State Aware Multiple Choice Routing for Load Balancing in Data

This paper presents the network load balancing problem, a challenging real-world task for multi-agent reinforcement learning (MARL) methods. Traditional heuristic solutions like Weighted-Cost Multi-Path (WCMP) and Local Shortest Queue (LSQ) are less flexible to the changing workload distributions and arrival rates, with a poor balance among multiple load balancers. The cooperative network load balancing task is formulated as a Dec-POMDP problem, which naturally induces the MARL methods. To bridge the reality gap for applying learning-based methods, all methods are directly trained and evaluated on an emulation system from moderate-to large-scale. Experiments on realistic testbeds show that the independent and "selfish" load balancing strategies are not necessarily the globally optimal ones, while the proposed MARL solution has a superior performance over different realistic settings. Additionally, the potential difficulties of MARL methods for network load balancing are analysed, which helps to draw the attention of the learning and network communities to such challenges.

The massive energy usage of data centers may be traced in part to the growing number of data centers and workstations because of economies of scale for cloud computing. It does, though, indicate wasteful power usage. Consequently, researching ways to improve the energy efficiency of datacenter equipment is now a crucial aspect in reducing datacenter power consumption. In this study, we describe methodologies and algorithms for flexible, energy-efficient, and effective load balancing in data centers, resulting in lower energy consumption by systems. Because of difficulties related with energy consumption, such as capital expenditure, operational costs, and environmental effect, renewable energy is becoming an increasing major consideration in data centers. With the rise in environmental consequences around the world, data centers must consider energy efficiency as one of the most critical factors. Cloud computing techniques have made a huge impact all over the world. Green DC is a concept that has been tested. We suggested an algorithm in this paper based on certain current algorithm limitations that will help to reduce environmental effect. We investigated the efficiency of our program further and used a load balancing method to improve its performance.

Virtualization is the key technology of cloud computing, by which the physical resources in data center can be abstracted as a virtual resource pool for flexible resource allocation. To make full use of the data center resources, efficient virtual network embedding (VNE) is both an efficient and challenging solution. Traditional VNE algorithms focused on the generic topology in the Internet, while the data center network topology is regular and symmetric. In this paper, the most widely used fat-tree topology is considered, and a proximal adaptive VNE algorithm based on multi-dimensional load balancing is proposed. In this VNE algorithm, the bandwidth resource cost is reduced by proximal mapping, and the joint balancing of computation and communication loads is taken into account to achieve multi-dimension resource allocation balance. Simulation results show that the proposed algorithm can increase the resource utilization, and keep both the load balance of each single-dimension resource and the joint balance of different dimensions of resources.

Emerging network applications (augmented reality, industrial Internet, etc.) introduce stringent new requirements on the performance, dependability, and adaptability of communication networks. Programmable data planes (e.g., based on P4) provide new opportunities to meet these requirements, by enabling adaptive network reconfigurations. However, ensuring consistency during such reconfigurations remains challenging. This paper makes a first step toward a more automated state management of adaptive data planes. In particular, we present an efficient P4 state management framework, P4State, which allows to quickly identify the network states from the source code that are critical for data plane reconfigurations (e.g., due to scaling, failure recovery). We report on first promising evaluation results of our prototype implementation in terms of correctness and efficiency, also considering two case studies using HULA (load balancing in data center) and HashPipe (line-rate measurement in data plane).

The routing based on load balancing and shortest path routing (LB-SPR) was recently introduced, and it was shown to have advantages over the standard shortest path routing (SPR) algorithms: it can support higher guaranteed traffic loads, and it simplifies the resource reservation processes. In this paper, the cost of setting up the reliable network which is resilient to node or link failures is analyzed. In such a network, enough capacity has to be provisioned on all the links for routing the traffic even in the case of a failure. The cost of setting up the reliable network using the shortest path routing (SPR) is compared to the cost of setting up the reliable network using the load balanced shortest path routing (LB-SPR). It is shown that LB-SPR can significantly decrease the network cost.

The stability of load balancing in the routers is a major problem for traffic flow survivability. Arrangements usually utilize an Equal Cost Multi Path (ECMP) mechanism, which basically emphasizes the load balancing network by equally splitting flows to the accessible concise paths. Empirical research conducted by previous researchers has provided new proof of techniques and methods solutions for more efficient load balancing schemes. However, a survey that examines the practical recommendation has rarely been considered. The surveys on which previous papers have focused give an experimental confirmation for load balancing in a datacenter with regards to descriptive investigation to order to suggest a scheme of best practices. The investigation of load balancing in every telecommunication layer technology is the premise for this descriptive survey. The outline arrangement of the advance level path parallelism in a datacenter and redundancy of transmission to accomplish low Flow Completion Times (FCTs) are the goals of the researcher. More research is expected to assess the effect of packet distribution against FCT. The terms usually used are load balancing, survivability, ECMP, network congestion, link criticality, and link failure protection. Load balancing or ECMP mechanism that are not focused on alternate congestion scheme of the network were exempted from the review.

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.

High-speed routers and high-performance data centers share a common system-level architecture in which multiple processing nodes are connected by an interconnection network for high-speed communications. Load balancing is an important technique for maximizing throughput and minimizing delay of the interconnection network. In this thesis, efficient load balancing schemes are designed and analyzed for next-generation routers and data centers.
\n
\nIn high-speed router design, two preferred switch architectures are input-queued switch and load-balanced switch. In an input-queued switch, time-domain load balancing can be carried out by an iterative algorithm that schedules packets for sending in different time slots. The complexity of an iterative algorithm increases rapidly with the number of scheduling iterations. To address this problem, a single-iteration scheduling algorithm called D-LQF is designed, in which exhaustive service policy is adopted for reusing the matched input-output pairs in the previous time slots to grow the match size. 
\n
\nUnlike an input-queued switch, a load-balanced switch consists of two stages of crossbar switch fabrics, where load balancing is carried out in both time and space domains. Among various load-balanced switches, the feedback-based switch gives the best delay-throughput performance. In this thesis, the feedback-based switch is enhanced in three aspects. Firstly, we focus on reducing its switch fabric complexity. Instead of using crossbars, a dual-banyan network is proposed. The complexity of dual-banyan can be further reduced by merging the two banyans to form a Clos network, resulting in a Clos-banyan network. Secondly, we target at improving the delay performance of the feedback-based switch. A Clos-feedback switch architecture is devised where each switch module in the Clos network is a small feedback-based switch. With application-flow based load balancing, packet order is ensured and the average packet delay is reduced from O(N) to O(n), where N and n are the switch and switch module sizes, respectively. Thirdly, we extend the feedback-based switch to support multicast traffic. Based on the notion of pointer-based multicast VOQ, an efficient multicast scheduling algorithm with packet replication at the middle-stage ports only is proposed. In order to provide close-to-100% throughput for any admissible multicast traffic patterns, a three-stage implementation of feedback-based switch is also designed.
\n
\nIn designing load balancing schemes for data centers, we focus on the most popular fat-tree based data centers. Notably, packet-based load balancing is widely considered infeasible for data centers. This is because the associated packet out-of-order problem will cause unnecessary TCP fast retransmits, and as a result, severely undermine TCP performance. In this thesis, we show that if packet-based load balancing is performed properly, the packet out-of-order problem can be easily addressed by slightly increasing the number of duplicate ACKs required for triggering fast retransmit. Admittedly, in case of a real packet loss, the loss recovery time will be increased. But our simulation results show that such an increase is far less than the reduction in the network queueing delay (due to a better load-balanced network). As compared to a flow-based load balancing scheme, our packet-based scheme consistently provides significantly higher goodput and noticeably smaller delay.

Many existing load balancing mechanisms work effectively in lossy datacenter networks (DCNs), but they suffer from serious packet reordering in lossless Ethernet DCNs deployed with the hop-by-hop Priority-based Flow Control (PFC). The key reason is that the prior solutions are not able to perceive PFC triggering correctly and in a timely manner when making load balancing decisions. Once the forwarding path pauses transmission due to PFC triggering, the packets allocated on it are blocked, inevitably leading to out-of-order packets and retransmission. In this paper, we present an Reordering-robust Load Balancing (RLB) scheme with PFC prediction in lossless DCNs. At its heart, RLB leverages the derivative of ingress queue length to predict PFC triggering and proactively notifies the upstream switches to choose an appropriate rerouting path or perform packet recirculation to avoid reordering. Furthermore, under switch failure scenarios, RLB adjusts the recirculation threshold adaptively to mitigate the risk of packets over-recirculation. We have implemented RLB in the hardware programmable switch. As a building block for existing load balancing mechanisms, we have integrated RLB into Presto, LetFlow, Hermes and DRILL. The evaluation results show that the RLB-enhanced solutions deliver significant performance by avoiding packet reordering. For example, it reduces the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$99^{th}$</tex-math> </inline-formula> percentile flow completion time (FCT) by up to 72%, 67%, 58% and 54% over DRILL, Presto, LetFlow and Hermes, respectively.

With the emerging diverse applications in data centers, the demands on quality of service in data centers also become diverse, such as high throughput of elephant flows and low latency of deadline-sensitive flows. However, traditional TCPs are ill-suited to such situations and always result in the inefficiency (e.g. missing the flow deadline, inevitable throughput collapse) of data transfers. This further degrades the user-perceived quality of service (QoS) in data centers. To reduce the flow completion time of mice and deadline-sensitive flows along with promoting the throughput of elephant flows, an efficient and deadline-aware priority-driven congestion control (PCC) protocol, which grants mice and deadline-sensitive flows the highest priority, is proposed in this paper. Specifically, PCC computes the priority of different flows according to the size of transmitted data, the remaining data volume, and the flows' deadline. Then PCC adjusts the congestion window according to the flow priority and the degree of network congestion. Furthermore, switches in data centers control the input/output of packets based on the flow priority and the queue length. Different from existing TCPs, to speed up the data transfers of mice and deadline-sensitive flows, PCC provides an effective method to compute and encode the flow priority explicitly. According to the flow priority, switches can manage packets efficiently and ensure the data transfers of high priority flows through a weighted priority scheduling with minor modification. The experimental results prove that PCC can improve the data transfer performance of mice and deadline-sensitive flows while guaranting the throughput of elephant flows.

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.

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.

Routing and load balancing are two important factors that guarantee Quality of Service (QoS) in the Internet. Meanwhile Software Defined Networking (SDN) has introduced a new networking paradigm that make routing and load balancing policies implemented more quickly and flexibly in the system. In the SDN architecture, the "intelligence" of the network is centralized in the application layer and control layer instead of distributing in the network device in traditional network. Current studies using SDN-based routing and load balancing still stop at the scope of finding appropriate network resource for one single data flow but not considering network resource for all flows coming from different sources as the whole. In this paper, we propose a global load-balanced routing scheme, which can take advantages of global view of the SDN controller to make a global policy for routing and load balancing. The solution has been shown to outperform traditional algorithms and recent SDN-based artificial intelligence algorithms in terms of delay and network utilization.

Modern data center topologies often take the form of a multi-rooted tree with rich parallel paths to provide high bandwidth. However, various path diversities caused by traffic dynamics, link failures and heterogeneous switching equipments widely exist in production datacenter network. Therefore, the multi-path load balancer in data center should be robust to these diversities. Although prior fine-grained schemes such as RPS and Presto make full use of available paths, they are prone to experience packet reordering problem under asymmetric topology. The coarse-grained solutions such as ECMP and LetFlow effectively avoid packet reordering, but easily lead to under-utilization of multiple paths. To cope with these inefficiencies, we propose a load balancing mechanism called AG, which adaptively adjusts switching granularity according to the asymmetric degree of multiple paths. AG increases switching granularity to alleviate packet reordering under large degrees of topology asymmetry, while reducing switching granularity to obtain high link utilization under small degrees of topology asymmetry. AG is deployed on the switches with negligible overhead, while making no modification on end-hosts. We evaluate AG through both Mininet testbed and large-scale NS2 simulations. The experimental results show that AG reduces the average and 99th flow completion time by up to 51% and 56% over the state-of-the-art load balancing schemes, respectively.