Sieve: A flow scheduling framework in SDN based data center networks

Zaher, Maiass; Alawadi, Aymen Hasan; Molnár, Sándor

doi:10.1016/j.comcom.2021.02.013

Cited by 21 publications

(13 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We envision SDN-enabled edge networks spread out across geographical locations in the form of "regions", as shown in Figure 1. Those regional networks can be in many forms such as residential [23], enterprise [24], or cellular [25]. Regional administrators, through SDN controllers, allocate a finite amount of bandwidth reserve to a region, which will then be shared with all in-region consumers.…”

Section: Preliminaries a Problem Statementmentioning

confidence: 99%

“…The controllers then form FLOW-MOD messages to increase buyers' bandwidth according to how much they buy (line 5-9). Similarly, they also form Once the trading duration is over, the controllers form FLOW-MOD messages to revert the bandwidth state of buyers and sellers (line [18][19][20][21][22][23][24][25][26][27][28][29]. The controller reset the bandwidth reserve to the original values (line 25).…”

Section: Stage 3: Provisioning Stagementioning

confidence: 99%

See 1 more Smart Citation

Blockchain-Powered Bandwidth Trading on SDN-Enabled Edge Network

Oktian

et al. 2022

IEEE Access

View full text Add to dashboard Cite

Bandwidth trading procedures can be made to incentivize users to sell their needless traffic and indirectly reduce the probability of traffic congestion. However, implementation of bandwidth trading is opex-heavy from Internet Service Provider (ISP) perspective, while on the other hand, users also do not trust network executions from the ISP due to its heavily centralized control. These issues hinder the applicability of bandwidth trading and become our motivation to propose this paper. Our bandwidth-trading framework utilize software-defined networking (SDN) and blockchain. SDN automates the bandwidth trading executions from the ISP side and reduces the opex. Meanwhile, the smart contract is a trusted platform for building a trading marketplace where buyers, sellers, and SDN controllers can negotiate the trading terms. Once the trading is executed, SDN controllers generate proof of trading that must be submitted to the smart contract as proof of provisioning. We implement our works using Ethereum and POX SDN controllers, and the results prove that it can provide a seamless bandwidth trading experience with reasonable overhead. Furthermore, by committing to our framework, bandwidth trading can be executed fairly and securely because all previous provisioning can be cross-checked through the provided proof-oftrading.

show abstract

Section: Preliminaries a Problem Statementmentioning

confidence: 99%

Section: Stage 3: Provisioning Stagementioning

confidence: 99%

Blockchain-Powered Bandwidth Trading on SDN-Enabled Edge Network

Oktian

et al. 2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…These schemes commonly split a single flow over multiple paths at a fine granularity to avoid network congestion and reduce flow completion time (FCT). 2,3 However, multipath schemes can result in load unbalance and packet disorder owing to the different flow sizes and various delays over multiple paths. These disadvantages increase the FCT and result in network congestion.…”

Section: Introductionmentioning

confidence: 99%

“…Meanwhile, traffic engineering (TE) is widely employed to optimize inter‐DC network performance, and multipath schemes are commonly employed in TE. These schemes commonly split a single flow over multiple paths at a fine granularity to avoid network congestion and reduce flow completion time (FCT) 2,3 . However, multipath schemes can result in load unbalance and packet disorder owing to the different flow sizes and various delays over multiple paths.…”

Section: Introductionmentioning

confidence: 99%

Flow splitting scheme over link‐disjoint multiple paths in software‐defined networking

Liu

Zhou

Hai-feng

et al. 2022

Concurrency and Computation

View full text Add to dashboard Cite

Multipath networking schemes can be employed to efficiently balance traffic and address link failures. However, these schemes can also cause packet disorder, thereby deteriorating the overall network performance. To address these problems, we propose splitting a single flow over a multipath scheme (SFMS) at a fine granularity based on software-defined networking. The SFMS splits an elephant flow into several subflows according to the residual bandwidths and delays of the link-disjointed paths inter the data centers (inter-DC) network to enhance the transmission efficiency of traffic. The experimental results confirm that the SFMS not only reduces the flow completion time required from the time the first packet of an elephant flow is sent until the last packet is received while avoiding packet disorder.

show abstract

“…This leads to poor scheduling results for other flows. In addition, many works of mixed flows scheduling only focus on coarse-grained flows, and seldom consider fine-grained flow [12], [15]- [18]. There is a lack of research on finegrained flow scheduling algorithms, which does not match the extensive research on fine-grained flow classification.…”

Section: Introductionmentioning

confidence: 99%

Max-Min Fairness based Scheduling Optimization Mechanism on Switches

Wang

Jia

et al. 2022

2022 IEEE International Performance, Computing, and Communications Conference (IPCCC)

View full text Add to dashboard Cite

Multiple types of flows with contradictory service requirements, namely mixed flows, coexist in the data center network. Similar flows will be aggregated into the same queue after flow classification and are scheduled in switches by using fair queueing and its extension schemes which are capable of flow isolation. These schemes allocate different bandwidths by adjusting weights to realize differentiated services. However, existing solutions only focus on the requirements of some flows, which leads to the failure to satisfy the requirements of other flows. Therefore, it is necessary to make a trade-off between the service requirements of different flows when allocating bandwidth. In this paper, a max-min fairness based scheduling optimization (MMFSO) mechanism is proposed to schedule mixed flows. First, the bandwidth requirements of each queue are calculated by statistics of the switch. To reduce the influence of sampled statistics while forecasting bandwidth requirements, we introduce the exponentially weighted moving average for bandwidth requirements computation. Second, the bandwidth is allocated to each queue according to the max-mix fairness. The queue weight is determined by the allocated bandwidth of the queue. Finally, the performance of the proposed mechanism is evaluated on the hardware testbed in which workloads include coarse-grained flows and fine-grained flows. The results show that MMFSO can effectively schedule mixed flows.

show abstract

Sieve: A flow scheduling framework in SDN based data center networks

Cited by 21 publications

References 27 publications

Blockchain-Powered Bandwidth Trading on SDN-Enabled Edge Network

Blockchain-Powered Bandwidth Trading on SDN-Enabled Edge Network

Flow splitting scheme over link‐disjoint multiple paths in software‐defined networking

Max-Min Fairness based Scheduling Optimization Mechanism on Switches

Contact Info

Product

Resources

About