Oddlab: fault-tolerant aware load-balancing framework for data center networks

Data center networks (DCNs) act as critical infrastructures for emerging technologies. In general, a DCN involves a multi-rooted tree with various shortest paths of equal length from end to end. The DCN fabric must be maintained and monitored to guarantee high availability and better QoS. Traditional traffic engineering (TE) methods frequently reroute large flows based on the shortest and least-congested paths to maintain high service availability. This procedure results in a weak link utilization with frequent packet reordering. Moreover, DCN link failures are typical problems. State-of-the-art approaches address such challenges by modifying the network components (switches or hosts) to discover and avoid broken connections. This study proposes Oddlab (Odds labels), a novel deployable TE method to guarantee the QoS of multi-rooted data center (DC) traffic in symmetric and asymmetric modes. Oddlab creatively builds a heuristic model for efficient flow scheduling and faulty link detection by exclusively using the gathered statistics from the DCN data plane, such as residual bandwidth and the number of installed elephant flows. Besides, the proposed method is implemented in an SDN-based DCN without altering the network components. Our findings indicate that Oddlab can minimize the flow completion time, maximize bisection bandwidth, improve network utilization, and recognize faulty links with sufficient accuracy to improve DC productivity.

Multi-Link Operation with Enhanced Synchronous Channel Access in IEEE 802.11be Wireless LANs: Coexistence Issue and Solutions

Multi-link operation is a new feature of IEEE 802.11be Extremely High Throughput (EHT) that enables the utilization of multiple links using individual frequency channels to transmit and receive between EHT devices. This paper aims to illustrate enhanced multi-link channel access schemes, identify the associated coexistence challenge, and propose solutions. First, we describe the multi-link operation of IEEE 802.11be and how the asynchronous and synchronous channel access schemes facilitate multi-link utilization. Next, we describe the design variants of the synchronous channel access scheme and demonstrate the associated coexistence challenge. Subsequently, we propose four features to address this challenge by assigning penalties to multi-link devices (repicking a backoff count, doubling the contention window size, switching to another contention window set, and compensating the backoff count) as well as five coexistence solutions derived from combinations of these features. Comparative simulation results are provided and analyzed for dense single-spot and indoor random deployment scenarios, demonstrating that the throughput and latency gains of multi-link operation differ between schemes. At the same time, we investigate the coexistence performance of multi-link operation with and without the capability of simultaneous transmission and reception and demonstrate that the proposed solutions mitigate the coexistence problem. In particular, compensating the backoff count achieves the highest coexistence performance among the proposed solutions, with a marginal throughput decrease of multi-link devices. A metric for evaluating both the throughput and latency gains and the coexistence performance of a multi-link channel access scheme using a single value is also proposed.

Fine-grained load balancing with traffic-aware rerouting in datacenter networks

Modern datacenters provide a wide variety of application services, which generate a mix of delay-sensitive short flows and throughput-oriented long flows, transmitting in the multi-path datacenter network. Though the existing load balancing designs successfully make full use of available parallel paths and attain high bisection network bandwidth, they reroute flows regardless of their dissimilar performance requirements. The short flows suffer from the problems of large queuing delay and packet reordering, while the long flows fail to obtain high throughput due to low link utilization and packet reordering. To address these inefficiency, we design a fine-grained load balancing scheme, namely TR (Traffic-aware Rerouting), which identifies flow types and executes flexible and traffic-aware rerouting to balance the performances of both short and long flows. Besides, to avoid packet reordering, TR leverages the reverse ACKs to estimate the switch-to-switch delay, thus excluding paths that potentially cause packet reordering. Moreover, TR is only deployed on the switch without any modification on end-hosts. The experimental results of large-scale NS2 simulations show that TR reduces the average and tail flow completion time for short flows by up to 60% and 80%, as well as provides up to 3.02x gain in throughput of long flows compared to the state-of-the-art load balancing schemes.

Hybrid Multicarrier Random Space Vector PWM for the Mitigation of Acoustic Noise

The pulse width modulation (PWM) inverter is an obvious choice for any industrial and power sector application. Particularly, industrial drives benefit from the higher DC-link utilization, acoustic noise, and vibration industrial standards. Many PWM techniques have been proposed to meet the drives’ demand for higher DC-link utilization and lower harmonics suppression and noise reductions. Still, random PWM (RPWM) is the best candidate for reducing the acoustic noises. Few RPWM (RPWM) methods have been developed and investigated for the AC drive’s PWM inverter. However, due to the lower randomness of the multiple frequency harmonics spectrum, reducing the drive noise is still challenging. These PWMs dealt with the spreading harmonics, thereby decreasing the harmonic effects on the system. However, these techniques are unsuccessful at maintaining the higher DC-link utilizations. Existing RPWM methods have less randomness and need complex digital circuitry. Therefore, this paper mainly deals with a combined RPWM principle in space vector PWM (SVPWM) to generate random PWM generation using an asymmetric frequency multicarrier called multicarrier random space vector PWM (MCRSVPWM). he SVPWM switching vectors with different frequency carrier are chosen with the aid of a random bi-nary bit generator. The proposed MCRSVPWM generates the pulses with a randomized triangular carrier (1 to 4 kHz), while the conventional RPWM method contains a random pulse position with a fixed frequency triangular carrier. The proposed PWM is capable of eradicating the high-frequency unpleasant acoustic noise more effectually than conventional RPWM with a shorter random frequency range. The simulation study is performed through MATLAB/Simulink for a 2 kW asynchronous induction motor drive. Experimental validation of the proposed MCRSVPWM is tested with a 2 kW six-switch (Power MOSFET–SCH2080KE) inverter power module-fed induction motor drive.

Energy-aware routing based on link utilization in domain network

Aimed at the characteristic of the Software Defined Network (SDN), several green routing algorithms are proposed. However, there are many drawbacks consisted in the existing algorithms. Therefore, we propose a self-adaptive energy saving routing algorithm (LAR) which is based on residual bandwidth of links and SDN. The proposed algorithm makes the link utilization which is changing in real time as the link cost. It would obtain the topology information and link status to optimize and prune the topology for reducing the computing time of routing algorithm before selecting routing path. After a period of time, the incoming flows will automatically be gathered in heavily-loaded links. The links without traffic will be switched off while the whole network connectivity and QoS are guaranteed. Simulation results show that it is possible to reduce considerable energy consumption during off-peak hours and link energy saving can be up to 55%. And, the algorithm has the distinct advantage in terms of complexity and network performance comparing with related schemes.

Fine-grained Load Balancing with Traffic-aware Rerouting in DataCenter Networks

Modern datacenters provide a wide variety of application services, which generate a mix of delay-sensitive short flows and throughput-oriented long flows, transmitting in the multi-path datacenter network. Though the existing load balancing designs successfully make full use of available parallel paths and attain high bisection network bandwidth, they reroute flows regardless of their dissimilar performance requirements. The short flows suffer from the problems of large queuing delay and packet reordering, while the long flows fail to obtain high throughput due to low link utilization and packet reordering. To address these inefficiency, we design a fine-grained load balancing scheme, namely TR (Traffic-aware Rerouting), which identifies flow types and executes flexible and traffic-aware rerouting to balance the performances of both short and long flows. Besides, to avoid packet reordering, TR leverages the reverse ACKs to estimate the switch-to-switch delay, thus excluding paths that potentially cause packet reordering. Moreover, TR is only deployed on the switch without any modification on end-hosts. The experimental results of large-scale NS2 simulations show that TR reduces the average and tail flow completion time for short flows by up to 60% and 80%, as well as provides up to 3.02x gain in throughput of long flows compared to the state-of-the-art load balancing schemes.

(0091) Control of Open end Induction Motor by Multi-objective GA based Selective Harmonic Elimination PWM to reduce Zero Sequence Currents and Torque Ripples

A Double inverter powered induction motor with open stator winding has few benefits, including excessive error forbearance functionality, great flexibility and lesser rating of dc input voltage etc. For this Configuration, two types of Modules can implement they are Non-isolated DC link and Isolated DC link. In these two, Non-isolated DC link is a good choice due to effective DC-link utilization and ruggedness, which is very beneficial in many applications. However, this module produces more zero sequence currents (Z-SC) by means of common mode (CMMD) voltage, which flows through Dc bus. The circulation of Z-SC must as little as feasible since it merely does rise the amplitude of currents in all phases. High ripple frequency of currents and torque, In addition resulting extra loss, which not alone reduces the efficiency, but loading ability and quickens the aging of drive. The triplen harmonics can denote meticulously as harmonics with integer of three times the frequency at fundamental, when they are in Phase in all Phases forms the Z-SC. In this paper, a novel SHE method is chosen to target triplen harmonics in Single DC Source Module (Non- isolated) and holding preferred fundamental quantity, which aids in improving the torque handling ability of the motor. In addition, the investigation of dual inverter fed OEW-IM with both common DC source as well as separate DC sources also explored by SHE for different number of switching angles and variable Modulation Index (MI) towards the torque ripples and Z-SC reduction are given. The foremost challenge related with SHE method is that resolving a set of higher order nonlinear equations with number of variables. A Multi-objective GA method provided for that challenge which effects the reduction in Z-SC so that torque ripples will be minimised. Moreover, the novel SHE method reduces more number of harmonics than the conventional SHE, which further decreases TH-D with decent fundamental quantity. For validation, the essential mathematical formulations and simulation results presented.

A port-based forwarding load-balancing scheduling approach for cloud datacenter networks

Today’s datacenter networks (DCNs) scale is rapidly increasing because of the wide deployment of cloud services and the rapid rise of edge computing. The bandwidth consumption and cost of a DCN are growing sharply with the extensions of network size. Thus, how to keep the traffic balanced is a key and challenging issue. However, the traditional load balancing algorithms such as Equal-Cost Multi-Path routing (ECMP) are not suitable for high dynamic traffic in cloud DCNs. In this paper, we propose a port-based forwarding load balancing scheduling (PFLBS) approach for Fat-tree based DCNs with some new features which can overcome the disadvantages of the existing load balancing methods in the following aspects. Firstly, we define a port-based source-routing addressing scheme, which decreases the switch complexity and makes the table-lookup operation unnecessary. Secondly, based on this addressing scheme, we proposed an effective routing mechanism which can obtain multiple available paths for flow scheduling based in Fat-tree. All the path information is saved in servers and each server only needs to maintain its own path information. Thirdly, we propose an efficient algorithm to implement large flows scheduling dynamically in terms of current link utilization ratio. This method is suitable for cloud DCNs and edge computing, which can reduce the complexity of the switches and the power consumption of the whole network. The experiment results indicate that the PFLBS approach has better performance compared with the ECMP, Hedera and MPTCP approaches, which decreases the flow completion time and improves the average throughput significantly. PFLBS is simple and can be implemented with a few signaling overheads.