Modern datacenter networks host diverse services and workloads with varying quality-of-service (QoS) requirements, yet are constrained by hardware limitations. Most notably, the small number of physical priority queues available in commodity switches. Existing scheduling mechanisms, whether end-host or in-network based, struggle to scale under these constraints due to their reliance on global priority information or complex queue management. This paper presents P2CS (Priority-based Probabilistic Congestion Signaling), a lightweight and scalable approach that enables fine-grained traffic prioritization using only a single FIFO queue. P2CS combines priority-aware probabilistic congestion signaling, priority-aware packet dropping, and simple switch-side arbitration to enforce prioritization across flows. P2CS supports a range of scheduling objectives, and requires minimal software changes making it readily deployable in today's datacenter infrastructure. Evaluation on representative workloads, including multi-tenant ML training, HPC, and mixed spray/ECMP traffic, demonstrates that P2CS achieves performance comparable to in-network mechanisms while significantly reducing complexity and cost.

Next generation applications for datacenters, such as Distributed Machine Learning (DML) and Big Data, have complex communication patterns that demand a scalable, stateless and application-aware optimal transport protocol to maximize network utilization and improve application performance. Recent transport protocols either provide limited benefits due to lack of information sharing between application and network; or implement complex stateful mechanisms to improve the application performance. In this paper, we present Omni- Transport Mechanism (Omni-TM) as a message-based congestion control protocol. Omni-TM allows exchanging message information with the network to negotiate the optimal transmission rate without maintaining a per-flow state at the switches (i.e., stateless). Omni- Tmis designed to reach maximum link capacity in one shot. Our simulation results show that Omni- Tmdemonstrates better traffic control decisions (i.e., close to zero queue length while maintaining high link utilization). Furthermore, Omni- Tmreduces Flow Completion Time (FCT) up to 45 % in a realistic workload compared to DCTCP.

To improve the performance of data-intensive applications, existing datacenter schedulers optimize either the placement of tasks or the scheduling of network flows. The task scheduler strives to place tasks close to their input data (i.e., maximize data locality) to minimize network traffic, while assuming fair sharing of the network. The network scheduler strives to finish flows as quickly as possible based on their sources and destinations determined by the task scheduler, while the scheduling is based on flow properties (e.g., size, deadline, and correlation) and not bound to fair sharing. Inconsistent assumptions of the two schedulers can compromise the overall application performance. In this paper, we propose NEAT+, a task scheduling framework that leverages information from the underlying network scheduler and available compute resources to make task placement decisions. The core of NEAT+ is a task completion time predictor that estimates the completion time of a task under given network condition and a given network scheduling policy. NEAT+ leverages the predicted task completion times to minimize the average completion time of active tasks. Evaluation using ns2 simulations and real-testbed shows that NEAT+ improves application performance by up to 3.7x for the suboptimal network scheduling policies and up to 33% for the optimal network scheduling policy.

Recent trends show that cloud computing is growing to span more and more globally distributed datacenters. For geo-distributed datacenters, there is an increasingly need for scheduling algorithms to place tasks across datacenters, by jointly considering WAN traffic and computation. This scheduling must deal with situations such as wide-area distributed data, data sharing, WAN bandwidth costs and datacenter capacity limits, while also minimizing makespan. However, this scheduling problem is NP-hard. We propose a new resource allocation algorithm called HPS+, an extension to Hypergraph Partition-based Scheduling. HPS+ models the combined task-data dependencies and data-datacenter dependencies as an augmented hypergraph, and adopts an improved hypergraph partition technique to minimize WAN traffic. It further uses a coordination mechanism to allocate network resources closely following the guidelines of task requirements, for minimizing the makespan. Evaluation across the real China-Astronomy-Cloud model and Google datacenter model show that HPS+ saves the amount of data transfers by upto 53 percent and reduces the makespan by 39 percent compared to existing algorithms.

Inefficient bandwidth sharing in a datacenter network, between different application frameworks, e.g., MapReduce and Spark, can lead to inelastic and skewed usage of link bandwidth and increased completion times for the applications. Existing work, however, either solely focuses on managing computation and storage resources or controlling only sending/receiving rate at hosts. In this paper, we present CoMan, a solution that provides global in-network bandwidth management in multiplexed data centers, with two goals: improving bandwidth utilization and reducing application completion time. CoMan first designs a novel abstraction of virtual link groups (VLGs) to establish a shared bandwidth resource pool. Based on this pool, CoMan implements a three-level bandwidth allocation model, which enables elastic bandwidth sharing among computing frameworks as well as guarantees network performance for the applications. CoMan further improves the bandwidth utilization by devising a VLG dependency graph and solves an optimization problem to guide the path selection using a 32-approximation algorithm. We conduct comprehensive trace-driven simulations as well as small-scale testbed experiments to evaluate the performance of CoMan. Extensive simulation results show that CoMan improves the bandwidth utilization and speeds up the application completion time by up to 2.83× and 6.68×, respectively, compared to the ECMP + ElasticSwitch solution. Our implementation also verifies that CoMan can realistically speed up the application completion times by 2.32× on average.

Multipath TCP (MPTCP) is an IETF standardized suite of TCP extensions that allow two endpoints to simultaneously use multiple paths between them. In this paper, we report vulnerabilities in MPTCP that arise because of cross-path interactions between MPTCP subflows. First, an attacker eavesdropping one MPTCP subflow can infer throughput of other subflows. Second, an attacker can inject forged MPTCP packets to change priorities of any MPTCP subflow. We present two attacks to exploit these vulnerabilities. In the connection hijack attack, an attacker takes full control of the MPTCP connection by suspending the subflows he has no access to. In the traffic diversion attack, an attacker diverts traffic from one path to other paths. Proposed vulnerabilities fixes, changes to MPTCP specification, provide the guarantees that MPTCP is at least as secure as TCP and the original MPTCP. We validate attacks and prevention mechanism, using MPTCP Linux implementation (v0.91), on a real-network testbed.

In datacenter networks, flows can have different performance objectives. We use a tenant-objective division to denote all flows of a tenant that share the same objective. Bandwidth allocation in datacenters should support not only performance isolation among divisions but also objective-oriented scheduling among flows within the same division. This paper studies the Multi-Tenant Multi-Objective (MT-MO) bandwidth allocation problem. To our best knowledge, no existing practical work support performance isolation and objective scheduling simultaneously. We propose Stacked Congestion Control (SCC), a distributed host-based bandwidth allocation design, where an underlay congestion control (UCC) layer handles contention among divisions, and a private congestion control (PCC) layer for each division optimizes its performance objective. Via the tenant-objective tunnel abstraction, SCC achieves weighted bandwidth sharing for each division in a distributed and transparent way. By adding a rate-limiting send queue in the ingress of each tunnel, mechanisms between performance isolation and objective scheduling are completely decoupled. We evaluate SCC both on a small-scale testbed and with large-scale NS-2 simulations. Compared to the direct coexistence cases, SCC reduces latency by up to 40% for Latency-Sensitive flows, deadline miss ratio by up to 3.2× for Deadline-Sensitive flows, and average flow-completion-time by up to 53% for Completion-Sensitive flows.

Outdoor localization is of great importance for driving navigation, attracting many research efforts in past decades. Prevailing GPS achieves meter-level localization accuracy under general outdoor conditions. Yet, GPS service performs poorly in urban canyons where skyscrapers blocks GPS signals and drain mobile phone battery quickly within few hours. In this work, we exploit common city facilities, i.e. speed humps, as an indicator for vehicle location. The key insight is that when the vehicle passes through the speed bump, it experiences significant fluctuations, causing larger acceleration in the vertical direction. On this basis, we design a localization scheme that utilizes the accelerator equipped on modern smart phones to track sequence of speed bumps, which is further transferred into sequence of moving directions of the vehicle, and adopt effective road mapping technology to derive real-time location. As we have concerned, it is the first attempt to exploit the spatiotemporal characteristics generated by the speed humps to recover the trajectory of the travelling route and infer the current position. Experimental results in typical outdoor environment (campus) demonstrate a comparable performance with GPS method, yet achieve lower energy consumption.

Several data center transport protocols have been proposed in recent years (e.g., DCTCP, PDQ, and pFabric). In this paper, we first identify the underlying strategies used by the existing data center transports, namely, in-network Prioritization (used in pFabric), Arbitration (used in PDQ), and Self-adjusting at Endpoints (PASE) (used in DCTCP). We show that these strategies are complimentary to each other, rather than substitutes, as they have different strengths and can address each other's limitations. Unfortunately, prior data center transports use only one of these strategies. As a result, they either achieve near-optimal performance or deployment friendliness (i.e., require no changes to the data plane) but not both. Based on this insight, we design a data center transport protocol called PASE, which carefully synthesizes these strategies by assigning different transport responsibilities to each strategy. The key advantage of PASE over prior art is that it achieves both near-optimal performance as well as deployment friendliness. PASE does not require any changes in network switches (hardware or software); yet, it achieves comparable, or even better, performance than the state-of-the-art protocols (such as pFabric) that require changes to network elements. Our evaluation results show that the PASE performs well for a wide range of application workloads and network settings.

To improve the performance of data-intensive applications, existing datacenter schedulers optimize either the placement of tasks or the scheduling of network flows. The task scheduler strives to place tasks close to their input data (i.e., maximize data locality) to minimize network traffic, while assuming fair sharing of the network. The network scheduler strives to finish flows as quickly as possible based on their sources and destinations determined by the task scheduler, while the scheduling is based on flow properties (e.g., size, deadline, and correlation) and not bound to fair sharing. Inconsistent assumptions of the two schedulers can compromise the overall application performance. In this paper, we propose NEAT, a task scheduling framework that leverages information from the underlying network scheduler to make task placement decisions. The core of NEAT is a task completion time predictor that estimates the completion time of a task under given network condition and a given network scheduling policy. NEAT leverages the predicted task completion times to minimize the average completion time of active tasks. Evaluation using ns2 simulations and real-testbed shows that NEAT improves application performance by up to 3.7x for the suboptimal network scheduling policies and up to 30% for the optimal network scheduling policy.

We propose OpenFunction, an extensible data plane abstraction protocol for platform-independent software-defined middleboxes. The main challenge is how to abstract packet operations, flow states and event generations with elements. The key decision of OpenFunction is: actions/states/events operations should be defined in a uniform pattern and independent from each other. We implemented a working SDM system including one OpenFunction controller and OpenFunction boxes based on Netmap, DPDK and FPGA to verify OpenFunction abstraction.

Many data center transports have been proposed in recent times (e.g., DCTCP, PDQ, pFabric, etc). Contrary to the common perception that they are competitors (i.e., protocol A vs. protocol B), we claim that the underlying strategies used in these protocols are, in fact, complementary. Based on this insight, we design PASE, a transport framework that synthesizes existing transport strategies, namely, self-adjusting endpoints (used in TCP style protocols), innetwork prioritization (used in pFabric), and arbitration (used in PDQ). PASE is deployment friendly: it does not require any changes to the network fabric; yet, its performance is comparable to, or better than, the state-of-the-art protocols that require changes to network elements (e.g., pFabric). We evaluate PASE using simulations and testbed experiments. Our results show that PASE performs well for a wide range of application workloads and network settings.

Today's data centers face extreme challenges in providing low latency for online services such as web search, social networking, and recommendation systems. Achieving low latency is important as it impacts user experience, which in turn impacts operator revenue. However, most current congestion control protocols approximate Processor Sharing (PS), which is known to be sub-optimal for minimizing latency. In this paper, we propose Router Assisted Capacity Sharing (RACS), a data center transport protocol that minimizes flow completion times by approximating the Shortest Remaining Processing Time (SRPT) scheduling policy, which is known to be optimal, in a distributed manner. With RACS, flows are assigned weights which determine their relative priority and thus the rate assigned to them. By changing these weights, RACS can approximate a range of scheduling disciplines. Through extensive ns-2 simulations, we demonstrate that RACS outperforms TCP, DCTCP, and RCP in data center environments. In particular, it improves completion times by up to 95% over TCP, 88% over DCTCP, and 80% over RCP. Our results also show that RACS can outperform deadline-aware transport protocols for typical data center workloads.

For provisioning large-scale online applications such as web search, social networks and advertisement systems, data centers face extreme challenges in providing low latency for short flows (that result from end-user actions) and high throughput for background flows (that are needed to maintain data consistency and structure across massively distributed systems). We propose L2DCT, a practical data center transport protocol that targets a reduction in flow completion times for short flows by approximating the Least Attained Service (LAS) scheduling discipline, without requiring any changes in application software or router hardware, and without adversely affecting the long flows. L2DCT can co-exist with TCP and works by adapting flow rates to the extent of network congestion inferred via Explicit Congestion Notification (ECN) marking, a feature widely supported by the installed router base. Though L2DCT is deadline unaware, our results indicate that, for typical data center traffic patterns and deadlines and over a wide range of traffic load, its deadline miss rate is consistently smaller compared to existing deadline-driven data center transport protocols. L2DCT reduces the mean flow completion time by up to 50% over DCTCP and by up to 95% over TCP. In addition, it reduces the completion for 99th percentile flows by 37% over DCTCP. We present the design and analysis of L2DCT, evaluate its performance, and discuss an implementation built upon standard Linux protocol stack.