Various datacenter network (DCN) load balancing schemes have been proposed in the past decade. Unfortunately, most of these solutions designed for lossy DCNs do not work well Priority Flow Control (PFC) enabled lossless DCNs, primarily due to reason that individual congestion signals used solutions, e.g., link load, queue length, Round Trip Time (RTT) and Explicit Congestion Notification (ECN), may be able correctly or timely reflect hop-by-hop PFC pausing. This paper first reveals above...

Many existing load balancing mechanisms work effectively in lossy datacenter networks (DCNs), but they suffer from serious packet reordering lossless Ethernet DCNs deployed with the hop-by-hop Priority-based Flow Control (PFC). The key reason is that prior solutions are not able to perceive PFC triggering correctly and a timely manner when making decisions. Once forwarding path pauses transmission due triggering, packets allocated on it blocked, inevitably leading out-of-order...

Modern data-center applications generate a diverse mix of short and long flows with different performance requirements weaknesses. The are typically delay-sensitive but to suffer the head-of-line blocking out-of-order problems. Recent solutions prioritize meet their latency requirements, while damaging throughput-sensitive flows. To solve these problems, we design Coding-based Adaptive Packet Spraying (CAPS) that effectively mitigates negative impact on each other. exploit availability...

Various datacenter network (DCN) load balancing schemes have been proposed in the past decade. Unfortunately, most of these solutions designed for lossy DCNs do not work well Priority Flow Control (PFC) enabled lossless DCNs, primarily due to reason that individual congestion signals used solutions, e.g., link load, queue length, Round Trip Time (RTT) and Explicit Congestion Notification (ECN), may be able correctly or timely reflect hop-by-hop PFC pausing. This paper first reveals above...

Existing reactive or proactive congestion control protocols are hard to simultaneously achieve ultra-low latency and high link utilization across all workloads ranging from delay-sensitive flows bandwidth-hungry ones in datacenter networks. We present an Anti-ECN (Explicit Congestion Notification) Marking Receiver-driven Transport protocol called AMRT, which achieves both near-zero queueing delay full by reasonably increasing sending rate the case of under-utilization. Specifically, switches...

Data security and user privacy have become crucial elements in multi-tenant data centers. Various traffic types the center cloud environment their characteristics requirements. In network (DCN), short long flows are sensitive to low latency high throughput, respectively. The traditional processing approaches, however, neglect these This paper proposes a fine-grained enhancement mechanism (SEM) solve problem of heterogeneous reduce completion time (FCT) while ensuring transmission....

Many existing load balancing mechanisms work effectively in lossy datacenter networks (DCNs), but they suffer from serious packet reordering lossless Ethernet DCNs deployed with the hop-by-hop Priority-based Flow Control (PFC). The key reason is that prior solutions are not able to correctly and timely perceive PFC triggering when making decisions. Once forwarding path pauses transmission due triggering, packets allocated on it blocked, inevitably leading out-of-order retransmission. In this...

The state-of-the-art datacenter load balancing designs commonly optimize bisection bandwidth with homogeneous switching granularity. Their performances surprisingly degrade under mixed traffic containing both short and long flows. Specifically, the flows suffer from long-tailed delay, while throughputs of also dramatically due to low link utilization packet reordering. To solve these problems, we design a traffic-aware (TLB) scheme adaptively adjust granularity according strength ones. Under...

In the Ethernet lossless Data Center Networks (DCNs) deployed with Priority-based Flow Control (PFC), head-of-line blocking problem is still difficult to prevent due PFC triggering under burst traffic scenarios even existing congestion control solutions. To address of PFC, we propose a new mechanism. The key point Congestion Using In-Network Telemetry for Lossless Datacenters (ICC) use (INT) technology obtain comprehensive information, which then fed back sender adjust sending rate timely...

Due to the explosive growth of all kinds Internet services, data centers have become an irreplaceable and vital infrastructure support this soaring trend. Compared with traditional networks, center networks (DCNs) unique features, such as high bandwidth, low latency, many-to-one communication mode, shallow buffered switches, multi-root topology. These new characteristics pose a lot challenges previous network technics (e.g., Ethernet, Equal Cost Multi-path (ECMP), TCP), making them hard...

In Priority Flow Control (PFC) enabled datacenter networks (DCNs), PFC is inevitably triggered due to bursty traffic even with end-to-end congestion control. Load balancing as a complementary mechanism transport protocols can make rerouting decisions in time alleviate PFC's head-of-line (HoL) blocking problem. However, prior solutions designed for lossy DCNs do not work well PFC-enabled networks, because the unreliable signals such separate local queue length, round-trip (RTT), explicit...

Modern data-center applications generate a diverse mix of short and long flows with different performance requirements weaknesses. The are typically delay-sensitive but to suffer the head-of-line blocking out-of-order problems. Recent solutions prioritize meet their latency requirements, while damaging throughput-sensitive flows. To solve these problems, we design Coding-based Adaptive Packet Spraying (CAPS) that effectively mitigates negative impact on each other. exploit availability...

Modern datacenter topologies typically are multi-rooted trees consisting of multiple paths between any given pair hosts. Recent load balancing designs focus on making full use available parallel to provide high bisection bandwidth. However, they agnostic the mixed traffic generated by diverse applications in data centers and respectively same granularity rerouting flows regardless flow type. Therefore, short suffer long-tailed queueing delay reordering problems, while throughputs long also...

In Vehicular Ad Hoc Networks (VANETs), the high mobility of vehicle nodes makes network topology change frequently, reducing forwarding efficiency MAC protocol. existing enhanced TDMA-based protocol, farthest node in current transmission range is chosen as to accelerate multi-hop transmission. However, we use probabilistic model show that there potentially exist better nodes, which could effectively improve efficiency. Therefore, propose a motion-prediction based TDMA predicts next frame...

As datacenter networks continue to support a wider range of applications and faster link speeds, they face the challenge managing bursty traffic transient congestion. End-to-end congestion controls (CCs) find it increasingly difficult maintain effective due inherent feedback delay. To address this issue, per-hop flow control (FC) has gained popularity its ability react promptly However, existing FC mechanisms either lack fine-grained (i.e., per-flow granularity) or require an impractical...