Overview of Stream-Aware Scheduling for QUIC-Based Proxying Using Multiple Paths

A TCP flow is expected to traverse over a single path. This is in order to avoid performance problems especially on account of reordering of packets and acknowledgements. Reordering can occur on account of the use of different paths with diverse bandwidth and delay characteristics. We argue that the present approach prevents us from getting the advantages associated with multiple paths. Specifically, use of multiple paths for a flow can be advantageous in improving the performance as well as the dependability of a flow. We model the effects of reordering and derive conditions under which the performance of a TCP flow improves with the use of multiple paths. We also provide recommendations for choosing amongst multiple paths. We then verify the conditions using simulations. We also illustrate the applicability of the derived conditions to multiple flows.

Large scale InfiniBand clusters are becoming increasingly popular, as reflected by the TOP 500 Supercomputer rankings. At the same time, fat tree has become a popular interconnection topology for these clusters, since it allows multiple paths to be available in between a pair of nodes. However, even with fat tree, hot-spots may occur in the network depending upon the route configuration between end nodes and communication pattern(s) in the application. To make matters worse, the deterministic routing nature of InfiniBand limits the application from effective use of multiple paths transparently and avoid the hot-spots in the network. Simulation based studies for switches and adapters to implement congestion control have been proposed in the literature. However, these studies have focused on providing congestion control for the communication path, and not on utilizing multiple paths in the network for hot-spot avoidance. In this paper, we design an MPI functionality, which provides hot-spot avoidance for different communications, without a priori knowledge of the pattern. We leverage LMC (LID Mask Count) mechanism of InfiniBand to create multiple paths in the network and present the design issues (scheduling policies, selecting number of paths, scalability aspects) of our design. We implement our design and evaluate it with Pallas collective communication and MPI applications. On an InfiniBand cluster with 48 processes, collective operations like MPI All-to-all Personalized and MPI Reduce Scatter show an improvement of 27% and 19% respectively. Our evaluation with MPI applications like NAS Parallel Benchmarks and PSTSWM on 64 processes shows significant improvement in execution time with this functionality.

We introduce the game-theoretic stochastic routing (GTSR) framework, a proactive alternative to today's reactive approaches to route repair. GTSR minimizes the impact of link and router failure by 1) computing multiple paths between source and destination and 2) selecting among these paths randomly to forward packets. Besides improving fault tolerance, the fact that GTSR makes packets take random paths from source to destination also improves security. In particular, it makes connection eavesdropping attacks maximally difficult as the attacker would have to listen on all possible routes. The approaches developed are suitable for network layer routing, as well as for application layer overlay routing and multipath transport protocols such as the stream control transmission protocol (SCTP). Through simulations, we validate our theoretical results and show how the resulting routing algorithms perform in terms of the security/fault-tolerant/delay/throughput trade-off. We also show that a beneficial side effect of these algorithms is an increase in throughput, as they make use of multiple paths.

Path diversity exploits multiple routes simultaneously, achieving higher aggregated bandwidth and potentially decreasing delay and packet loss. Unfortunately, for TCP, naive load splitting often results in inaccurate estimation of round trip time (RTT) and packet reordering. As a result, it can suffer from significant instability or even throughput reduction. This is particular severe in Wireless Mesh Networks (WMNs), as validated by our analysis and simulation. To make multi-path TCP viable over WMNs, we propose a novel cross-layer design with a smart traffic split scheme, namely, Path Diversified Retransmission (PDR). PDR differentiates the original data packets and the retransmitted packets, and works with a novel QoS-aware multi-path routing protocol, QAOMDV, to distribute them separately. PDR does not suffer from the RTT underestimation and extra packet reordering, which ensures stable throughput improvement over single path routing. Through extensive simulations, we further demonstrate that, as compared to state-of-the-art multi-path protocols, our PDR with QAOMDV noticeably enhances the TCP throughput and reduces bandwidth fluctuation, with no obvious impact to fairness.

Multipath Transmission Control Protocol (MPTCP) promises higher bandwidth and higher resilience against network path failures as an evolving technology for the future Internet. MPTCP allows multiple paths between two devices to be pooled and appear to the application as a single end-to-end transport connection. In this paper, we propose a solution for the packet reordering problem, to improve path utilization and aggregate throughput. The proposed method maintains the congestion window size when three duplicate acknowledgements occur to avoid unnecessary reductions in transmission rate. We evaluate the performance of the proposed solution using an NS-3 in different wireless scenarios and compare it against the performance of other existing reordering methods. The results show that the proposed method improves the aggregate throughput and path utilization when the packet drop rate is high.

The current inter-domain routing in the Internet, which is based on BGP-4, does not allow for the use of multiple paths, but rather restricts the routing to a single path for each destination prefix. This fact is especially unfortunate considering the vast route diversity which is inherently present in the global Internet graph. Therefore, we propose Inter-Domain Route Diversity (IDRD) as an overlay mechanism which enables efficient, backwards compatible and incrementally deployable introduction of route diversity in the Internet. Beyond presenting the architecture of IDRD, this paper also presents the conditions which ensure the stability of the proposed mechanism as a fundamental prerequisite for its deployment in real-world scenarios.

Reservation-based wireless TDMA networks can provide deterministic quality-of-service guarantees to multi-hop data flows. The challenge is to route the flows and schedule appropriate transmission slots in a resource-efficient way. In this paper, we study the fundamental question whether multi-path routing, i.e. concurrent use of multiple paths for transmission of data belonging to the same flow, may yield better solutions to the joint routing and slot scheduling problem than single-path routing approaches. To answer this question, we first define an accurate network model. We then devise a branch-and-bound algorithm that computes optimal multi-path and single-path solutions to the joint routing and slot scheduling problem. We apply the algorithm to a large number of scenarios in randomly generated networks and compare optimal network load and end-to-end delay metric values in order to quantify the benefit of multi-path routing over the single-path approach. Our results show that in the studied settings, it is quite rare for multi-path solutions to surpass the optimal single-path solution. Multi-path routing should therefore not be considered an option to improve performance in reservation-based TDMA networks with random topologies.

Mobile Ad hoc Networks (MANET) are wireless networks consisting of a collection of mobile nodes with no fixed infrastructure. Due to their decentralized, self-configuring and dynamic nature, MANETs offer many advantages and are easy to install. But with this dynamic topology, mobile ad hoc networks have some challenges like the design of an efficient routing protocol. An example for this challenge is load balancing. The multipath routing protocol with load balancing provides a solution for the congestion network and increases it's capacity. To consider that the use of multiple paths simultaneously for transmission data allows to improve the network performance, we propose a new protocol LB-AOMDV (Load Balancing-AOMDV), a solution to achieve better load balancing mechanism. The simulation's result shows the significant performance improvement of the network for the multipath routing protocol with load balancing. The proposed solution LB-AOMDV works better than other protocols in terms of average delay, capacity and load balance.

Human Positioning Estimation Method Using Received Signal Strength Indicator (RSSI) in a Wireless Sensor Network

Reducing traffic burstiness for MPTCP in data center networks

The self-organizing nature of the Mobile Ad hoc Networks (MANETs) provide a communication channel anywhere, anytime without any pre-existing network infrastructure. However, it is exposed to various vulnerabilities that may be exploited by the malicious nodes. One such malicious behavior is introduced by blackhole nodes, which can be easily introduced in the network and, in turn, such nodes try to crumble the working of the network by dropping the maximum data under transmission. In this paper, a new protocol is proposed which is based on the widely used Ad hoc On-Demand Distance Vector (AODV) protocol, Enhanced Secure Trusted AODV (ESTA), which makes use of multiple paths along with use of trust and asymmetric cryptography to ensure data security. The results, based on NS-3 simulation, reveal that the proposed protocol is effectively able to counter the blackhole nodes in three different scenarios.

The number of mobile terminals that have both a cellular interface and a Wi-Fi interface has been growing in the recent years. Meanwhile, Multipath TCP (MPTCP) that enables the simultaneous use of multiple paths with multiple TCP flows while showing just one regular TCP connection with the application has been standardized by the IETF and has drawn more attention recently. One feature of MPTCP is that it can be realized only by operating system layers supporting multipath communication without any modifications of the application. However, it might be difficult to install MPTCP in all server operating systems. In this paper, we propose a method for applying a proxy that forwards TCP communication from a server to MPTCP communication on a mobile core network, and show the results of evaluation with the existing equipment.

A storage-area (SAN) comprises computers (initiators), storage block devices (targets), and a controller(s). Most SANs use the SCSI protocol over various communication infrastructures. Presently, all initiator-target traffic must pass through the controller, severely limiting scalability. We extend the SCSI-3 transport layer to support distribution, and combine this with SCSIs support for data-control split to create DSDC, a novel architecture that can be used over any networking infrastructure: data may be sent directly between initiators and targets, relieving the controller communication bottleneck; the use of multiple paths for data moreover relieves traffic bottlenecks on network links; finally, passing all commands through the Controller retains simplicity. DSDC thus enables the construction of much larger SANs while retaining the simplicity of a single controller. A prototype SAN using Ethernet and Linux nodes, with DSDC implemented in the iSCSI transport layer protocol and in the controller's SCSI application layer, has been constructed.

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.

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 equipment widely exist in the 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 the best use of available paths, However, they are prone to experiencing packet reordering problem under the 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 PDLB, which adaptively adjusts flowcell granularity according to path diversity. PDLB increases flowcell granularity to alleviate packet reordering under large degrees of topology asymmetry, while reducing flowcell granularity to obtain high link utilization under small degrees of topology asymmetry. PDLB is only deployed on the sender without any modification on switch. We evaluate PDLB through large-scale NS2 simulations. The experimental results show that PDLB reduces the average flow completion time by up to ∼\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\sim$$\\end{document}8-53% over the state-of-the-art load balancing schemes.