Low Latency Software Rate Limiters for Cloud Networks

He, Keqiang; Qin, Weite; Zhang, Qiwei; Wu, Wenfei; Yang, Junjie; Pan, Tian; Hu, Chengchen; Zhang, Jiao; Stephens, Brent; Zhang, Ying

doi:10.1145/3106989.3107005

Cited by 15 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They evaluate latency only as a function of flow throughput and ignore the performance impacts of bandwidth slicing in multi-queue scenarios. He et al [14] evaluate a software switch bandwidth slicing mechanism; however, their tests were performed in simple scenarios and their results are presented without a thorough analysis of the latency values. The same shortcoming is exhibited in [10]- [13] in terms of latency evaluation: their models of packet latency are for simple, synthetic testing scenarios.…”

Section: Related Workmentioning

confidence: 99%

Predictable Bandwidth Slicing with Open vSwitch

Chen¹,

Dezfouli

2021

Preprint

View full text Add to dashboard Cite

Software switching, a.k.a virtual switching, plays a vital role in network virtualization and network function virtualization, enhances configurability, and reduces deployment and operational costs. Software switching also facilitates the development of edge and fog computing networks by allowing the use of commodity hardware for both data processing and packet switching. Despite these benefits, characterizing and ensuring deterministic performance with software switches is harder, compared to physical switching appliances. In particular, achieving deterministic performance is essential to adopt software switching in mission-critical applications, especially those deployed in edge and fog computing architectures. In this paper, we study the impact of switch configurations on bandwidth slicing and predictable packet latency. We demonstrate that latency and predictability are dependent on the implementation of the bandwidth slicing mechanism and that the packet schedulers used in OVS Kernel-Path and OVS-DPDK each focus on different aspects of switching performance.

show abstract

Section: Related Workmentioning

confidence: 99%

Predictable Bandwidth Slicing with Open vSwitch

Chen¹,

Dezfouli

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In particular, zD allows for setting a small queue size, handling a large number of ⇥ows without causing any packet drops. End host queue buildup can be handed in a similar manner to in-network queue buildup through congestion control algorithms [16]. This approach does not eliminate packet drops but helps improve tail latency.…”

Section: Related Workmentioning

confidence: 99%

zD

Zhao

Ahmed

Zegura

et al. 2019

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

View full text Add to dashboard Cite

Modern end-host network stacks have to handle tra c from tens of thousands of ⇥ows and hundreds of virtual machines per single host, to keep up with the scale of modern clouds. This can cause congestion for tra c egressing from the end host. The e⇤ects of this congestion have received little attention. Currently, an over⇥owing queue, like a kernel queuing discipline, will drop incoming packets. Packet drops lead to worse network and CPU performance by in⇥ating the time to transmit the packet as well as spending extra e⇤ort on retansmissions. In this paper, we show that current end-host mechanisms can lead to high CPU utilization, high tail latency, and low throughput in cases of congestion of egress tra c within the end host. We present zD, a framework for applying backpressure from a congested queue to tra c sources at end hosts that can scale to thousands of ⇥ows. We implement zD to apply backpressure in two settings: i) between TCP sources and kernel queuing discipline, and ii) between VMs as tra c sources and kernel queuing discipline in the hypervisor. zD improves throughput by up to 60%, and improves tail RTT by at least 10x at high loads, compared to standard kernel implementation. CCS CONCEPTS • Networks ⌅ Network architectures; • Software and its engineering ⌅ Communications management.

show abstract

“…[130], [131], [132] 2 5 Rate limiting can improve performance isolation across tenants and users by preventing any tenant/flow from taking too much bandwidth starving other tenants/flows (a tenant may initiate many flows).…”

Section: B Traffic Shapingmentioning

confidence: 99%

Datacenter Traffic Control: Understanding Techniques and Trade-offs

Noormohammadpour¹,

Raghavendra²

2017

Preprint

View full text Add to dashboard Cite

Datacenters provide cost-effective and flexible access to scalable compute and storage resources necessary for today's cloud computing needs. A typical datacenter is made up of thousands of servers connected with a large network and usually managed by one operator. To provide quality access to the variety of applications and services hosted on datacenters and maximize performance, it deems necessary to use datacenter networks effectively and efficiently. Datacenter traffic is often a mix of several classes with different priorities and requirements. This includes user-generated interactive traffic, traffic with deadlines, and long-running traffic. To this end, custom transport protocols and traffic management techniques have been developed to improve datacenter network performance.In this tutorial paper, we review the general architecture of datacenter networks, various topologies proposed for them, their traffic properties, general traffic control challenges in datacenters and general traffic control objectives. The purpose of this paper is to bring out the important characteristics of traffic control in datacenters and not to survey all existing solutions (as it is virtually impossible due to massive body of existing research). We hope to provide readers with a wide range of options and factors while considering a variety of traffic control mechanisms. We discuss various characteristics of datacenter traffic control including management schemes, transmission control, traffic shaping, prioritization, load balancing, multipathing, and traffic scheduling. Next, we point to several open challenges as well as new and interesting networking paradigms. At the end of this paper, we briefly review inter-datacenter networks that connect geographically dispersed datacenters which have been receiving increasing attention recently and pose interesting and novel research problems.

show abstract

Low Latency Software Rate Limiters for Cloud Networks

Cited by 15 publications

References 17 publications

Predictable Bandwidth Slicing with Open vSwitch

Predictable Bandwidth Slicing with Open vSwitch

zD

Datacenter Traffic Control: Understanding Techniques and Trade-offs

Contact Info

Product

Resources

About