Association for Computing Machinery
Datacenter networks deal with a variety of workloads, ranging from latency-sensitive small flows to bandwidth-hungry large flows. Load balancing schemes based on flow hashing, e.g., ECMP, cause congestion when hash collisions occur and can perform poorly in asymmetric topologies. Recent proposals to load balance the network require centralized traffic engineering, multipath-aware transport, or expensive specialized hardware. The authors propose a mechanism that avoids these limitations by pushing load-balancing functionality into the soft network edge (e.g., virtual switches) such that no changes are required in the transport layer, customer VMs, or networking hardware and load balancing on fine-grained, near-uniform units of data (flowcells) that fit within end-host segment offload optimizations used to support fast networking speeds.