Fast Packet Processing with eBPF and XDP Vieira, Marcos A. M.; Castanho, Matheus S.; Pacífico, Racyus D. G. ...
ACM computing surveys,
02/2020, Letnik:
53, Številka:
1
Journal Article
Recenzirano
Extended Berkeley Packet Filter (eBPF) is an instruction set and an execution environment inside the Linux kernel. It enables modification, interaction, and kernel programmability at runtime. eBPF ...can be used to program the eXpress Data Path (XDP), a kernel network layer that processes packets closer to the NIC for fast packet processing. Developers can write programs in C or P4 languages and then compile to eBPF instructions, which can be processed by the kernel or by programmable devices (e.g., SmartNICs). Since its introduction in 2014, eBPF has been rapidly adopted by major companies such as Facebook, Cloudflare, and Netronome. Use cases include network monitoring, network traffic manipulation, load balancing, and system profiling. This work aims to present eBPF to an inexpert audience, covering the main theoretical and fundamental aspects of eBPF and XDP, as well as introducing the reader to simple examples to give insight into the general operation and use of both technologies.
Underwater sensor networks are used to collect data from aquatic environments. Nodes from these networks usually communicate via acoustic or optical transmissions due to the poor performance of ...radiofrequency communication in these environments. Acoustic transmissions achieve longer distances than optical ones, but also consume more energy and have lower bandwidth. Underwater optical-acoustic sensor networks (UOASNs) combine both types of communication to explore the best of each one. In this paper, we propose CAPTAIN, an algorithm to perform data collection with data aggregation in UOASNs. CAPTAIN divides the network into clusters, builds a routing tree, and uses data aggregation to deliver all data collected to the sink node. Experiments showed that, compared to the shortest path algorithm, CAPTAIN led to lower network energy consumption, especially in denser networks, where it was able to consume, on average, up to 73% less. CAPTAIN could also achieve lower average latencies (up to almost 83% lower) and higher rates of data collected per hour by the sink node using fewer acoustic transmissions in clustered networks.
Energy is often a critical resource for underwater sensor networks. Underwater nodes usually have their life restricted to the initial charge of their battery, since recharging them may be ...impractical. This fact directly impacts sensing and monitoring applications, where a set of nodes is expected to cover and sense some region as long as possible. So, energy saving methods are very important for these networks. In this sense, this paper presents 3DVS: a method to perform node scheduling in underwater sensor networks that aims to extend network lifetime, while maintaining connectivity. 3DVS chooses only a subset of the network nodes to remain active so that the others can save their energy for later use. To do so, it uses 3D Voronoi Diagrams to decompose the space into regions around each of the nodes and so be able to determine those that can sleep for some time. Through simulations, we show that the proposed method is able to increase the network lifetime, guarantee connectivity and sensed data delivery and reduce network traffic.
This paper presents the novel routing protocol called Underwater Greedy Geographic Routing by Network Embedding (UW-GRE) for underwater wireless networks. Geographic routing is a promising network ...routing protocol strategy for wireless networks because it is stateless. However, this routing strategy faces various problems in the underwater environment. This is caused by some challenging characteristics of this environment, such as its 3D nature and the no propagation of GPS signal due to high radio frequency signal attenuation. Accurate localization, void regions, and guaranteed delivery are then some of the problems faced by this routing strategy in this environment. In this paper, we solve those problems efficiently by using geographic routing over virtual coordinates obtained through a network embedding in an n-dimensional (n≥ 2) Euclidean virtual space. By properly embedding the network in this virtual Euclidean space, no geographic information is needed. We implemented the protocol on ns-3. Our results show that, when compared to a conventional geographic routing protocol with perfect localization information, our protocol improves the number of transmissions and energy consumption by up to or higher than 30%, end-to-end delay by up to 28%, while achieving similar network throughput.
Many Wireless Sensor Networks (WSNs) applications demand to collect large volumes of data in a coordinated approach. To do this, a many-to-one pattern (convergecast) communication in tree-based WSN ...is used, but traffic near the sink node commonly becomes the network bottleneck. Thus, we propose an extension to a wireless network standard to provide wider bandwidth channels. We explore how fast the information can be collected in a WSN organized as a tree, with the radios operating with these wider bandwidth channels. As consequence, we proposed and implemented Funneling Wider Bandwidth (FWB), an algorithm that minimizes schedule length in networks with radios operating with the proposed extension. We proved that the algorithm is optimal in calculating a minimum number of time slots. In this way, we obtain a higher average throughput and a lesser number of time slots. Results from simulations and experiments on a real testbed confirm these gains. The proposed approach could be adapted and used with other related standards such as WirelessHARTTM, ISA 100.11a and IEEE 802.15.4e TSCH. Experiment results showed that with this wider bandwidth emulated in a given topology, the packets received per second by the sink node increased by 122.8% in comparison with just one bandwidth.
Scheduling Nodes in Underwater Networks using Voronoi Diagram Câmara Júnior, Eduardo P.M.; Vieira, Luiz F.M.; Vieira, Marcos A.M.
Proceedings of the 20th ACM International Conference on Modelling, Analysis and Simulation of Wireless and Mobile Systems,
11/2017
Conference Proceeding
Underwater networks are used for monitoring water resources and underwater environments. Thus, it is important that the underwater sensor nodes cover the largest region possible, during the largest ...amount of time. This paper presents a method to perform node scheduling in underwater stratified networks. It aims to maintain the network active for longer, maintaining its connectivity. Voronoi Diagrams are used to decompose the space into regions around each node in order to determine which one should be scheduled to sleep. Simulation results show that the proposed method achieves the desired objectives, more than doubling the network lifetime while guaranteeing connectivity.
There is evidence showing an inverse association between steps/day and arterial stiffness in adults. However, the relationship of steps/day and peak cadence with arterial stiffness is poorly ...understood in older adults. This study aimed to investigate the association between steps/day and peak cadence with arterial stiffness in older adults.
This cross-sectional study included 222 community-dwelling older adults (66 ± 5 years; 81.5% females; 70.3% with hypertension). Arterial stiffness was measured by aortic pulse wave velocity (aPWV). Steps/day and peak cadence were assessed by accelerometry. The participants were categorized according to the number of steps/day: sedentary <5000; low active 5000–7499; active 7500–9999; highly active 10,000+. Peak cadence was defined as the average of steps/day of the highest 30 min (not necessarily consecutive) for all valid days. Generalized linear models were used for data analyses.
The active (β = −0.34 m/s, 95% CI –0.60, −0.08) and highly active (β = −0.51 m/s, 95% CI –0.83, −0.20) groups had lower aPWV compared to the sedentary group. No significant difference was found between the low active group and the sedentary group (β = −0.21 m/s, 95% CI –0.46, 0.05). Every increment of 1000 steps/day was associated with a decrease of 0.05 m/s in the aPWV (95% CI –0.08, −0.02). Every increment of 10 steps/min in peak 30-min cadence was associated with a decrease of 0.05 m/s in aPWV (95% CI –0.09, −0.01).
Our findings show that easy-to-use proxies of the volume (steps/day) and intensity (peak cadence) of ambulatory behavior are inversely associated with arterial stiffness in older adults. The inverse association of steps/day and peak cadence with arterial stiffness is dose-response.
•Steps/day (volume) and peak cadence (intensity) are proxies of ambulatory behavior.•Both proxies were inversely associated with arterial stiffness in older adults.•Every 1000 steps/day was associated with 0.05 m/s lower pulse wave velocity (PWV).•Every 10 steps/min in peak 30-min cadence was associated with 0.05 m/s lower PWV.
Pectis brevipedunculata is native species and widely available in dry and semi-arid ecosystems showing high biotechnological potential. The objective of this study was to evaluate the circadian and ...seasonal chemical variation of the essential oil (EO) of P. brevipedunculata, as well as its acaricide effect on Rhipicephalus microplus larvae. Aerial parts were collected and submitted to the hydrodistillation process, and the chemical composition was determined by gas chromatography mass spectrometry (GC-MS). For the assays with R. microplus, the larval immersion test was performed. The main constituents were citral (75% of the EO), followed by α-pinene and limonene. In the seasonality analysis, the highest yields were in the months of April (2.08%) and August (2.05%), while in the study of circadian rhythm, the percentage was 2.0% at 6 p.m. in the rainy season, and 1.2%, dry season at 6 p.m. Concerning acaricidal activity (50% lethal concentration (LC50)), the April (1.17 mg mL-1), March (1.28 mg mL-1), June (1.37 mg mL-1), and October (1.27 mg mL-1) oils obtained were the most active and assays performed with circadian rhythm revealed in the rain season (April) at 6 p.m. and dry season (September) at 12 a.m. LC50 values of 1.75 and 1.75 mg mL-1, respectively. Additionally, this EO is selective to non-target organisms, i.e., ladybeetles and lacewing.
Underwater wireless acoustic networks (UWANs) take advantage of acoustic communication to enable many distinct applications. The peculiar features of underwater acoustic channels, such as long ...propagation delay, high bit error rates, and severely limited bandwidth, make the use of medium access control (MAC) solutions designed for terrestrial radio networks inefficient in UWANs. This paper proposes UW-SEEDEX, a MAC protocol for UWANs that employs random time slot schedules, created from seeds, to avoid collisions. After exchanging seeds, nodes can know other's entire schedules, allowing them to then better plan their transmissions. Simulations evaluate how each of UW-SEEDEX's parameter affects its performance in metrics such as end-to-end delay, energy consumption, transmissions per data reception, and reception rate using different test scenarios. Simulations also show that UW-SEEDEX can perform better than other MAC solutions, delivering more messages than protocols such as UW-Aloha and Slotted FAMA and reduced, on average, up to three times the number of transmissions required for each message reception in networks with grid topologies. UW-SEEDEX presented reception rates close to 100 percent, low energy cost, and fewer transmissions per data reception. Our code is available at https://gitlab.com/epmcj/ns-3-dev/-/tree/new-uan-mac-protocols .
Abstract only Introduction: Medical students, susceptible to sleep disorders, have irregular sleep-awake cycle, with repercussions on the quality of life and reduced academic performance, often with ...greater incidence of psychiatric disorders, estimated at 15% to 25% during your academic training. Objective: To evaluate the sleep habits in students of medical school in a private college of Rio de Janeiro, Brazil. Method: This subset study is a part of an observational study with cross-sectional delineation, with data collected through the application of an anonymous questionnaire, where they were asked about the number (and modification) of hours they sleep daily, in addition to reports of “stress and anxiety”. These data collected formed the basis of an instrument for assessing the quality of life on the medical students of this College. Results: We analyzed data from 481 students: 82 (17%) at the first year, 118 (24.5%) at the second year, 99 (20.6%) at the third, 64 (13.3%) and 118 (24.5%) within the fifth and sixth years. The average age was 21.7 years (16-42) and 306 (63.6%) of female gender. As for the hours of sleep, 445 students (92.5%) report 5 to 8 hours of sleep, and 216 (44.9%) a minimum of 6 hours. In the sleep of the weekends, 394 (81.9%) reported change in the number of hours and, 313 (65.1%) referred to a reduction after the entrance into college. It was reported "some level of stress and anxiety,“ not related to the least number of hours sleepping (0.07). Comparing the data obtained among the 199 students of the first and second years (62.8% female-group A) to 117 at fifth and sixth years (70.9% female-group B), we observed similarity in relation to (1) number of 5 to 8 hours sleepping (A 92.9% x 90.5% B), (2) change the hours of sleepping on the weekends (A 81.9% x 79.5% B), and (3) reduction of sleepping hours after the entrance into College (A 81.9% x 78.6% B). With regard to the report of "stress and anxiety”, we observed statistically significant difference between the groups (A 100% x 88% B, p < 0.03). Conclusion: The change in lifestyle imposed on joining a new school learning model, generates anxiety and loss of sleep hours among students. The development of resilience and adaptation to change, both individual and institutional, may have been responsible to varying degrees of stress.