Hi, Tony! I think the better way to compare Redis, Memcached and no cache is to look at disk reads. The more complex the content, more impact on the server performance. The network speed is not so important in this case. I would like see you redo this video.
The only factor that should matter here is TTFB (time to first byte) since we’re talking about the server-side response here, and even that is subject to the variability of the network. Not to say the conclusion ultimately isn’t correct, but the data backing it is important (if comparing caching solutions, not just caching vs. non-caching). That said, another major factor for me was the configuration of a cluster of memcached backing a cluster of web-app containers, and for me a few ms difference at large isn’t as significant as setup/configuration overhead. For me memcached seemed simpler and a more obvious choice (considering I didn’t need the extra advanced features and the speed improvements were negligible at the scale of the clusters I’m dealing with), particularly since the sharding is happening client side with no need for slave/master replication on the back-end. So… “better” will definitely depend on the situation! 😃
there are a lot of other stuff faster than redis but redis seams to be an all around solution just small-medium scale work load but its not by the best
Hi Tony, thanks for another great video. Could you please point out a good guide how to setup Redis for WP on vps? I'm playing with Google Cloud with aaPanel and can't for the life of me get it to work. Cheers
hi, how are you measuring speed of redis/memcache on the client-side by comparing request and resource loaded? If it was about TTFB then it was fine but resource size and number of request from client-side doesn't make a good comparison. just curious
The test is a page speed test in order to determine which caching solution is able to deliver a web page faster. I only mentioned the page size and number of requests to show how the sites with W3 Total Cache compared to the non-caching site
Hi Tony, Can you help me sort out all the Apache module and PHP extension options for cache? I'm getting great results with the OPcache PHP extension. Can I use that with the memcached extension? Also, I have mod_cache, mod_cache_disk, and mod_file_cache modules installed. Do I need these as well or do they fight each other? Thanks!
@Tony - Out of interest, do you prefer W3TC for your caching/optimising? It's been a while since I've last touched it, but would be interested to hear your thoughts.
both are geat brother. but if your hosting provider uses litesepped as web server then definitely litespeed cache is better for you. otherwise, use w3 total cache.
with the exception that my server was hacked due to an old Redis server. I guess that's what you get when you build your own and expect apt to auto update it:)
hey Tony - thanks for this and your very informative videos. wanted to ask - im on a 4gb linode, it looks like the recommended server specs are >8cores and > 30gb ram, is redis even an option for me? what would be a good alternative instead?
Hi, Tony! I think the better way to compare Redis, Memcached and no cache is to look at disk reads. The more complex the content, more impact on the server performance. The network speed is not so important in this case. I would like see you redo this video.
That's a good idea Marcos
@@TonyTeachesTech, as always, I wish a good job!
The only factor that should matter here is TTFB (time to first byte) since we’re talking about the server-side response here, and even that is subject to the variability of the network. Not to say the conclusion ultimately isn’t correct, but the data backing it is important (if comparing caching solutions, not just caching vs. non-caching). That said, another major factor for me was the configuration of a cluster of memcached backing a cluster of web-app containers, and for me a few ms difference at large isn’t as significant as setup/configuration overhead.
For me memcached seemed simpler and a more obvious choice (considering I didn’t need the extra advanced features and the speed improvements were negligible at the scale of the clusters I’m dealing with), particularly since the sharding is happening client side with no need for slave/master replication on the back-end. So… “better” will definitely depend on the situation! 😃
I agree Patrick. As with most things in the industry... it depends
Nice Antony!
there are a lot of other stuff faster than redis but redis seams to be an all around solution just small-medium scale work load but its not by the best
Excellent video!
Thank you very much!
Really excellent video. Thanks a lot.
You are welcome!
Hi Tony, thanks for another great video. Could you please point out a good guide how to setup Redis for WP on vps? I'm playing with Google Cloud with aaPanel and can't for the life of me get it to work. Cheers
ua-cam.com/video/jZo97S85gWE/v-deo.html
Thank you
hi, how are you measuring speed of redis/memcache on the client-side by comparing request and resource loaded? If it was about TTFB then it was fine but resource size and number of request from client-side doesn't make a good comparison. just curious
The test is a page speed test in order to determine which caching solution is able to deliver a web page faster. I only mentioned the page size and number of requests to show how the sites with W3 Total Cache compared to the non-caching site
Hi Tony,
Can you help me sort out all the Apache module and PHP extension options for cache?
I'm getting great results with the OPcache PHP extension. Can I use that with the memcached extension? Also, I have mod_cache, mod_cache_disk, and mod_file_cache modules installed. Do I need these as well or do they fight each other?
Thanks!
Sorry Steve to be honest I’m not sure
@Tony - Out of interest, do you prefer W3TC for your caching/optimising? It's been a while since I've last touched it, but would be interested to hear your thoughts.
I do! Use it on most all of my site. There's a few videos on my channel related to W3 ua-cam.com/channels/WPJwoVXJhv0-ucr3pUs1dA.htmlsearch?query=w3
both are geat brother. but if your hosting provider uses litesepped as web server then definitely litespeed cache is better for you. otherwise, use w3 total cache.
100% redis every, single, time.
As a object cache, it also works wonders for Dynamic Content (for the most part) with minimal exclusions required.
with the exception that my server was hacked due to an old Redis server. I guess that's what you get when you build your own and expect apt to auto update it:)
Redis FTW 😃
Ouch 😔
@@laci272 Ooof! Yes; it's definitely something to consider if you're maintaining the servers yourself. Good reminder *checks versions*!
hey Tony - thanks for this and your very informative videos. wanted to ask - im on a 4gb linode, it looks like the recommended server specs are >8cores and > 30gb ram, is redis even an option for me? what would be a good alternative instead?
Where did these recommended specs come from?
Is it possible to use Redis and memcached together ? Any improvement will give ?
so if i had to do everything by myself, why you made this video in the first place? to show your cool warehouse?