Modern data center networks are usually constructed in multi-rooted tree topologies, which require the highly efficient multi-path load balancing to achieve high link utilization. Recent packet-level load balancer obtains high throughput by spraying packets to all paths, but it easily leads to the packet reordering under network asymmetry. The flow-level or flowlet-level load balancer avoids the packet reordering, while reducing the link utilization due to their inflexibility. To solve these problems, we design a Queueing Delay Aware Packet Spraying (QDAPS), that effectively mitigates the packet reordering for packet-level load balancer. QDAPS selects paths for packets according to the queueing delay of output buffer, and lets the packet arriving earlier be forwarded before the later packets to avoid packet reordering. Moreover, we adopt the “power-of- $n$ -choices” paradigm on QDAPS to alleviate the impact of herd behavior under multiple forwarding engines. We compare QDAPS with ECMP, LetFlow and RPS through NS2 simulation and Mininet implementation. The test results show that QDAPS reduces flow completion time (FCT) by ~30%-50% over the state-of-the-art load balancing mechanism.

中文翻译：

---

缓解数据中心网络中随机数据包喷射的数据包重新排序

现代数据中心网络通常采用多根树形拓扑结构，需要高效的多路径负载均衡来实现高链路利用率。最近的包级负载均衡器通过向所有路径喷射数据包来获得高吞吐量，但它容易导致网络不对称下的数据包重新排序。流级或流级负载均衡器避免了数据包重新排序，同时由于它们的不灵活性而降低了链路利用率。为了解决这些问题，我们设计了一种队列延迟感知数据包喷射（QDAPS），它有效地缓解了数据包级负载均衡器的数据包重新排序。QDAPS 根据输出缓冲区的排队延迟为数据包选择路径，并让较早到达的数据包在较晚的数据包之前转发，以避免数据包重新排序。而且， $n$ -choices”范式在 QDAPS 上减轻群体行为在多个转发引擎下的影响。我们通过 NS2 模拟和 Mininet 实现将 QDAPS 与 ECMP、LetFlow 和 RPS 进行了比较。测试结果表明，与最先进的负载平衡机制相比，QDAPS 将流完成时间 (FCT) 减少了约 30%-50%。