High-Resolution Timer-Based Packet Pacing Mechanism on the Linux Operating System

Takano, Ryousei; Kudoh, Tomohiro; Kodama, Y.; Okazaki, Fumihiro

doi:10.1587/transcom.e94.b.2199

Cited by 7 publications

(7 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Implementing Hrtimer We adopt the mechanism used in [72] and implement a Qdisc that transmits packets by raising qdisc watchdog timer interrupts. The data structures used in the Qdisc are illustrated in Fig.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

TCP Rapid: From theory to practice

Yin

Kaur

Smith

2017

IEEE INFOCOM 2017 - IEEE Conference on Computer Communications

View full text Add to dashboard Cite

Delay and bandwidth-based alternatives to TCP congestion-control have been around for nearly three decades and have seen a recent surge in interest. However, such designs have faced significant resistance in being deployed on a wide-scale across the Internet-this has been mostly due to serious concerns about noise in delay measurements, pacing inter-packet gaps, and required changes to the standard TCP stack. With the advent of high-speed networking, some of these concerns become even more significant. This thesis considers Rapid, a recent proposal for ultra-high speed congestion control, which perhaps stretches each of these challenges to the greatest extent. Rapid adopts a framework of continuous finescale bandwidth probing and rate adapting. It requires finely-controlled inter-packet gaps, high-precision timestamping of received packets, and reliance on fine-scale changes in inter-packet gaps. While simulationbased evaluations of Rapid show that it has outstanding performance gains along several important dimensions, these will not translate to the real-world unless the above challenges are addressed. This thesis identifies the key challenges TCP Rapid faces on real high-speed networks, including deployability in standard protocol stacks, precise inter-packet gap creation, achieving robust bandwidth estimation in the presence of noise, and a stability/adaptability trade-off. A Linux implementation of Rapid is designed and developed after carefully considering each of these challenges. The evaluations on a 10Gbps testbed confirm that the implementation can indeed achieve the claimed performance gains, and that it would not have been possible unless each of the above challenges was addressed.

show abstract

Section: Methodsmentioning

confidence: 99%

“…TRC-TCP [68] and [71] realize inter-packet gaps in the transport layer. The former [68] Pspacer/HT [72] and the Fair Queue (FQ) Qdisc [70] follow another strategy-they create gaps after the protocol stack processing. They implement a Qdisc, which shares a data structure with the protocol stack.…”

Section: Hrtimer-based Interruptsmentioning

confidence: 99%

TCP Rapid: From theory to practice

Yin

Kaur

Smith

2017

IEEE INFOCOM 2017 - IEEE Conference on Computer Communications

View full text Add to dashboard Cite

show abstract

“…In the experiments using ImageNet, we used nodes of the same quality on both the client side and the Cloud side; however, we used only the CPU on the client side and the GPGPU on the Cloud side. We used PSPacer [7] to control the network bandwidth to represent various sensor and Cloud network environments.…”

Section: Machine Performancementioning

confidence: 99%

Performance Evaluation of Pipeline-Based Processing for the Caffe Deep Learning Framework

Ichinose

Takefusa

Nakada

et al. 2018

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

Many life-log analysis applications, which transfer data from cameras and sensors to a Cloud and analyze them in the Cloud, have been developed as the use of various sensors and Cloud computing technologies has spread. However, difficulties arise because of the limited network bandwidth between such sensors and the Cloud. In addition, sending raw sensor data to a Cloud may introduce privacy issues. Therefore, we propose a pipelined method for distributed deep learning processing between sensors and the Cloud to reduce the amount of data sent to the Cloud and protect the privacy of users. In this study, we measured the processing times and evaluated the performance of our method using two different datasets. In addition, we performed experiments using three types of machines with different performance characteristics on the client side and compared the processing times. The experimental results show that the accuracy of deep learning with coarse-grained data is comparable to that achieved with the default parameter settings, and the proposed distributed processing method has performance advantages in cases of insufficient network bandwidth between realistic sensors and a Cloud environment. In addition, it is confirmed that the process that most affects the overall processing time varies depending on the machine performance on the client side, and the most efficient distribution method similarly differs.

show abstract

“…In exp2), as shown in Table 2, the number of Spout threads is set to two, the number of Bolt threads is set to 16, the inter-arrival time of image data is set to 10 ms/tuple, and the network bandwidth between sensors and cloud varies 10, 50, 100, and 1000 Mbps. We use PSPacer [9] for network bandwidth control. The constructed Storm cluster is shown in Figure 4.…”

Section: Overview Of Experimentsmentioning

confidence: 99%

A Study of Load Balancing between Sensors and the Cloud for a Real-Time Video Streaming Analysis Application Framework

Yuko

Takefusa

Nakada

et al. 2016

Proceedings of the 10th International Conference on Ubiquitous Information Management and Communication

View full text Add to dashboard Cite

For the home and office, many life-log analysis applications for transferring data from cameras and sensors to the cloud and analyzing the data have been developed. However, because of limitation of the resources of the cloud and the network bandwidth between the sensor and the cloud, it is difficult to execute large load processing, such as video streaming analysis, in real time on the cloud. Moreover, taking into account the execution environment from the sensor to the cloud, it is necessary to set an appropriate degree of parallelism in the processing from the pre-processing, such as feature extraction, to the analysis on the information. In this paper, we propose a video streaming analysis application framework for load balancing between sensors and the cloud, and investigate the performance of the application in a cluster environment that simulates the sensor and the cloud. From the experiments, we show that the processing performance is improved by increasing the number of processing threads, and we demonstrate the effectiveness of load balancing between the sensors and the cloud.

show abstract

High-Resolution Timer-Based Packet Pacing Mechanism on the Linux Operating System

Cited by 7 publications

References 7 publications

TCP Rapid: From theory to practice

TCP Rapid: From theory to practice

Performance Evaluation of Pipeline-Based Processing for the Caffe Deep Learning Framework

A Study of Load Balancing between Sensors and the Cloud for a Real-Time Video Streaming Analysis Application Framework

Contact Info

Product

Resources

About