VPS Network Performance: Bandwidth and Latency Optimization

Are your API response times high, or are file download speeds falling below expectations? Even if your VPS has sufficient CPU and RAM, default Linux kernel network settings are not optimized for most server workloads. VPS network performance optimization - changing the TCP congestion control algorithm, adjusting buffer sizes, and configuring connection limits - increases throughput and reduces latency. In this guide, we'll improve your network performance with concrete sysctl settings and monitoring tools.

Essential Network Metrics

Increasing Throughput with TCP BBR

Developed by Google, BBR (Bottleneck Bandwidth and Round-trip propagation time) works by modeling bandwidth and RTT, unlike traditional loss-based congestion control algorithms. It significantly increases throughput, especially on connections with high latency or packet loss.

# Check current congestion control algorithm
sysctl net.ipv4.tcp_congestion_control

# Enable BBR
sudo sysctl -w net.core.default_qdisc=fq
sudo sysctl -w net.ipv4.tcp_congestion_control=bbr

# Make it persistent
echo 'net.core.default_qdisc=fq' | sudo tee -a /etc/sysctl.conf
echo 'net.ipv4.tcp_congestion_control=bbr' | sudo tee -a /etc/sysctl.conf

# Verify
sysctl net.ipv4.tcp_congestion_control
# Output: net.ipv4.tcp_congestion_control = bbr

Kernel Network Parameters

Default Linux kernel network settings are designed for low-resource systems. Optimize the following sysctl settings for web servers and APIs:

# TCP buffer sizes (for high bandwidth)
net.core.rmem_max = 16777216
net.core.wmem_max = 16777216
net.ipv4.tcp_rmem = 4096 87380 16777216
net.ipv4.tcp_wmem = 4096 65536 16777216

# Connection queue (for heavy traffic)
net.core.somaxconn = 65535
net.core.netdev_max_backlog = 65535

# Quickly recycle TIME_WAIT sockets
net.ipv4.tcp_tw_reuse = 1
net.ipv4.tcp_fin_timeout = 15

# TCP keepalive (connection liveness check)
net.ipv4.tcp_keepalive_time = 300
net.ipv4.tcp_keepalive_intvl = 30
net.ipv4.tcp_keepalive_probes = 5

# SYN flood protection
net.ipv4.tcp_syncookies = 1
net.ipv4.tcp_max_syn_backlog = 65535

DNS Optimization

Every HTTP request starts with a DNS resolution. Slow DNS slows down all connections. You can speed up repeated queries by setting up a local DNS cache:

# Local DNS cache with systemd-resolved
sudo systemctl enable --now systemd-resolved

# Test DNS resolution time
dig hosted.cloud | grep "Query time"

# Use fast DNS servers (/etc/systemd/resolved.conf)
# DNS=1.1.1.1 8.8.8.8
# FallbackDNS=9.9.9.9 8.8.4.4

Network Monitoring Tools

# Bandwidth test (iperf3)
sudo apt install iperf3 -y
iperf3 -c speedtest.serverius.net -p 5002

# Latency and packet loss test
mtr --report --report-cycles 100 google.com

# Real-time bandwidth monitoring
nload eth0

# Connection status summary
ss -s

For comprehensive server performance monitoring, you can also check out our resource monitoring guide. Network performance is optimized with high bandwidth and low latency on Hosted Cloud cloud servers.

Frequently Asked Questions

Conclusion

VPS network performance optimization starts with concrete steps like enabling TCP BBR, adjusting kernel buffer sizes, and setting up DNS caching. These changes are typically applied in a few minutes and provide noticeable improvement in throughput. Back up your current settings before making changes and compare before/after results with iperf3.