performance optimizations.txt

Optimizing Linux
Disable THP

https://www.pingcap.com/blog/transparent-huge-pages-why-we-disable-it-for-databases/

Tuned-ADM

https://manpages.ubuntu.com/manpages/jammy/man7/tuned-profiles.7.html

Change Init Level to 3 **(ONLY FOR VPS's HOSTING WORDPRESS SITES, this disables the Desktop GUI)**

Set your init level to 3 if the default is set to 5. Use the command “init 3” to set the init level to 3, then reboot for the changes to take effect. You can click the link below for instructions on how to do so.

https://www.cyberciti.biz/howto/question/linux/changing-run-levels-3-5.php

More info on that here

This link is how to do the reverse, changing 5 to 3 if the “init 3” command does not work. If your run level is set to 5, change it to 3.
Optimizing Linux Github

https://github.com/sn99/Optimizing-linux

Great repo for how to optimize linux performance.

Stacer

https://www.site24x7.com/learn/linux/linux-performance-optimization.html

Stacer is a tool used for optimizing Linux that deletes unwanted cache and sorting processes by looking at memory usage and CPU. Stacer reveals how services and processes are performing and which user is executing the processes. 

Performance Optimization Tools and Techniques

https://linuxconfig.org/linux-performance-optimization-tools-and-techniques

Enable Z-Swap
https://blog.jacobstoner.com/zswap-improve-memory-performance-on-linux/
---------------------
sudo perl -pi -e 's/^(GRUB_CMDLINE_LINUX_DEFAULT.*)"$/$1 zswap.enabled=1 

zswap.compressor=lz4 zswap.zpool=z3fold zswap.max_pool_percent=25"/' /etc/default/grub

sudo update-grub

sudo echo -e 'lz4\nlz4_compress\nz3fold' >> /etc/initramfs-tools/modules

sudo update-initramfs -u

sudo reboot
---------------------
Tune the Linux Filesystem for increased performance
Foss Linux Article

https://www.fosslinux.com/111937/tips-and-tricks-for-optimizing-linux-device-performance.htm

Semi-suface level, but slightly different info than the linux.com article, best to use both of the many listed techniques together.
Use a Custom Kernel

https://www.fosslinux.com/111937/tips-and-tricks-for-optimizing-linux-device-performance.htm

Disable Unnecessary Linux Services

To list all services, you can use this command:

systemctl list-unit-files --type=service

Then, to disable a service, use:

sudo systemctl disable SERVICE_NAME

Jumbo Frames
Enabling Jumbo Frames:

●	Check if your network adapter supports Jumbo Frames: ethtool eth0

●	Enable Jumbo Frames by setting a larger MTU (Maximum Transmission Unit), typically 9000 bytes for Ethernet: ifconfig eth0 mtu 9000 or ip link set dev eth0 mtu 9000

●	Verify the changes: ifconfig eth0 or ip addr show eth0

●	Remember to configure Jumbo Frames on all devices in your network to avoid packet fragmentation.

Jumbo Frames can help reduce CPU load by decreasing the number of packets processed, but it's crucial to ensure that all network equipment supports them to avoid potential network issues. Contact your hosting provider to ensure enabling this feature is supported.

TCP BBR
Increase Linux Internet speed with TCP BBR congestion control

https://www.cyberciti.biz/cloud-computing/increase-your-linux-server-internet-speed-with-tcp-bbr-congestion-control/

https://www.techrepublic.com/article/how-to-enable-tcp-bbr-to-improve-network-speed-on-linux/

https://www.linuxbabe.com/ubuntu/enable-google-tcp-bbr-ubuntu

https://github.com/systemd/systemd/issues/9725#issuecomment-413369212

https://djangocas.dev/blog/huge-improve-network-performance-by-change-tcp-congestion-control-to-bbr/

Linux Network Performance Parameters

https://github.com/leandromoreira/linux-network-performance-parameters

Intel’s VPS Wordpress Tuning guide for Xeon Systems

https://www.intel.com/content/www/us/en/developer/articles/guide/wordpress-tuning-guide-on-xeon-systems.html

Medium Article

https://medium.com/@a.b.t./here-are-some-possibly-useful-tweaks-for-steamos-on-the-steam-deck-fcb6b571b577#6c57
Switching over to the “performance” CPU governor at start-up

(paste this into your terminal as one giant code block, do not run the commands individually)

-------------------------------------------

cat << EOF | sudo tee /etc/systemd/system/cpu_performance.service
[Unit]
Description=CPU performance governor
[Service]
Type=oneshot
ExecStart=/usr/bin/cpupower frequency-set -g performance
[Install]
WantedBy=multi-user.target
EOF
sudo systemctl daemon-reload
sudo systemctl enable cpu_performance.service

-------------------------------------------

To undo or revert this change, run: 

sudo systemctl disable cpu_performance.service
MGLRU (Multi-Gen. Least Recently Used)

This is one of the most recent additions to the upstream Linux kernel, literally just merged for the 6.1 LTS (Long-Term Support). However, the upstream kernel doesn’t enable this functionality by default, yet.

(paste this into your terminal as one giant code block, do not run the commands individually)

To enable this feature, run the terminal command:

--------------------------------

cat << EOF | sudo tee /etc/tmpfiles.d/mglru.conf
w /sys/kernel/mm/lru_gen/enabled - - - - 7
w /sys/kernel/mm/lru_gen/min_ttl_ms - - - - 0
EOF
--------------------------------
Reversion command:

sudo rm /etc/tmpfiles.d/mglru.conf
Unlocking the memory lock
By default, the Linux kernel sets this particular value to just 64KiB, which is the maximum amount of memory the kernel will lock within a single operation. This is an old recommendation. 

This command will set this value to 2 GB, adjust as needed depending on your server’s RAM capacity:

(paste this into your terminal as one giant code block, do not run the commands individually)
-------------------
cat << EOF | sudo tee /etc/security/limits.d/memlock.conf
* hard memlock 2147484
* soft memlock 2147484
EOF
-------------------
This site will help you convert the values so you can replace the above value “2147484” with an appropriate value for your VPS’s memory.

http://www.csgnetwork.com/memconv.html

To revert the change, run:

sudo rm /etc/security/limits.d/memlock.conf

Changing the I/O (Input/Output) Scheduler
We are going to permanently swap the I/O-scheduler to Kyber with the following block (note that this will work with all cases, meaning NVMe SSD, eMMC & microSD):

(paste this into your terminal as one giant code block, do not run the commands individually)
------------------

cat << EOF | sudo tee /etc/udev/rules.d/64-ioschedulers.rules
ACTION=="add|change", KERNEL=="nvme[0-9]*", ATTR{queue/scheduler}="kyber"
ACTION=="add|change", KERNEL=="sd[a-z]|mmcblk[0-9]*", ATTR{queue/rotational}=="0", ATTR{queue/scheduler}="kyber"
EOF
------------------
To undo/revert this change:

sudo rm /etc/udev/rules.d/64-ioschedulers.rules

TCP Network Stack Optimization
TCP Keepalive 
Adjusts how long TCP should wait before sending keepalive messages to confirm that an idle connection is still active. This can help in managing connection overloads.

(Adjust command values as needed)

sysctl -w net.ipv4.tcp_keepalive_time=600
sysctl -w net.ipv4.tcp_keepalive_probes=5
sysctl -w net.ipv4.tcp_keepalive_intvl=60

TCP Window Scaling 
Enables TCP window size adjustments, allowing for larger buffers, which can improve data throughput for connections.

(Adjust command value as needed)

sysctl -w net.ipv4.tcp_window_scaling=1
TCP SACK 
The Selective Acknowledgment (SACK) feature allows devices to inform senders about lost packets, facilitating more efficient retransmissions.

(Adjust command value as needed)

sysctl -w net.ipv4.tcp_sack=1
TCP FIN Timeout 
Adjusts the timeout to close a connection, freeing up resources sooner.

(Adjust command value as needed)

sysctl -w net.ipv4.tcp_fin_timeout=30
Increase TCP Max Syn Backlog 
Determines the number of incoming connections that can be queued for handling.

(Adjust command value as needed)

sysctl -w net.ipv4.tcp_max_syn_backlog=2048
Increase TCP Max Tw Buckets 
Controls the maximum number of TCP sockets in TIME_WAIT state, managing memory use.

(Adjust command value as needed)

sysctl -w net.ipv4.tcp_max_tw_buckets=400000
TCP Fast Open 
Allows data to be sent during the initial SYN handshake, reducing latency for connections.

(Adjust command value as needed)

sysctl -w net.ipv4.tcp_fastopen=3

Reversion Commands
To revert TCP settings, you'll need to know the default values, which might vary based on your Linux distribution and version. Here are commands to set typical default values:

sysctl -w net.ipv4.tcp_keepalive_time=7200
sysctl -w net.ipv4.tcp_keepalive_probes=9
sysctl -w net.ipv4.tcp_keepalive_intvl=75
sysctl -w net.ipv4.tcp_window_scaling=1
sysctl -w net.ipv4.tcp_sack=1
sysctl -w net.ipv4.tcp_fin_timeout=60
sysctl -w net.ipv4.tcp_max_syn_backlog=128
sysctl -w net.ipv4.tcp_max_tw_buckets=5000
sysctl -w net.ipv4.tcp_fastopen=0

Enable Asynchronous I/O (AIO)
Enable AIO to allow multiple I/O operations to execute concurrently, improving performance for database servers like MySQL (WORDPRESS AND WEBSERVERS ONLY)

innodb_use_native_aio = 1

Offload Processing 
Enable offloading features on your network interface to reduce CPU load

ethtool -K eth0 gro on
ethtool -K eth0 gso on
ethtool -K eth0 tso on

Increase Open File Limits
Ensure your system can handle a large number of open files

echo "fs.file-max = 500000" >> /etc/sysctl.conf
ulimit -n 500000

Enable Writeback Caching for Disk I/O
Improve write performance by enabling writeback caching

hdparm -W1 /dev/sda

Store temporary files in memory to speed up read/write operations

mount -t tmpfs -o size=512M tmpfs /tmp

Optimize Memory Usage
Use zram to compress RAM data to increase the amount of available memory

apt-get install zram-config

Optimize Virtual Memory Settings
Adjust virtual memory settings for better performance

sysctl -w vm.dirty_background_ratio=5
sysctl -w vm.dirty_ratio=10

Use CPU Isolation
Isolate CPU cores for critical applications to ensure they have dedicated resources

isolcpus=1,2

Configure HugeTLB Pages
For applications requiring large memory pages, configure HugeTLB to reduce TLB misses

echo 'vm.nr_hugepages=128' >> /etc/sysctl.conf

Optimize DNS Resolution
Use a Local DNS Cache
Install and configure a local DNS caching server like dnsmasq

apt-get install dnsmasq

Adjust Out-of-Memory (OOM) Killer Settings

Prevent the OOM killer from terminating critical processes by adjusting oom_score_adj

echo '-1000' > /proc/$(pidof critical_process)/oom_score_adj

Use RAM Disks for High-Performance Temporary Storage
Mount a portion of your RAM as a disk for extremely fast I/O operations

mount -t tmpfs -o size=1G tmpfs /mnt/ramdisk

Optimize Swap Settings
Use zswap

Enable and configure zswap for better swap performance
echo 1 > /sys/module/zswap/parameters/enabled
echo lz4 > /sys/module/zswap/parameters/compressor
echo 20 > /sys/module/zswap/parameters/max_pool_percent

Optimize Boot Parameters
Modify boot parameters to improve performance (ONLY FOR THE GRUB BOOTLOADER, CONFIRM WHAT BOOTLOADER YOU ARE USING BEFORE RUNNING THIS).

GRUB_CMDLINE_LINUX_DEFAULT="quiet splash noapic noacpi nosplash irqpoll"

Reduce Kernel Latency
Enable kernel preemption and configure low-latency settings

echo 1 > /sys/kernel/debug/sched_features

Enable and Configure HugeTLBfs
Use HugeTLBfs to allocate large pages for applications that can benefit from reduced TLB misses

mount -t hugetlbfs none /dev/hugepages

Configure Network Bonding
Combine multiple network interfaces for increased bandwidth and redundancy

apt-get install ifenslave
echo "bonding" >> /etc/modules

Fine-Tune TCP SYN Cookies
Enable SYN cookies to protect against SYN flood attacks while allowing legitimate traffic.

sysctl -w net.ipv4.tcp_syncookies=1

Optimize ARP Cache Settings
Tune ARP cache settings to handle large numbers of IP addresses efficiently

sysctl -w net.ipv4.neigh.default.gc_thresh1=1024
sysctl -w net.ipv4.neigh.default.gc_thresh2=2048
sysctl -w net.ipv4.neigh.default.gc_thresh3=4096

Enable IP Spoofing Protection
Configure settings to prevent IP spoofing

sysctl -w net.ipv4.conf.all.rp_filter=1
sysctl -w net.ipv4.conf.default.rp_filter=1

Optimize Kernel Shared Memory Parameters
Adjust shared memory settings to improve performance for applications that use IPC (Inter-Process Communication)

sysctl -w kernel.shmmax=68719476736
sysctl -w kernel.shmall=4294967296

Adjust VM Overcommit Settings
Fine-tune memory overcommit behavior for better memory management

sysctl -w vm.overcommit_memory=1
sysctl -w vm.overcommit_ratio=50

Enable and configure query caching in MySQL/MariaDB for faster query responses

query_cache_size = 64M
query_cache_type = ON

Configure TCP Timestamps
Enable or disable TCP timestamps based on your workload to improve performance

sysctl -w net.ipv4.tcp_timestamps=1

Adjust UDP Buffer Sizes
Increase UDP buffer sizes for applications that use UDP heavily

sysctl -w net.core.rmem_max=26214400
sysctl -w net.core.wmem_max=26214400

Optimize nf_conntrack Settings
Adjust connection tracking settings for high-volume traffic

sysctl -w net.netfilter.nf_conntrack_max=131072
sysctl -w net.netfilter.nf_conntrack_tcp_timeout_established=1200

Enable Disk Write Caching
Enable write caching on disks to improve write performance

hdparm -W1 /dev/sda

Tune Systemd for Performance
Reduce systemd's default timeout values for faster boot and shutdown

DefaultTimeoutStartSec=10s
DefaultTimeoutStopSec=10s

Optimize NTP Synchronization
Ensure accurate time synchronization with minimal overhead using chrony instead of ntpd

apt-get install chrony

Enable and Configure Network Interface Offloading
Enable offloading features on network interfaces to reduce CPU usage

ethtool -K eth0 gro on
ethtool -K eth0 gso on
ethtool -K eth0 tso on

Implement Kernel Same-Page Merging (KSM)
Enable KSM to merge identical memory pages between virtual machines

echo 1 > /sys/kernel/mm/ksm/run

Use High-Resolution Timers (ONLY FOR THE GRUB BOOTLOADER, CONFIRM WHAT BOOTLOADER YOUR DEVICE IS USING)
Enable high-resolution timers for applications that require precise timing

echo "highres=on" >> /etc/default/grub
update-grub

Optimize sysctl Parameters for Network Performance

Adjust various sysctl parameters to enhance network throughput and reduce latency
sysctl -w net.ipv4.tcp_rfc1337=1
sysctl -w net.ipv4.tcp_mtu_probing=1
sysctl -w net.ipv4.tcp_base_mss=536

Implement Disk Quotas for User Management
Enable and configure disk quotas to manage user disk usage efficiently

apt-get install quota
Use Page-Cache Management Tools
Use tools like vmtouch to manage and lock frequently accessed files in memory (ONLY FOR WEBSERVERS/VPS'S)

apt-get install vmtouch
vmtouch -dl /path/to/important/files

Optimize Swapping Behavior
Configure vm.dirty_expire_centisecs and vm.dirty_writeback_centisecs for better swap performance

sysctl -w vm.dirty_expire_centisecs=2000
sysctl -w vm.dirty_writeback_centisecs=1000

Implement Software RAID for Disk Performance
Use software RAID to improve disk performance and redundancy

mdadm --create --verbose /dev/md0 --level=1 --raid-devices=2 /dev/sda1 /dev/sdb1

Optimize Kernel Timer Frequency
Adjust the kernel timer frequency to reduce overhead

echo 1000 > /proc/sys/kernel/timer_frequency

Enable IPv6
Enable and optimize IPv6 for better network performance and future-proofing

sysctl -w net.ipv6.conf.all.disable_ipv6=0

Use xfs_fsr for XFS File System Defragmentation
Defragment XFS file systems to improve performance

xfs_fsr /dev/sda1

Optimize TCP Backlog Settings
Increase TCP backlog settings to handle more simultaneous connections

sysctl -w net.core.somaxconn=1024
sysctl -w net.core.netdev_max_backlog=5000

Use mdadm for Software RAID Management
Manage software RAID arrays with mdadm

mdadm --create --verbose /dev/md0 --level=1 --raid-devices=2 /dev/sda1 /dev/sdb1

Implement TCP MSS Clamping
Adjust TCP MSS (Maximum Segment Size) to avoid fragmentation

iptables -t mangle -A FORWARD -p tcp --tcp-flags SYN,RST SYN -j TCPMSS --clamp-mss-to-pmtu

Optimize TCP SYN Flood Protection
Configure TCP SYN flood protection to prevent DoS attacks

sysctl -w net.ipv4.tcp_syncookies=1

Enable and Configure dnsmasq for DNS Caching
Use dnsmasq to cache DNS queries and reduce lookup times

apt-get install dnsmasq
systemctl enable dnsmasq
systemctl start dnsmasq

Enable tcp_tw_recycle for Faster Connection Reuse
Enable tcp_tw_recycle to speed up connection reuse

sysctl -w net.ipv4.tcp_tw_recycle=1

Use blktrace for Block Layer Tracing
Install and use blktrace to trace block I/O operations

apt-get install blktrace
blktrace -d /dev/sda

Adjust Kernel Panic Settings

Configure the system to automatically reboot after a kernel panic

sysctl -w kernel.panic=10
sysctl -w kernel.panic_on_oops=1

Optimize Inotify Watches
Increase the number of inotify watches to handle more file events

sysctl -w fs.inotify.max_user_watches=524288

Enable TCP Window Scaling
Improve TCP performance by enabling window scaling

sysctl -w net.ipv4.tcp_window_scaling=1

Enable TCP Keepalive Probes
Configure TCP keepalive probes to manage idle connections

sysctl -w net.ipv4.tcp_keepalive_time=300
sysctl -w net.ipv4.tcp_keepalive_intvl=60
sysctl -w net.ipv4.tcp_keepalive_probes=5

Optimize TCP Congestion Control Algorithms
Experiment with different TCP congestion control algorithms to find the best fit for your workload

sysctl -w net.ipv4.tcp_congestion_control=bbr

Use schedtool for Advanced CPU Scheduling
Install and use schedtool to control CPU scheduling policies

apt-get install schedtool
schedtool -R -p 20 -e myapp

Enable Kernel Samepage Merging (KSM)
Use KSM to reduce memory usage by merging identical memory pages

echo 1 > /sys/kernel/mm/ksm/run

Optimize Network Buffer Sizes
Increase network buffer sizes to handle high traffic volumes

sysctl -w net.core.rmem_max=16777216
sysctl -w net.core.wmem_max=16777216

Optimize Kernel Scheduler
Adjust kernel scheduler settings for better performance

echo 1 > /proc/sys/kernel/sched_autogroup_enabled

Adjust rcu Settings for Low-Latency Systems
Tune Read-Copy Update (RCU) settings for low-latency performance

echo 0 > /sys/kernel/rcu_expedited

Optimize Filesystem Journaling
Choose the appropriate journaling mode for your filesystem (e.g., data=writeback for ext4)

mount -o remount,data=writeback /dev/sda1

Enable and Configure BBRv2 TCP Congestion Control
Use BBRv2 for even better TCP congestion control performance

echo "net.ipv4.tcp_congestion_control=bbr" >> /etc/sysctl.conf
sysctl -p

Optimize File System Performance with tune2fs
Use tune2fs to optimize ext4 filesystem parameters

tune2fs -o journal_data_writeback /dev/sda1

Use ethtool to Optimize NIC Settings
Fine-tune NIC settings for better network performance

ethtool -K eth0 tso off

Enable and Configure Network Interface Teaming
Use network interface teaming for load balancing and failover

apt-get install ifenslave

Use irqbalance for Balancing Interrupts
Install and configure irqbalance to distribute interrupts across CPUs

apt-get install irqbalance

Tune Kernel Semaphore Parameters
Adjust semaphore settings for better performance with IPC-heavy applications

sysctl -w kernel.sem="250 32000 100 128"

Configure and Use ethtool for Network Optimization
Use ethtool to optimize network interface settings

ethtool -K eth0 rx off tx off

Enable TCP Small Queues (TSQ)
Optimize the TCP stack by enabling TSQ

sysctl -w net.ipv4.tcp_limit_output_bytes=131072

Enable and Configure EarlyOOM
Use EarlyOOM to handle out-of-memory situations more gracefully

apt-get install earlyoom

Optimize Inode Cache
Adjust inode cache settings to handle large numbers of files

sysctl -w fs.inode-nr
sysctl -w fs.inode-state

Optimize Disk Caching
Adjust vm.dirty_ratio and vm.dirty_background_ratio to control disk caching behavior

sysctl -w vm.dirty_ratio=20
sysctl -w vm.dirty_background_ratio=10

Implement IP Address Aliasing
Use IP address aliasing to manage multiple IP addresses on a single network interface, improving network performance and management

ip addr add 192.168.1.2/24 dev eth0

Enable Asynchronous I/O for Database Servers
Enable asynchronous I/O (AIO) for database servers like MySQL or PostgreSQL to improve disk I/O performance

innodb_use_native_aio=1

Optimize TCP Reassembly Settings
Adjust TCP reassembly settings to handle fragmented packets more efficiently

sysctl -w net.ipv4.tcp_reordering=3

Enable jemalloc for Memory Allocation
Use jemalloc as a memory allocator for applications like Nginx, MySQL, or Redis to improve memory management performance **(ONLY FOR VPS'S)**

LD_PRELOAD=/usr/lib/x86_64-linux-gnu/libjemalloc.so.2

Adjust IO Elevator Algorithms
Switch to a more suitable I/O elevator algorithm like mq-deadline or none for SSDs

echo mq-deadline > /sys/block/sda/queue/scheduler

Use fstrim for SSDs
Periodically run fstrim to inform the SSD of unused blocks, improving performance

fstrim -v /

Optimize TCP Keepalive Settings
Fine-tune TCP keepalive settings to manage idle connections better

sysctl -w net.ipv4.tcp_keepalive_time=120
sysctl -w net.ipv4.tcp_keepalive_intvl=30
sysctl -w net.ipv4.tcp_keepalive_probes=5

Optimize Virtual Memory Overcommit
Adjust virtual memory overcommit settings to better handle memory allocation

sysctl -w vm.overcommit_memory=2
sysctl -w vm.overcommit_ratio=50

Use ethtool for Network Interface Tuning
Adjust network interface settings for better performance

ethtool -K eth0 rx off tx off

Implement BBRv2 TCP Congestion Control
Use BBRv2 for enhanced TCP congestion control performance

sysctl -w net.ipv4.tcp_congestion_control=bbr

Optimize I/O Performance with blk-mq (Multi-Queue Block I/O Queueing Mechanism)
Enable and configure blk-mq for better I/O performance on SSDs

echo "mq-deadline" > /sys/block/sda/queue/scheduler

Use schedtool to Control CPU Scheduling Policies
Install and use schedtool to manage CPU scheduling

apt-get install schedtool
schedtool -R -p 20 -e /usr/bin/myapp

Optimize TCP Retransmission Timeout (RTO)
Adjust TCP RTO settings for better performance

sysctl -w net.ipv4.tcp_retries2=5

Optimize TCP SYN Cookies
Enable TCP SYN cookies to protect against SYN flood attacks

sysctl -w net.ipv4.tcp_syncookies=1

Optimize TCP Window Scaling
Enable TCP window scaling for better performance

sysctl -w net.ipv4.tcp_window_scaling=1

Use cgroups for Resource Management
Implement cgroups to limit, prioritize, and isolate resource usage of processes

cgcreate -g cpu,memory:mygroup
cgset -r cpu.shares=512 mygroup
cgset -r memory.limit_in_bytes=512M mygroup

Optimize TCP Memory Buffers
Adjust TCP memory buffer settings for better performance

sysctl -w net.ipv4.tcp_rmem='4096 87380 16777216'
sysctl -w net.ipv4.tcp_wmem='4096 65536 16777216'

Optimize Kernel Schedulers
Choose the appropriate I/O scheduler for your workload (noop for SSD)

echo "noop" > /sys/block/sda/queue/scheduler

Optimize Kernel Shared Memory Parameters
Adjust shared memory settings for better performance with IPC-heavy applications

sysctl -w kernel.shmmax=68719476736
sysctl -w kernel.shmall=4294967296

Implement ZFS for Advanced File System Features
Use ZFS for better data integrity, snapshots, and performance

apt-get install zfsutils-linux
zpool create mypool /dev/sda

Enable TCP Selective Acknowledgements (SACK)
Optimize TCP performance with SACK

sysctl -w net.ipv4.tcp_sack=1

Enable and Configure zram for Compressed Swap
Use zram to create a compressed block device in RAM

apt-get install zram-config

Implement TCP Timestamping
Enable TCP timestamps for better performance and reliability

sysctl -w net.ipv4.tcp_timestamps=1

Use zpool for ZFS Pool Management
Manage ZFS storage pools with zpool

zpool create mypool /dev/sda

Optimize File System Access Time
Use noatime and nodiratime mount options to reduce disk I/O

mount -o remount,noatime,nodiratime /

Optimize TCP Backlog Size
Increase TCP backlog size to handle more incoming connections

sysctl -w net.core.somaxconn=1024

Optimize Kernel Virtual Memory Settings
Adjust Virtual Memory Settings For Better Performance and Optimize Kernel File Handle Limits
Increase the number of file handles the kernel can manage

sysctl -w fs.file-max=2097152

Enable TCP Low Latency
Configure TCP low latency mode for better responsiveness

sysctl -w net.ipv4.tcp_low_latency=1

Use pm2 for Node.js Process Management **(VPS'S USING NODE.JS ONLY)**
Install and use pm2 to manage Node.js processes
npm install pm2@latest -g
pm2 start app.js

Note: Only applicable if using Node.js in Wordpress

Implement Squid for Caching and Proxying (VPS'S ONLY)
Install and configure Squid for web caching and proxying

apt-get install squid

Optimize TCP Congestion Control for High Latency Networks
Use a suitable TCP congestion control algorithm for high-latency networks

sysctl -w net.ipv4.tcp_congestion_control=htcp

Use cpulimit to Limit CPU Usage of Processes
Install and use cpulimit to control the CPU usage of processes

apt-get install cpulimit
cpulimit -e myapp -l 50

Optimize Disk Write Caching
Enable disk write caching for better write performance

hdparm -W1 /dev/sda

Use noatime and nodiratime Mount Options
Reduce disk I/O by disabling access time updates on filesystems

mount -o remount,noatime,nodiratime /

Adjust Maximum Transmission Unit (MTU) for Network Interfaces
Increase MTU size for better performance on high-speed networks

ip link set dev eth0 mtu 9000

Tune Kernel Preemption Model
Configure the kernel preemption model for low-latency desktop or high-throughput server workloads

echo 0 > /proc/sys/kernel/preempt

Optimize Disk I/O
Increase Read-Ahead Buffer 
Adjust the read-ahead buffer size for disk I/O

blockdev --setra 2048 /dev/sda

Optimize IRQ (Interrupt Request) Balancing
Distribute IRQ load across multiple CPU cores to prevent bottlenecks

apt-get install irqbalance

systemctl enable irqbalance

Use HugePages for Memory-Intensive Applications

echo 'vm.nr_hugepages=128' >> /etc/sysctl.conf

Use Lightweight Logging
Reduce the verbosity of system logs to decrease disk I/O

sed -i 's/^#LogLevel=info/LogLevel=error/' /etc/systemd/journald.conf

Network Queue Management
Queue Length

Increasing the network interface queue length can help handle more packets during traffic bursts, reducing packet drops.
(Adjust value as needed)

ifconfig eth0 txqueuelen 10000

Reversion Command
Reset the transmission queue length to its default value (usually 1000):

ifconfig eth0 txqueuelen 1000

I/O Scheduler
Choosing the right I/O scheduler can improve performance based on your workload characteristics (e.g., SSD vs HDD).
Set I/O Scheduler

For SSDs, noop or deadline might be better.

echo "deadline" > /sys/block/sda/queue/scheduler

Swappiness
The vm.swappiness parameter controls the relative weight given to swapping out runtime memory, versus dropping pages from the system cache.

Adjust Swappiness

sysctl -w vm.swappiness=10

Preventing the superfluous book-keeping of File Access Times
The Linux kernel updates the access time of every file every time it would access it, meaning that even a read-only operation would immediately result in a write operation, because the access time of the file in question had to be updated.
Therefore, let’s permanently switch over to the noatime setting with the following command:

sudo sed -i -e '/home/s/\bdefaults\b/&,noatime/' /etc/fstab

To undo/revert this change:

sudo sed -i -e 's/,noatime//' /etc/fstab

Reversion Commands

File Descriptor Limits
Reverting file descriptor limits involves setting them back to typical defaults:

# System-wide
echo "fs.file-max = 65535" >> /etc/sysctl.conf

# Per-user
echo "* soft nofile 1024" >> /etc/security/limits.conf
echo "* hard nofile 4096" >> /etc/security/limits.conf

I/O Scheduler

To revert the I/O scheduler, you can switch back to the default. For many Linux systems, cfq (Completely Fair Queuing) is a common default for HDDs, and noop or mq-deadline for SSDs:

echo "cfq" > /sys/block/sda/queue/scheduler

Swappiness
The default swappiness value is usually set at 60, which is suitable for most general-use cases:

sysctl -w vm.swappiness=60

Making Reversions Permanent

To ensure these settings are permanent, update the /etc/sysctl.conf file with the reverted values and run:

sysctl -p

This command reloads the configurations from the sysctl.conf file.