Traffic shaping is a critical function for the efficient operation of modern networks, with applications ranging from data center networks to home routers. To correctly fulfill their expected task, traffic shapers must keep up with increasing line rates. While it is possible to scale traffic shapers across multiple CPUs via hardware queues, in some cases — such as when enforcing a global rate limit — these algorithms underperform due to lock contentions. This is especially true within the Linux kernel, where scheduling policies are realized as so-called queuing disciplines (qdiscs) and enforcing a global rate limit can only be achieved on what is effectively a single CPU core.In this work, we design and implement a lockless synchronization mechanism that allows qdiscs to efficiently scale rate limiting across multiple hardware queues. To demonstrate its practicality, we integrate this mechanism into the CAKE qdisc, enabling multiple CAKE instances to operate under the MQ qdisc while maintaining a global rate limit. We perform an extensive performance evaluation and find that the implementation achieves close to perfect scaling across cores, with an accuracy deviation with less that 0.25% of the configured rate, while keeping latencies low.

While Internet Service Providers (ISPs) have traditionally focused on marketing network throughput, it is becoming increasingly recognized that network latency also plays a significant role for the quality of experience. However, many ISPs lack the means to continuously monitor the latency of their network. In this work, we present a method to continuously monitor and aggregate network latency per subnet directly in the Linux kernel by leveraging eBPF. We deploy this solution on a middlebox in an ISP network and collect an extensive dataset of latency measurements for both the internal and external parts of the network. We find that our monitoring solution can monitor all subscriber traffic while maintaining a low overhead of only around 1% additional CPU utilization. Our analysis of the latency data reveals a wide latency tail in the last-mile access, which grows during busy periods in the evening. Furthermore, we dissect the external network latency and uncover the latency profiles for the most popular autonomous systems.

Abstract Network latency is a critical factor for the perceived quality of experience for many applications. With an increasing focus on interactive and real-time applications, which require reliable and low latency, the ability to continuously and efficiently monitor latency is becoming more important than ever. Always-on passive monitoring of latency can provide continuous latency metrics without injecting any traffic into the network. However, software-based monitoring tools often struggle to keep up with traffic as packet rates increase, especially on contemporary multi-Gbps interfaces. We investigate the feasibility of using eBPF to enable efficient passive network latency monitoring by implementing an evolved Passive Ping (ePPing). Our evaluation shows that ePPing delivers accurate RTT measurements and can handle over 1 Mpps, or correspondingly over 10 Gbps, on a single core, greatly improving on state-of-the-art software based solutions, such as PPing.

Since the definition of the bufferbloat phenomenon, several Linux kernel modules have been introduced in its TCP/IP stack, and there is a lack of experimental studies on their effects when coupled with WLAN technologies, in particular, IEEE 802.11n and IEEE 802.11ac. One essential algorithm introduced is named TCP Small Queues (TSQ) and has the role of limiting the number of packets that a TCP socket can enqueue in the stack, waiting for the physical layer to send the packets before enqueueing extra data. A second significant TCP algorithm is named TCP Pacing (TP) and regulates the pace used by the socket to enqueue packets in the stack, regulating the formation of bursts of data. These mechanisms affect the frame aggregation logic on WLAN networks and compromise the throughput-latency tradeoff of all the TCP variants. This paper presents an experimental evaluation of these techniques investigating the wireless network performance of several TCP congestion control variants under the presence of different TSQ and TP policies, modeling also their interaction.

As WiFi grows ever more popular, airtime contention becomes an increasing problem. One way to alleviate this is through network policy enforcement. Unfortunately, WiFi lacks protocol support for configuring policies for its usage, and since network-wide coordination cannot generally be ensured, enforcing policy is challenging. However, as we have shown in previous work, an access point can influence the behaviour of connected devices by changing its scheduling of transmission opportunities, which can be used to achieve airtime fairness. In this work, we show that this mechanism can be extended to successfully enforce airtime usage policies in WiFi networks. We implement this as an extension our previous airtime fairness work, and present PoliFi, the resulting policy enforcement system. Our evaluation shows that PoliFi makes it possible to express a range of useful policies. These include prioritisation of specific devices; balancing groups of devices for sharing between different logical networks or network slices; and limiting groups of devices to implement guest networks or other low-priority services. We also show how these can be used to improve the performance of a real-world DASH video streaming application.

Programmable packet processing is increasingly implemented using kernel bypass techniques, where a userspace application takes complete control of the networking hardware to avoid expensive context switches between kernel and userspace. However, as the operating system is bypassed, so are its application isolation and security mechanisms; and well-tested configuration, deployment and management tools cease to function.

The last several years has seen a renewed interest in smart queue management to curb excessive network queueing delay, as people have realised the prevalence of bufferbloat in real networks. However, for an effective deployment at today's last mile connections, an improved queueing algorithm is not enough in itself, as often the bottleneck queue is situated in legacy systems that cannot be upgraded. In addition, features such as per-user fairness and the ability to de-prioritise background traffic are often desirable in a home gateway.In this paper we present Common Applications Kept Enhanced (CAKE), a comprehensive network queue management system designed specifically for home Internet gateways. CAKE packs several compelling features into an integrated solution, thus easing deployment. These features include: bandwidth shaping with overhead compensation for various link layers; reasonable DiffServ handling; improved flow hashing with both per-flow and per-host queueing fairness; and filtering of TCP ACKs. Our evaluation shows that these features offer compelling advantages, and that CAKE has the potential to significantly improve performance of last-mile internet connections.

In recent years, the Linux kernel has adopted an algorithm called TCP Small Queues (TSQ) for reducing queueing latency by controlling buffering in the networking stack. This solution consists of a back-pressure mechanism that limits the number of TCP segments within the sender TCP/IP stack, waiting for packets to actually be transmitted onto the wire before enqueueing further segments. Unfortunately, TSQ prevents the frame aggregation mechanism in the IEEE 802.11n/ac standards from achieving its maximum aggregation, because not enough packets are available in the queue to build aggregates from, which severely limits achievable throughput over wireless links. This paper demonstrates this limitation of TSQ in wireless networks and proposes Controlled TSQ (CoTSQ), a solution that improves TSQ so that it controls the amount of data buffered while allowing the IEEE 802.11n/ac aggregation logic to fully exploit the available channel and achieve high throughput. Results on a real testbed show that CoTSQ leads to a doubling of throughput on 802.11n and up to an order of magnitude improvement in 802.11ac networks, with a negligible latency increase.

We analyse two complementary datasets to quantify the latency variation experienced by internet end-users: (i) a large-scale active measurement dataset (from the Measurement Lab Network Diagnostic Tool) which shed light on long-term trends and regional differences; and (ii) passive measurement data from an access aggregation link which is used to analyse the edge links closest to the user.

Several new active queue management (AQM) and hybrid AQM/fairness queueing algorithms have been proposed recently. They seek to ensure low queueing delay and high network goodput without requiring parameter tuning of the algorithms themselves. However, extensive experimental evaluations of these algorithms are still lacking. This paper evaluates a selection of bottleneck queue management schemes in a test-bed representative of residential Internet connections of both symmetrical and asymmetrical bandwidths as well as WiFi. Latency under load and the performance of VoIP and web traffic patterns are evaluated under steady state conditions. Furthermore, the impact of the algorithms on fairness between TCP flows with different RTTs, and also the transient behaviour of the algorithms at flow startup is examined. The results show that while the AQM algorithms can significantly improve steady state performance, they exacerbate TCP flow unfairness. In addition, the evaluated AQMs severely struggle to quickly control queueing latency at flow startup, which can lead to large latency spikes that hurt the perceived performance. The fairness queueing algorithms almost completely alleviate the algorithm performance problems, providing the best balance of low latency and high throughput in the tested scenarios. However, on WiFi the performance of all the tested algorithms is hampered by large amounts of queueing in lower layers of the network stack inducing significant latency outside of the algorithms’ control.