WP Hosting Performance Benchmark: Page 3 - Enhance Control Panel

WP Hosting Performance Benchmark

Adrien

Zoinkies Great advice. Thank you.
Are you by any chance Martin, the guy who crushed last year's Review Signal enterprise-level benchmark using GP? I'm curious because you said you participated in Kevin's testing.

Regarding the 7950X, I researched a lot and selected two providers:

1/ Pebblehost in the UK to cover Europe. The backup in case they fail would be, as of date, MechanicWeb.
2 Mechanicweb in Dallas to cover the USA. I did not find any backup yet, but I'm pretty sure soon others will renew their hardware and offer alternatives.

Pebblehot : $270/month
Mechanicweb: $290/month (less RAM but they said RAM prices will be cheaper soon and I should wait for this before adding more memory).

P.S: I would have liked this Community to be private so that I could share more of my business insights only with fellow members. Anyways...

Zoinkies

Adrien I try to keep private, I think it would be better if this community was private for only enhance members who pay then we could be less secretive. I think we should all band together and create a sub market, I.e those of us who offer more expensive products refer down to cheaper businesses etc not like there’s a shortage of customers who need hosting.

Adrien

Zoinkies I'm down for it! If you want to vote for this privacy feature, I posted this request in the past (without much success yet ;-)

https://community.enhance.com/d/92-hide-this-community-from-the-public/3

prasad0889

Adrien Zoinkies @Adam

If its not possible to make the whole community private, it would be better that some part of it is made private and access is given to paying Enhance users!

Thoughts?

Adrien

prasad0889 Excellent idea! This is even better.

AdamM

I have a problem and am wondering what it could be caused by, maybe someone has an idea.
When I use the mysql performance testing plug-in "Post Insert, Query & Delete Benchmark", I have poorer results with dragonfly/redis cache enabled in the LiteSpeed Cache plug-ins.

With dragonfly cache enabled, my scores are around the 8 point mark and times are getting longer. Insert time increases.
With the cache dragonfly turned off, the results are around 9 points.

I have some observations but I don't know how to describe them professionally.
it seems to me that it is a problem of creating data for mysql which goes through dragonfly and it takes more time because it is an additional operation

anyone have any thoughts?

fbaye

Never really used Dragonfly, prefer Relay, however, there are some circumstances where object caching isn't going to improve performance, object caching is usually beneficial in high concurrency environments not single request threads.

Relay is actually slower than directly connecting to MySQL8, object caching isn't designed for single request performance, more for the anility to process more requests CPU efficiently.

It's why the new Cloudlinux AccelerateWP platform has Smart Advice that analyses whether or not a site would benefit from Object Caching.

fbaye

Adding onto this, Redis API's often require more base CPU to operate, and don't work very well in shared environments, that's why most companies don't offer shared object caching. If Enhance in the future offers containerised, or service provided Redis this will change, but I presume you're running Dragonfly outside of Enhance, and allowing all your customers to share it - this will end in tears, trust me.

AdamM

fbaye "this will end in tears, trust me."
What exactly do you have in mind?

fbaye

AdamM Same issue that Relay has running in a shared environment, the database salts etc can be programmatically predicted and you can expose each customers databases. It's why lsmcd was made for Memcached, there's nothing really stopping someone from accessing another customer's data.

AdamM

fbaye
Hmm. So overall it's about security as I thought. So we need dedicated isolated docker redis images for each website.
https://www.reddit.com/r/redis/comments/yrnt6x/securing_redis_for_a_shared_hosting_environment/

fbaye

AdamM Correct, reach Redis instance needs to be unique to the website, and then Dragonfly can be used as a API, but until then you're practically sharing all your customers data between them with no access rules.

Adrien

AdamM Adam I share your frustration about not being able to use Memcached/Redis/DF. If you read the roadmap, there's no mention of it yet. Adam wrote this in another post:
"Our plan is to make Redis a package setting which will enable customers to create a Redis instance within their PHP container. They will be able to connect to it on 127.0.0.1 and it won't be shared or accessible to any other website or container. We will also add the Redis PHP extension as a toggle. The feature is scoped but is not on the roadmap yet."

We have like 6 months to go before the current roadmap is completed if everything goes according to Enhance plans.

So it would be super if they could consider adding Redis feature before the end of 2023 ;-)

Zoinkies

Adrien I think we should have a voting system on the roadmap so the users of Enhance get to choose what is prioritised.

Adam

The items on the roadmap are the things which currently have development time scheduled. We have a much larger list of features which are approved for development but not yet scheduled and this includes redis. It will definitely happen this year, possibly even before some of the existing roadmap items are launched.

AdamM

Adam Very good news, Thx Adam.

Adrien

Adam This is great news. Thank you to everyone on the Enhance Team.

DracoBlue

Adam

Great message. Glad this post was made. I care about a well-optimized server environment. Enhance you will do everything in your power to make the panel very efficient. It is worth changing the default php version to 8.1.

dynamo

Adrien Did you find any good provider for performance? I have been testing Enhance extensively for the past couple of weeks and I am not able to get satisfactory performance out of Enhance.

I compared Runcloud and Enhance with OLS and same PHP version on the same server after reloading OS and the results with Enhance were not satisfactory.

Runcloud + OLS https://ibb.co/NrhrbtK
Enhance + OLS https://ibb.co/D1BrCNf

I thought this might be related to our setup but found this thread and noticed other users getting a little lower performance.

Zoinkies

dynamo Enhance's default settings on OLS aren't editable yet, that'll sort most issues out.

In order of performance:

Bare Metal:
OVH(Sometimes suspends you for clock changes)
Vultr(Does not allow clock changes)

Cloud Providers:
Google C2
Azure C/O
Linode

Avoid Vultr High Frequency, it's EOL this year, I've moved over 3000$ per month from them thanks to a heads up from my account executive, they're being replaced with C/O instead of HF, I believe Cloudways has also EOL Vultr because of this also.

Colo is always the best option, that's what we do, https://media.discordapp.net/attachments/1090988032041156641/1102205785578164234/jcTFZXO_d.png

Our setup on Plesk is above, and that's on our Colo infra with Intel chips.

dynamo

Zoinkies Thanks for the reply, have had bad experience with OVH in the past so we'd like to avoid it. Colo would surely be the best option, but it will limit us due to fund requirement. We will not be able to quickly scale to multiple locations. Thinking of going with Ryzen 7950X chips and I think they should give satisfactory performance. But the question still remains, why is the performance that low with Enhance? I tested with both Apache and OLS here and compared with Runcloud.

« Previous Page Next Page »