本文转自：https://magedirect.co/how-to-speed-up-magento-2-maximum-performance/
Introduction
In this article, we’ll show you how to run a Magento 2 store at maximum speed. We will review the most popular techniques that you can quickly implement today.
So, let’s imagine the situation:
- You are preparing for big sales and the development team is making final edits in the code.
- You are not happy with the current performance of your store on Magento 2 and want to finally get it over with, and to speed up it.
- You have many products and categories, and the store is slow
If you find your situation with this line, then read on.
1. Preliminary preparation of the test store
Let’s create a real situation, and simulate a large store with a huge amount of products.
Thanks to the Magento 2 developers, they took care of the fact that we were able to generate a set of test products. To do this, Magento 2 has a data generator that allows you to generate a large amount of data (products, websites, categories, customers, order, etc)
More detailed instructions can be found in the official documentation:
http://devdocs.magento.com/guides/v2.2/config-guide/cli/config-cli-subcommands-perf-data.html
So we begin the fun:
Install the fresh Magento 2 on the server and load the test data into the Magento.
In this example, we will use one server for the application and the database.
We use dedicated server – DELL PowerEdge ™ R730 DX151 with the following characteristics:
CPU(s): 16
Model name: Intel(R) Xeon(R) CPU E5-2620 v4 @ 2.10GHz
CPU MHz: 1200.281
L3 cache: 20480K
Memory: 64 GB
For speeding up the highload store, we will install the following software which is recommended by Magento Team:
- Mysql
- Nginx
- php7-fpm
- Redis
- Varnish
Next, we will configure Cache Settings:
We will use the Varnish as Caching Application
Let’s enable Varnish in Magento 2 admin panel
Stores -> Configuration -> Advanced -> System -> Full Page Cache and select next settings:
Then we will enable Redis as default cache (default array) and the full page cache.
For default settings add next records in your app/etc/env.php
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
|
array (
'frontend' =>
array (
'default' =>
array (
'backend' => 'Cm_Cache_Backend_Redis',
'backend_options' =>
array (
'server' => '127.0.0.1',
'port' => '6379',
'persistent' => '',
'database' => '0',
'force_standalone' => '0',
'connect_retries' => '1',
'read_timeout' => '10',
'automatic_cleaning_factor' => '0',
'compress_data' => '1',
'compress_tags' => '1',
'compress_threshold' => '20480',
'compression_lib' => 'gzip',
),
),
'page_cache' =>
array (
'backend' => 'Cm_Cache_Backend_Redis',
'backend_options' =>
array (
'server' => '127.0.0.1',
'port' => '6379',
'persistent' => '',
'database' => '1',
'force_standalone' => '0',
'connect_retries' => '1',
'read_timeout' => '10',
'automatic_cleaning_factor' => '0',
'compress_data' => '0',
'compress_tags' => '1',
'compress_threshold' => '20480',
'compression_lib' => 'gzip',
),
),
),
),
|
2. Generate test data for our store
We will use generated data from Magento 2. More details you can find here:
http://devdocs.magento.com/guides/v2.2/config-guide/cli/config-cli-subcommands-perf-data.html
We use so-called “Large profile”
It is:
| Name | Value |
| website | 5 |
| store group | 5 |
| store views | 5 |
| simple products | 300,000 |
| configurable_products | 8,000 with 24 options |
| categories | 3000 |
| categories_nesting_level | 5 |
| catalog_price_rules | 20 |
| catalog_target_rules | 5 |
| cart_price_rules | 20 |
| cart_price_rules_floor | 2 |
| customers | 5000 |
| tax rates | 40,000 |
| orders | 100,000 |
Start the generation with the command:
|
1
|
bin/magento setup:perf:generate-fixtures /var/www/html/magento2/setup/performance-toolkit/profiles/ce/large.xml
|
The execution of this script takes time. If you do a similar job, we advise you to set enough memory for PHP or only personally to the given script with the help of the directive “memory_limit”. For example:
|
1
|
php -d="memory_limit=-1" setup:perf:generate-fixtures /var/www/html/magento2/setup/performance-toolkit/profiles/ce/large.xml
|
In this example, we set up unlimited memory for php processes.
After the script is executed in our store, we load all the test data we need to check the performance. Let’s open some test category.
Cool! The store is filled with test data. Check the administrative panel.
Products:
Orders:
A lot of categories:
Customers:
3. Let’s setup Magento for high-load
Now we will show you the most priority and simplest actions to speed up your store and to turn into a high-performance mode.
Step 1. Enable production mode
Let’s enable Magento 2 in the combat mode so that your Magento works in high-efficiency mode. For this, you need to enable production mode.
Magento 2 has different modes:
- default
- developer
- production
More details about modes you can check here: http://devdocs.magento.com/guides/v2.0/config-guide/bootstrap/magento-modes.html
Run the following command:
|
1
|
magento deploy:mode:set production
|
Check:
|
1
|
magento deploy:mode:show
|
You should see the next output:
Current application mode: production. (Note: Environment variables may override this value.)
Then run the command
|
1
|
php bin/magento setup:di:compile
|
This will then re-compile the files. Lastly:
|
1
|
php bin/magento setup:static-content:deploy
|
If Magento has enabled maintenance mode – disable maintenance mode by the command:
|
1
|
./bin/magento maintenance:disable
|
Step 2. Enable Cache
We enable all cache types in the administrative panel in the section System -> Cache Management
Step 3. Flat Catalog
Let’s Enable Flat Catalog. With this setting, Magento 2 will make less complex MySQL queries in the database.
Don’t forget to clear configuration cache after this action.
Step 4. Merge & Minify CSS and JS Files
Further, we include the minification and Merge in Store -> Configuration -> Developer
Enable Merge CSS and JS Files, Minify CSS & JS. It will speed up the loading of CSS and JS
Step 5. Tune Server Configuration
Let’s show the most needed configurations for a store with a high load:
Mysql (config file: my.cnf)
The max_allowed_packet directive is chosen empirically to avoid the error “General error: 2006 MySQL server has gone away”. By studying the MySQL logs, you will be able to determine what size of packages your database needs.
For this amount of data we will use:
|
1
|
max_allowed_packet = 32M
|
The directives tmp_table_size and max_heap_table_size are chosen experimentally. With the default value of Magento, many temporary tables will be created. It is defined using the command:
|
1
|
show global status LIKE 'Created_tmp_disk_tables';
|
For this demo store we use:
|
1
2
|
tmp_table_size = 128M
max_heap_table_size = 128M
|
The innodb_buffer_pool_size directive is set relative to the amount of RAM. The recommended value is 70-80%.
For our server this value is:
|
1
|
innodb_buffer_pool_size = 48G
|
PHP (config file .php.ini)
The RAM limit must be at least 1G
|
1
|
memory_limit=1G
|
For cron jobs we personally set at least 4GB
|
1
2
3
4
5
|
* * * * * /usr/bin/php -d=”memory_limit=”4000M” /var/www/magento2/bin/magento cron:run | grep -v "Ran jobs by schedule" >> /var/www/magento2/var/log/magento.cron.log
* * * * * /usr/bin/php -d=”memory_limit=”4000M /var/www/magento2/update/cron.php >> /var/www/magento2/var/log/update.cron.log
* * * * * /usr/bin/php -d=”memory_limit=”4000M /var/www/magento2/bin/magento setup:cron:run >> /var/www/magento2/var/log/setup.cron.log
|
Note:
– Reindex requires more RAM on so many products, if 4GB is not enough for you – try using the bigger value.
– reindex on so many products is performed for a long time, so it’s better to set the reindex mode “update on save”
Step 6. Enable Gzip Compression
Turn on gzip compression in the nginx config:
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
|
gzip on;
gzip_disable "msie6";
gzip_comp_level 6;
gzip_min_length 1100;
gzip_buffers 16 8k;
gzip_proxied any;
gzip_types
text/plain
text/css
text/js
text/xml
text/javascript
application/javascript
application/x-javascript
application/json
application/xml
application/xml+rss
image/svg+xml;
gzip_vary on;
|
Step 7. Optimize Images.
There are different strategies for optimizing images. You can use third-party extensions or optimize before upload images to Magento. In this article, we use a simple script that uses open-source Linux libraries.
Just install next software:
|
1
2
3
|
apt-get install optipng
apt-get install jpegoptim
apt-get install gifsicle
|
|
1
2
3
4
5
6
7
|
#!/bin/bash
find ./media -iname '*.gif' -exec sh -c 'gifsicle -b -O3 -o "{}.out" "{}"; mv "{}.out" "{}"' ;
find ./media -iname '*.png' -exec optipng -o5 -keep -preserve '{}' ;
find ./media -type f -iname '*jpg' -exec sh -c 'jpegtran -outfile "{}.out" -optimize "{}"; mv "{}.out" "{}"' ;
|
4. Let’s make final performance tests
Preliminary information:
1) We made a basic tuning Magento 2. It can be done even better personally under the store.
2) The pages for which we do Magento site speed test, we open in the browser before. That would get the page in the Varnish cache, it said “warmed up.” Because on this demo site we do not have a module for cache warmer and there are no more users who “warm up” the site.
3) In this article, we will not discuss the optimization of render-blocking JavaScript and CSS and JS minifications.
We think that a good mark for the store is 70+. A real store can have many different external scripts (chat rooms, analytics, marketing tools, etc). They all influence Google PageSpeed Insight.
Testing Magento speed (site load time) is necessary. Speed is an important metric in SEO, and also affects the user experience.
When a Googlebot indexes a site, it does not see the PageSpeed indicator, but only knows the speed itself.
Did you know that Google PageSpeed Insights does not measure the speed of your site? Read more detailed:
Google PageSpeed Insights does not measure the speed of the site.
To measure the download speed, you can use GTMetrix or Pingdom Tools.
From our point of view, there is no point in scrolling for Google PageSpeed indicators. The main speed of the store, as this affects the behavioral factor, it is very important for SEO.
Let’s look at the results of Google PageSpeed
1) Google PageSpeed Insight
Category page:
Product Page:
Configurable Product
2) Pingdom
Location: Stockholm, Sweden
Speed of category page:
Speed of product page (configurable product):
5. Additional steps that you can use for performance optimization.
- CDN
- Cache Warmer
- Nginx professional tuning
- Docker Container & Kubernetes
- image optimization
All these methods we will consider in our next articles and we will reveal each method of Magento Speed Optimization in more details.
Thank you for your attention. Hope that now you understand how to speed up Magento site. Keep up with our new performance workshops!
Valentyn KubrakChief Operating Officer