Search for: All records

This paper describes a cloud infrastructure and virtual laboratories on P4 programmable data plane switches. P4 programmable data planes emerged as a technology that enables innovation in networking. P4 is a programming language used to describe how network packets are processed. This paper explains an entry-level training library on P4. The virtual laboratories introduce the learner to P4 and data plane concepts by providing step-by-step guides and exercises. The virtual laboratories are hosted in the Academic Cloud, a distributed platform that manages and orchestrates computing resources. Additionally, the paper describes a work in progress of P4 virtual laboratories that uses Intel Tofino switches. Lastly, the paper discusses the use of the Academic Cloud as a network testbed.

Google published the first release of the Bottleneck Bandwidth and Round-trip Time (BBR) congestion control algorithm in 2016. Since then, BBR has gained a widespread attention due to its ability to operate efficiently in the presence of packet loss and in scenarios where routers are equipped with small buffers. These characteristics were not attainable with traditional loss-based congestion control algorithms such as CUBIC and Reno. BBRv2 is a recent congestion control algorithm proposed as an improvement to its predecessor, BBRv1. Preliminary work suggests that BBRv2 maintains the high throughput and the bounded queueing delay properties of BBRv1. However, the literature has been missing an evaluation of BBRv2 under different network conditions. This paper presents an experimental evaluation of BBRv2 Alpha (v2alpha-2019-07-28) on Mininet, considering alternative active queue management (AQM) algorithms, routers with different buffer sizes, variable packet loss rates and round-trip times (RTTs), and small and large numbers of TCP flows. Emulation results show that BBRv2 tolerates much higher random packet loss rates than loss-based algorithms but slightly lower than BBRv1. The results also confirm that BBRv2 has better coexistence with loss-based algorithms and lower retransmission rates than BBRv1, and that it produces low queuing delay even with large buffers.more »When a Tail Drop policy is used with large buffers, an unfair bandwidth allocation is observed among BBRv2 and CUBIC flows. Such unfairness can be reduced by using advanced AQM schemes such as FQ-CoDel and CAKE. Regarding fairness among BBRv2 flows, results show that using small buffers produces better fairness, without compromising high throughput and link utilization. This observation applies to BBRv1 flows as well, which suggests that rate-based model-based algorithms work better with small buffers. BBRv2 also enhances the coexistence of flows with different RTTs, mitigating the RTT unfairness problem noted in BBRv1. Lastly, the paper presents the advantages of using TCP pacing with a loss-based algorithm, when the rate is manually configured a priori. Future algorithms could set the pacing rate using explicit feedback generated by modern programmable switches.« less

The alpha version of Bottleneck Bandwidth and Round-trip Time version 2 (BBRv2) has been recently presented, which aims to mitigate the shortcomings of its predecessor, BBR version 1 (BBRv1). Previous studies show that BBRv1 provides a high link utilization and low queuing delay by estimating the available bottleneck bandwidth. However, its aggressiveness induces unfairness when flows i) use different congestion control algorithms, such as CUBIC, and ii) have distinct round-trip times (RTTs). This paper presents an experimental evaluation of BBRv2, using Mininet. Results show that the coexistence between BBRv2-CUBIC is enhanced with respect to that of BBRv1-CUBIC, as measured by the fairness index. They also show that BBRv2 mitigates the RTT unfairness problem observed in BBRv1. Additionally, BBRv2 achieves a better fair share of the bandwidth than its predecessor when network conditions such as bandwidth and latency dynamically change. Results also indicate that the average flow completion time of concurrent flows is reduced when BBRv2 is used.

Ever since the inception of the networking industry, routing and switching devices have been limited to tightly-coupled hardware and software components. Vendors provide closed source proprietary stacks, restraining network operators from utilizing customized features, and hence hindering innovation. This aggregated model is costly, time consuming, and unscalable as changes in the devices require vendor's intervention. As a result, the industry started manufacturing white-box switches and developing Network Operating Systems (NOSs) that support multiple vendors and Application Specific Integrated Circuits (ASICs). This model is referred to as ”disaggregated” as the software and hardware are decoupled; essentially, vendors' switching silicons (e.g., Broadcom) are compatible with different NOS (e.g., SONiC). In this paper, we discuss the lessons learned while designing and implementing a testbed that consists of disaggregated network devices. We iterate over several open source Internet Protocol (IP) routing suites and NOSs that are vendor-agnostic. Additionally, we highlight a novel type of forwarding data planes that are programmable and explore their features. The testbed consists of two white-box switches provided by Edgecore that use programmable switching silicon (Tofino) manufactured by Barefoot Networks, an Intel Company. We installed SONiC NOS on top of the switches and tested static and BGP routing protocols. Wemore »report the configuration process and the prerequisites needed to deploy a working disaggregated environment. Finally, we discuss how open source NOSs and programmable switches can be extended to support campus networks, rather than being data center-oriented only.« less