NIC Multi-Queue

The Network Interface Cards (NICs) Multi-Queue feature enhances the network performance of Dedicated Core Servers and vCPU Servers by associating multiple queues with a VM based on the number of cores. It enables the VM to distribute network traffic across multiple queues, which improves throughput and reduces latency, particularly during high-traffic periods. By utilizing multiple CPU cores to process network packets, NIC Multi-Queue efficiently handles network Input or Output (I/O) and prevents bottlenecks commonly found in single-queue NICs. This feature benefits applications that require high network bandwidth and low latency, such as databases, analytics platforms, and high-performance computing workloads.

NIC Multi-Queue is available in all supported data centers. It is off by default; to use this feature, you must turn it on for each VM using the DCD or IONOS Cloud API for Dedicated Core Servers and vCPU Servers.

System requirements

The feature is supported on any VM configuration that meets the two-thread minimum, regardless of core count or processor architecture. A minimum of two CPU threads can be provided by:

  • Multiple CPU cores.

  • Hyper-threading on single-core systems. Examples:

    • A VM with one Intel Xeon core (providing two threads via hyper-threading) meets the minimum requirement for NIC Multi-Queue.

    • Certain CPU families, such as Intel Sierra Forest, do not support hyperthreads, and therefore require two cores to enable this feature.

Features

Developer Tools

Last updated

Was this helpful?