اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدBackpressure Flow Control
54
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Effective congestion control for data center networks is becoming increasingly challenging with a growing amount of latency sensitive traffic, much fatter links, and extremely bursty traffic. Widely deployed algorithms, such as DCTCP and DCQCN, are still far from optimal in many plausible scenarios, particularly for tail latency. Many operators compensate by running their networks at low average utilization, dramatically increasing costs. In this paper, we argue that we have reached the practical limits of end-to-end congestion control. Instead, we propose, implement, and evaluate a new congestion control architecture called Backpressure Flow Control (BFC). BFC provides per-hop per-flow flow control, but with bounded state, constant-time switch operations, and careful use of buffers. We demonstrate BFCs feasibility by implementing it on Tofino2, a state-of-the-art P4-based programmable hardware switch. In simulation, we show that BFC achieves near optimal throughput and tail latency behavior even under challenging conditions such as high network load and incast cross traffic. Compared to existing end-to-end schemes, BFC achieves 2.3 - 60 X lower tail latency for short flows and 1.6 - 5 X better average completion time for long flows.
قيم البحث
اقرأ أيضاً
A cross-layer cognitive radio system is designed to support unicast and multicast traffic with integration of dynamic spectrum access (DSA), backpressure algorithm, and network coding for multi-hop networking. The full protocol stack that operates wi
th distributed coordination and local information exchange is implemented with software-defined radios (SDRs) and assessed in a realistic test and evaluation (T&E) system based on a network emulation testbed. Without a common control channel, each SDR performs neighborhood discovery, spectrum sensing and channel estimation, and executes a distributed extension of backpressure algorithm that optimizes the spectrum utility (that represents link rates and traffic congestion) with joint DSA and routing. The backpressure algorithm is extended to support multicast traffic with network coding deployed over virtual queues (for multicast destinations). In addition to full rank decoding at destinations, rank deficient decoding is also considered to reduce the delay. Cognitive network functionalities are programmed with GNU Radio and Python modules are developed for different layers. USRP radios are used as RF front ends. A wireless network T&E system is presented to execute emulation tests, where radios communicate with each other through a wireless network emulator that controls physical channels according to path loss, fading, and topology effects. Emulation tests are presented for different topologies to evaluate the throughput, backlog and energy consumption. Results verify the SDR implementation and the joint effect of DSA, backpressure routing and network coding under realistic channel and radio hardware effects.
We consider a set of flows passing through a set of servers. The injection rate into each flow is governed by a flow control that increases the injection rate when all the servers on the flows path are empty and decreases the injection rate when some
server is congested. We show that if each servers congestion is governed by the arriving traffic at the server then the system can *oscillate*. This is in contrast to previous work on flow control where congestion was modeled as a function of the flow injection rates and the system was shown to converge to a steady state that maximizes an overall network utility.
Providing resilient network control is a critical concern for deploying Software-Defined Networking (SDN) into Wide-Area Networks (WANs). For performance reasons, a Software-Defined WAN is divided into multiple domains controlled by multiple controll
ers with a logically centralized view. Under controller failures, we need to remap the control of offline switches from failed controllers to other active controllers. Existing solutions could either overload active controllers to interrupt their normal operations or degrade network performance because of increasing the controller-switch communication overhead. In this paper, we propose RetroFlow to achieve low communication overhead without interrupting the normal processing of active controllers during controller failures. By intelligently configuring a set of selected offline switches working under the legacy routing mode, RetroFlow relieves the active controllers from controlling the selected offline switches while maintaining the flow programmability (e.g., the ability to change paths of flows) of SDN. RetroFlow also smartly transfers the control of offline switches with the SDN routing mode to active controllers to minimize the communication overhead from these offline switches to the active controllers. Simulation results show that compared with the baseline algorithm, RetroFlow can reduce the communication overhead up to 52.6% during a moderate controller failure by recovering 100% flows from offline switches and can reduce the communication overhead up to 61.2% during a serious controller failure by setting to recover 90% of flows from offline switches.
In a hybrid PON/xDSL access network, multiple Customer Premise Equipment (CPE) nodes connect over individual Digital Subscriber Lines (DSLs) to a drop-point device. The drop-point device, which is typically reverse powered from the customer, is co-lo
cated with an Optical Network Unit (ONU) of the Passive Optical Network (PON). We demonstrate that the drop-point experiences very high buffer occupancies when no flow control or standard Ethernet PAUSE frame flow control is employed. In order to reduce the buffer occupancies in the drop-point, we introduce two gated flow control protocols that extend the polling-based PON medium access control to the DSL segments between the CPEs and the ONUs. We analyze the timing of the gated flow control mechanisms to specify the latest possible time instant when CPEs can start the DSL upstream transmissions so that the ONU can forward the upstream transmissions at the full PON upstream transmission bit rate. Through extensive simulations for a wide range of bursty traffic models, we find that the gated flow control mechanisms, specifically, the ONU and CPE grant sizing policies, enable effective control of the maximum drop-point buffer occupancies.
Software Defined Networking (SDN) promises greater flexibility for directing packet flows, and Network Function Virtualization promises to enable dynamic management of software-based network functions. However, the current divide between an intellige
nt control plane and an overly simple, stateless data plane results in the inability to exploit the flexibility of a software based network. In this paper we propose SDNFV, a framework that expands the capabilities of network processing-and-forwarding elements to flexibly manage packet flows, while retaining both a high performance data plane and an easily managed control plane. SDNFV proposes a hierarchical control framework where decisions are made across the SDN controller, a host-level manager, and individual VMs to best exploit state available at each level. This increases the networks flexibility compared to existing SDNs where controllers often make decisions solely based on the first packet header of a flow. SDNFV intelligently places network services across hosts and connects them in sequential and parallel chains, giving both the SDN controller and individual network functions the ability to enhance and update flow rules to adapt to changing conditions. Our prototype demonstrates how to efficiently and flexibly reroute flows based on data plane state such as packet payloads and traffic characteristics.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
