang

Linguaglossa

Gallo

et al. 2019

Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies

View full text Add to dashboard Cite

show abstract

“…P4 is different from OpenFlow protocol; it has been designed to program and control the pipeline level of the data plane of SDN enabled routers and switches. The advent of P4 programming language introduced the new form of flexibility for users to control the network with a top-down approach [33][34][35][36].…”

Section: Related Workmentioning

confidence: 99%

A Dynamic Programmable Network for Large-Scale Scientific Data Transfer Using AmoebaNet

Shah

Noh

2019

Applied Sciences

View full text Add to dashboard Cite

Large scientific experimental facilities currently are generating a tremendous amount of data. In recent years, the significant growth of scientific data analysis has been observed across scientific research centers. Scientific experimental facilities are producing an unprecedented amount of data and facing new challenges to transfer the large data sets across multi continents. In particular, these days the data transfer is playing an important role in new scientific discoveries. The performance of distributed scientific environment is highly dependent on high-performance, adaptive, and robust network service infrastructures. To support large scale data transfer for extreme-scale distributed science, there is the need of high performance, scalable, end-to-end, and programmable networks that enable scientific applications to use the networks efficiently. We worked on the AmoebaNet solution to address the problems of a dynamic programmable network for bulk data transfer in extreme-scale distributed science environments. A major goal of the AmoebaNet project is to apply software-defined networking (SDN) technology to provide "Application-aware" network to facilitate bulk data transfer. We have prototyped AmoebaNet's SDN-enabled network service that allows application to dynamically program the networks at run-time for bulk data transfers. In this paper, we evaluated AmoebaNet solution with real world test cases and shown that how it efficiently and dynamically can use the networks for bulk data transfer in large-scale scientific environments. Appl. Sci. 2019, 9, 4541 2 of 17 for on-demand secure path reservation over the Wide Area Network (WAN), and these solutions also provide the automated, guaranteed bandwidth services for scientific workflows. The existing network paradigm for scientific workflows has been proven very successfully. However, it has been observed during our comprehensive study that the present network paradigm for an extreme-scale scientific data transfer job still needs some more improvements to reach its full potentials. We claimed that the current networking paradigm must address the following major problems: scalability, last mile, and dynamic programmability problems.The recent emerging concept in the network world is called Software-Defined Networking (SDN) [9][10][11]. This latest technology has been providing the new methods of configuration and management of networks. In SDN, the underlying network devices are simply considered as packets forwarding elements. It is possible to manage the control logic of the network centrally through a software program, which can effectively control the entire network behavior. In order to address such problems mentioned above, AmoebaNet [12] has been proposed.It is common to notice low efficiency in data movement when running the data transfer tools on DTN machines. Such inefficient data transmission is one of critical points in scientific data computing. Therefore, it is required to provide high-performance, predictable, and schedulable data movement...

show abstract

Software Defined Networking and Network Function Virtualization for Converged Access‐Metro Networks

Ruffini

Slyne

2019

Optical and Wireless Convergence for 5G Networks

View full text Add to dashboard Cite

High speed packet forwarding compiled from protocol independent data plane specifications

Cited by 24 publications

References 2 publications

Comparing the performance of state-of-the-art software switches for NFV

Comparing the performance of state-of-the-art software switches for NFV

A Dynamic Programmable Network for Large-Scale Scientific Data Transfer Using AmoebaNet

Software Defined Networking and Network Function Virtualization for Converged Access‐Metro Networks

Contact Info

Product

Resources

About