Optimizing Inter-Datacenter Tail Flow Completion Times using Best Worst-case Routing

To meet the high performance requirement of real time and critical applications in industrial Internet of Things, modern lossless ethernet data center networks (DCNs) deployed with remote direct memory access and priority-based flow control (PFC) are dedicated to delivering low latency and high bandwidth. However, existing load balancing schemes either lack sub-round-trip-time congestion sensing or fail to accurately detect and reroute flows that cause congestion in PFC-enabled lossless DCNs. Therefore, we propose LBoDSN, an in-network load balancing for lossless DCNs using direct switch notification (DSN) for fast congestion flow detection, to address above challenges. LBoDSN tracks the evolution of ingress queue lengths at destination switches to anticipate the initiation of PFC pause, precisely identifies congested flows before PFC pause, and then sends DSNs to source switches to perform rerouting. After rerouting, the congestion notification packet associated with the previous path is selectively discarded to improve transmission performance. Experiments under realistic workloads reveal that, LBoDSN outperforms CONGA by 13%–65%, and 25%–80% in average and tail Flow Completion Times (FCTs), respectively. Compared to ConWeave, LBoDSN achieves approximately 9% improvement in both average and tail FCTs, while reducing switch queue consumption for reordering.

RaceCC: A rapidly converging explicit congestion control for datacenter networks

Accelerating Distributed Deep Learning draws increasing attention recently. Existing solutions like gradient compression, pipelining computation/communication, flow scheduling based on gradient disparities, neglect load balancing for Distributed Deep Learning (DDL) flows in DataCenter Network. In this paper, we review the PS traffic pattern in usual DDL training and propose that minimizing tail Flow Completion Time (FCT) of DDL flows is objective for DDL traffic Load Balancing. Besides, traffics of DDL differ from traditional DataCenter in temporal and spatial distribution. Current DCN Load Balancing become unadaptable to optimization requirements of DDL traffic, so we propose Tailor, a brand new load balancing scheme for DDL traffics training in DCN. Tailor is a solution without complex switch modification that sufficiently leverages the known parameter amount of DDL flows to discover congestion and reroutes by ECMP-like hash function to compensate bandwidth for slower flows. With large-scale simulation, we validate the effectiveness of Tailor. Tailor behaves well in both symmetric and asymmetric topology. It reduces time cost of DDL training by decreasing about 20-30% tail FCT than other load balancing mechanisms.

Software-Defined Data Center Networks (SDDCNs) utilizes Software Defined Networking (SDN) as a network architecture to achieve highly flexible, programmable, and automated management of Data Center Networks (DCNs). The high energy consumption of DCNs remains a persistent and significant challenge. Thus, the energy saving is crucial and imperative for DCNs. Current energy-efficient solutions primarily rely on flow consolidation and dynamic device sleeping techniques to reduce energy consumption. However, these approaches often yield long Flow Completion Time (FCT), potentially resulting in violations of service-level agreements, particularly for delay-sensitive applications. In this paper, we formulate the problem of minimizing the power consumption in SDDCNs as key objective while ensuring timely FCT for delay-sensitive applications. To solve this problem, we first introduce the Active Network Generation (ANG) approach, which generates a minimal active subnet with the least number of active devices while meeting the current traffic demand. Subsequently, we propose two algorithms based on the type of applications: the Delay-Tolerant Flow Route (DTFR) algorithm for delay-tolerant applications and the Delay-Sensitive Flow Route (DSFR) algorithm for delay-sensitive applications. Simulation results demonstrate that our propose solution achieves an energy-saving rate of up to 67.77% and significantly reduces FCT compared to benchmark solutions.

Cloud applications have reshaped the model of services and infrastructure of the Internet. Search engines, social networks, content delivery and retail and e-commerce sites belong to this group of applications. An important element in the architecture of data centers where these applications run is the communication infrastructure, commonly known as data center networks (DCNs). A critical challenge DCNs have to address is the processing of the traffic of cloud applications, which due to its properties is essentially different to the traffic of other Internet applications. In order to improve the responsiveness and throughput of applications, DCNs should be able to prioritize short flows (a few kB) over long flows (several MB). However, given the time and space variations the traffic presents, the information about flow sizes is not available in advance in order to plan the flow scheduling. There has been a wealth of solutions developed in this space, and prior work includes flow scheduling mechanisms optimizing for a specific workload but fall short when workloads are not known in advance, or comprise a collection of applications changing dynamically. In this paper, we present an adaptable mechanism called Adaptable Workload-Agnostic Flow Scheduling (AWAFS). It is an adaptable approach that can adjust in an agnostic way the scheduling configuration of DCN forwarding devices. This agnostic adjustment contributes to reduce the Flow Completion Time (FCT) of those short flows, representing around 85% of the traffic handled by cloud applications. AWAFS operates by observing the traffic and detecting statistical properties that provide a hint to adapt the scheduling parameters. Our evaluation results based on simulation show that AWAFS can reduce the average FCT of short flows between 16.9% and 45.2% when compared to the best existing agnostic non-adaptable solution, without inducing starvation on long flows. Indeed, it can provide improvements as high as 39% for long flows. Additionally, AWAFS can improve the FCT for short flows in scenarios with high heterogeneity in the traffic present in the network, with a reduction up to 5% for the average FCT and 15% for the tail FCT.

Reducing the power consumption and maintaining the Flow Completion Time (FCT) for the Quality of Service (QoS) of applications in Data Center Networks (DCNs) are two major concerns for data center operators. However, existing works either fail in guaranteeing the QoS due to the neglect of the FCT constraints or achieve a less satisfying power efficiency. In this paper, we propose SmartFCT, which employs Software-Defined Networking (SDN) coupled with the Deep Reinforcement Learning (DRL) to improve the power efficiency of DCNs and guarantee the FCT. The DRL agent can generate a dynamic policy to consolidate traffic flows into fewer active switches in the DCN for power efficiency, and the policy also leaves different margins in different active links and switches to avoid FCT violation of unexpected short bursts of flows. Simulation results show that with similar FCT guarantee, SmartFCT can save 8% more of the power consumption compared to the state-of-the-art solutions.

Data center network (DCN) can consume a significant amount of power (e.g., 10% to 20%) in large-scale data centers. To reduce the power consumption of DCN, traffic consolidation has been recently proposed as an effective approach to reduce the number of DCN devices in use. However, existing consolidation approaches do not sufficiently consider the flow completion time (FCT) requirement. On one hand, missing the FCT deadlines can cause serious violation of service-level agreement, especially for delay-sensitive networking services, such as web search and E-commerce. On the other hand, keeping all the devices on to make FCTs much shorter than the desired requirements is unnecessary because 1) users may not be able to perceive the difference, and 2) such a greedy strategy can lead to unnecessarily high DCN power consumption and thus more electricity costs. In this paper, we propose FCTcon, a dynamic FCT control strategy for DCN power optimization. FCTcon is designed rigorously based on control theory to dynamically control the FCT of delay-sensitive traffic flows exactly to requirements, such that the desired FCT performance is guaranteed while the maximum amount of DCN power savings can be achieved. Results from both hardware experiments and simulation evaluations demonstrate that, compared to the state-of-the-art DCN power optimization schemes, FCTcon can improve the DCN FCT performance, while achieving nearly the same or even more power savings. Consequently, FCTcon can result in more than 22.0% to 62.2% extra net profits for a data center with 50K servers.

In the era of cloud computing, data center network (DCN) can consume a significant amount of power (e.g., 10 to 20 percent) in large-scale data centers. To reduce the power consumption of DCN, traffic consolidation has been recently proposed as an effective approach to reduce the number of DCN devices in use. However, existing consolidation approaches do not sufficiently consider the flow completion time (FCT) requirement. On one hand, missing the FCT deadlines can cause serious violation of service-level agreement, especially for delay-sensitive networking services, such as web search and E-commerce. On the other hand, keeping all the devices on to make FCTs much shorter than the desired requirements is unnecessary because 1) users may not be able to perceive the difference, and 2) such a greedy strategy can lead to unnecessarily high DCN power consumption and thus more electricity costs. In this paper, we propose FCTcon, a dynamic FCT control strategy for DCN power optimization. FCTcon is designed rigorously based on control theory to dynamically control the FCT of delay-sensitive traffic flows exactly to the requirements, such that the desired FCT performance is guaranteed while the maximum amount of DCN power savings can be achieved. Results from both hardware experiments and simulation evaluation demonstrate that, compared to the state-of-the-art DCN power optimization schemes, FCTcon can improve the DCN FCT performance, while achieving nearly the same or even more power savings. Consequently, FCTcon can result in more than 22.0 to 62.2 percent extra net profits for a data center with 50K servers. In addition, we further propose two extended designs of FCTcon to handle the coflow abstraction recently proposed for DCN, which successfully reduce the coflow deadline miss ratio by 12 to 15 percent.

In this paper, we investigate the use of instantly decodable network coding (IDNC) for improving two fundamental performance metrics, namely, the completion time (as a measure of throughput) and mean decoding delay, in multicast cooperative data exchange (CDE) systems, where a group of geographically close clients cooperate with each other to obtain their missing packets. Here, an IDNC scheme is used for the transmissions across these clients. We utilize the stochastic shortest path (SSP) technique to study the minimum mean completion time problem. However, since finding the optimum solution is intractable, we use the obtained formulation to draw some theoretical guidelines to heuristically find solutions that can efficiently reduce the completion time. Second, we formulate the minimum mean decoding delay problem as selecting the appropriate maximal clique over well-structured graphs at the clients, and in order to reduce its complexity, we propose a simple heuristic algorithm for it. The effectiveness of our proposed algorithms is verified through extensive simulations and comparisons with existing techniques.

The proliferation of cloud computing has forced data centers to support a large number of dynamic resources such as virtual machines and storage volumes from any physical location within the data center. This has led to highly volatile and unpredictable traffic patterns. The emergence of cloud applications that exchange large volumes of information have created large persistent flows that need to coexist with other types of traffic. As a result, a number of data center topologies have been proposed to support cloud applications. One of the topologies that has gained prominence is the Clos network. On one hand, Clos networks provide many nice properties, such as equidistant hops, bandwidth, and latency. On the other hand the routing of diverse traffic flows still remains an issue. Current routing methods such as equal-cost multi-path (ECMP) routing that rely on flow hashing over equal cost paths do not take into consideration traffic volume nor network state. In this paper, we investigate routing in Clos topologies. We formulate optimal flow routing in Clos networks as a binary multicommodity flow problem and present optimized flow re-routing, a modified version of the binary multicommodity flow problem that uses a heuristics approach to minimize hash collisions. We also propose FlowFit, a practical optimization method that uses network state to optimally re-assign flows to links. Our experimental and simulation results show a significant improvement in bisection bandwidth and flow completion time over ECMP.

Aim To assess the impact of simulated laparoscopic trainers on improving surgical trainees’ technical ability in completing a laparoscopic appendicectomy by measuring effect on completion time and distance travelled with standard instruments. Method Six CT1 (early-stage) trainees with little prior operative exposure were selected from the North-West of England Deanery to perform x10 appendicectomy simulated exercises with the Augmented Reality Laparoscopic Simulator (LapAR™), interspersed by x3 Lapass exercises. Metrics including ‘Time of completion’ and ‘Distance travelled’ were collected by the simulators and analysed. Results Repeated laparoscopic appendicectomies with the LapAR™ improved performance time for 100% of trainees, with an average of 55% (Regression coefficient -0.65) and improved distance travelled for 75% of the trainees with an average improvement of 39% (Coefficient -2.67). Improvement in instrument smoothness, acceleration and ambidexterity were also observed. Conclusions LapAR™ use has improved all performance metrics for most surgical trainees and proves to be an invaluable tool for surgical training.

The use of methcathinone (MCAT), a psychostimulant drug that can lead to long-term health risks and executive dysfunction, increased to an alarming rate in recent years. Impairments in low-level executive function have been reported in substance use disorder. However, little empirical evidence is available regarding high-level executive function (e.g., problem solving), which may act as a risk factor for relapse. The present study aimed to investigate whether the problem-solving ability was altered in abstinent individuals with methcathinone use disorder (MCUD). Here, we tested fifty male MCUD individuals (short-term MCUD group: twenty-nine patients with MCAT use less than 3years, long-term MCUD group: twenty-one patients with MCAT use longer than 3years, which were split by medium years of drug use) and twenty-four well-matched healthy controls (HC) in the Tower of Hanoi task (TOH) to assess the impact of task difficulty on drug-related changes in problem-solving performance. We used several measures to characterize problem-solving performance: the number of mistakes made, the completion time of the task, and the thinking time before the first move. In the low task difficulty condition, the MCUD group and HC group showed similar levels of mistakes and completion time, while in the high task difficulty condition, the MCUD group reported more mistakes (the mean number of mistakes in each trial: 1.41 ± 1.15 vs 0.79 ± 0.76, P = 0.019, Cohen's d = 0.635) and longer completion time in the task (the mean completion time in each trial: 45.83 ± 20.51s vs 33.40 ± 15.10s, P = 0.010, Cohen's d = 0.690) than the HC group. The thinking time before the first move did not differ significantly between groups (P = 0.257). We further found that the long-term (more than 3years) MCUD group made more mistakes than the short-term MCUD group and HC group, mainly in the highly difficult subtasks. The longer time than HCs was reported in the long-term MCUD group among high task difficulty of subtasks. In addition, there was a positive correlation between years of MCAT use and the number of mistakes made in high task difficulty TOH task (r = 0.326, P = 0.021). Chronic methcathinone use was associated with deficits in problem-solving performance, which depended on the degree of task difficulty. The impairment was more evident in the long-term (> 3years) MCAT group.

Numerous scheduling approaches have been proposed to improve user experiences in a data center network (DCN) by reducing flow completion time (FCT). Mimicking the shortest job first (SJF) has been proved to be the prominent way to improve FCT. To do so, some approaches require flow size or completion time information in advance, which is not possible in scenarios like HTTP chunk transfer or database query response. Some information-agnostic schemes require involving end-hosts for counting the number of bytes sent. We present Longer Stay Less Priority (LSLP), an information-agnostic flow scheduling scheme, like Multi-Level Feedback Queue (MLFQ) scheduler in operating systems, that aims to mimic SJF using P4 switches in a DCN. LSLP considers all the flows as short flows initially and assigns them to the highest priority queue, and flows get demoted to the lower priority queues over time. LSLP estimates the active time of a flow by leveraging the state-of-the-art P4 switch’s programmable nature. LSLP estimates the active time of a group of new flows that arrive during a time interval and assigns their packets to the highest priority. At the beginning of the next time interval, arriving packets of old flows are placed one priority lower except for those already in the lowest priority queue. Therefore, short flows can be completed in the few higher priority queues while long flows are demoted to lower priority queues. We have evaluated LSLP via a series of tests and shown that its performance is comparable to the existing scheduling schemes.

One of the major issues in Data Center Networks (DCNs) is related to elephant flows. These flows typically are primarily responsible for network congestion, causing impacts on applications that are intolerant to delays. Although there are several works in the literature that analyze this type of traffic, none of them consider the impact of elephant flow size on the throughput and on the Flow Completion Time (FCT). This article presents the EFM (Elephant Flow Manager) as a complete solution for monitoring and re-routing elephant flows in DCNs, by observing the size of these flows in the network. The tests performed show that the size of the elephant flows influences the throughput and the FCT of the applications and, therefore, deserves to be observed. The evaluation of the solution was performed comparing the joint use of EFM with ECMP (EqualCost Multi-path). In all tests, our re-routing solution proved to be better when compared to pure routing solely with ECMP. Then, this paper presents three main contributions: (1) it compares our solution with the ECMP, (2) highlights the importance of the size of the elephant flows in the re-routing of these in DCNs and (3) preliminary tests indicate that due to the decrease in FCT, it is possible to reduce the energy consumption in the data center.

This research paper focuses on the development and analysis of a web-based system which is designed to study road traffic conditions and their routing behaviour in a dynamic environment. Nowadays Traffic congestion has become a major issue in urban areas due to the increasing number of vehicles and frequent changes in road conditions such as congestion, accidents, and peak-hour delays. Efficient route planning during those conditions requires routing approaches that can calculate traffic variations in real time. In this thesis, the road network is modelled as a weighted graph, where intersections are represented as nodes and road segments are represented as edges with weights corresponding to traffic conditions. The proposed web application allows users to interact with the road network by selecting source and destination points. They can also modify traffic conditions, and routing techniques and observe how routes change dynamically. Both static shortest path computation and dynamic routing approaches are implemented to study their behaviour and performance under varying traffic scenarios. The study provides visual representation of road networks and routing paths and also their performance metrics such as computation time and route cost. Thus it helps understand the clear comparison between static and dynamic routing methods and the impact of traffic changes depending on routing decisions. The developed web application serves as an analytical and educational tool for studying road traffic analysis and routing algorithms. The study shows the usefulness of web-based platforms in visualizing traffic behaviour and evaluating routing performance. But the current approach is limited to simulated data. This leaves room for future work that could improve the system by integrating real-world traffic data and explore more advanced routing strategies.