Host Tuning
Background Information
Linux
Expert Guide
Measurement Host Tuning
Mac OSX
FreeBSD
MS Windows
Other OS
NIC Tuning
Interrupt Binding
Virtual Machines
Send us feedback about this page
Linux Tuning
Table of Contents
General Approach
TCP tuning
UDP Tuning
NIC Tuning
Virtual Machine Tuning
This page contains a quick reference guide for Linux 2.6+ tuning for Data
Transfer hosts connected at speeds of 1Gbps or higher. Note that most of the
tuning settings described here will actually decrease performance of hosts
connected at rates of OC3 (155 Mbps) or less, such as home users on Cable/DSL
connections.
For a detailed explanation of some of the advice on this page, see the Linux
Tuning Expert page. Note that the settings on this page are not attempting to
achieve full 10G with a single flow. These settings assume you are using
tools that support parallel streams, or have multiple data transfers
occurring in parallel, and want to have fair sharing between the flows.
If you are trying to optimize for a single flow, see the tuning advice for
test/measurement hosts page.
General Approach
To check what setting your system is using, use 'sysctl name' (e.g.: 'sysctl
net.ipv4.tcp_rmem'). To change a setting use 'sysctl -w'. To make the setting
permanent add the setting to the file 'sysctl.conf'.
Back to Top
TCP tuning
Like most modern OSes, Linux now does a good job of auto-tuning the TCP
buffers, but the default maximum Linux TCP buffer sizes are still too small.
Here are some example sysctl.conf commands for different types of hosts.
For a host with a 10G NIC, optimized for network paths up to 100ms RTT, add
this to /etc/sysctl.conf
# increase TCP max buffer size settable using setsockopt()
net.core.rmem_max = 16777216
net.core.wmem_max = 16777216
# increase Linux autotuning TCP buffer limit
net.ipv4.tcp_rmem = 4096 87380 16777216
net.ipv4.tcp_wmem = 4096 65536 16777216
# increase the length of the processor input queue
net.core.netdev_max_backlog = 30000
# recommended default congestion control is htcp
net.ipv4.tcp_congestion_control=htcp
# recommended for hosts with jumbo frames enabled
net.ipv4.tcp_mtu_probing=1
Also add this to /etc/rc.local (where N is the number for your 10G NIC):
/sbin/ifconfig ethN txqueuelen 10000
For a host with a 10G NIC optimized for network paths up to 200ms RTT, or a
40G NIC up on paths up to 50ms RTT:
# increase TCP max buffer size settable using setsockopt()
net.core.rmem_max = 67108864
net.core.wmem_max = 67108864
# increase Linux autotuning TCP buffer limit
net.ipv4.tcp_rmem = 4096 87380 67108864
net.ipv4.tcp_wmem = 4096 65536 67108864
# increase the length of the processor input queue
net.core.netdev_max_backlog = 250000
# recommended default congestion control is htcp
net.ipv4.tcp_congestion_control=htcp
# recommended for hosts with jumbo frames enabled
net.ipv4.tcp_mtu_probing=1
Also add this to /etc/rc.local (where N is the number for your 10G NIC):
/sbin/ifconfig ethN txqueuelen 10000
Notes: you should leave net.tcp_mem alone, as the defaults are fine. A number
of performance experts say to also increase net.core.optmem_max to match net.
core.rmem_max and net.core.wmem_max, but we have not found that makes any
difference. Some experts also say to set net.ipv4.tcp_timestamps and net.ipv4.
tcp_sack to 0, as doing that reduces CPU load. We disagree with that
recommendation, as we have observed that the default value of 1 helps in more
cases than it hurts. But if you are extremely CPU bound you might want to
experiment with turning those off.
Linux supports pluggable congestion control algorithms. To get a list of
congestion control algorithms that are available in your kernel (kernal 2.6.
20+), run:
sysctl net.ipv4.tcp_available_congestion_control
If cubic and/or htcp are not listed try the following, as most distributions
include them as loadable kernel modules:
/sbin/modprobe tcp_htcp
/sbin/modprobe tcp_cubic
NOTE: There seem to be bugs in both bic and cubic for a number of versions of
the Linux kernel up to version 2.6.33. We recommend using htcp with older
kernels to be safe. To set the congestion control do:
sysctl -w net.ipv4.tcp_congestion_control=htcp
If you are using Jumbo Frames, we recommend setting tcp_mtu_probing = 1 to
help avoid the problem of MTU black holes. Setting it to 2 sometimes causes
performance problems.
Back to Top
UDP Tuning
The TCP window size actually effects UDP as well on Linux. Be sure to use the
setsockopt() call to increase SO_SNDBUF/SO_RCVBUF to around 4MB if you want
to do UDP streams that are faster than 3-4 Gbps. The optimal data payload
size is the MTU size minus IP and UDP header size, or 1472 for a 1500 byte
MTU and 8972 for a 9000 byte MTU.
Back to Top
NIC Tuning
This can be added to /etc/rc.local to be run at boot time:
# increase txqueuelen for 10G NICS
/sbin/ifconfig ethN txqueuelen 10000
Note that this might have adverse affects for a 10G host sending to a 1G host
or slower.
Back to Top
Virtual Machine Tuning
Back to Top
Last edited: 2012-11-08 16:40:39
http://fasterdata.es.net/host-tuning/linux/