# How much of a performance booster is 2GB vs 1GB RAM?

**URL:** https://meta.discourse.org/t/how-much-of-a-performance-booster-is-2gb-vs-1gb-ram/58000
**Category:** Self-hosting
**Tags:** hosting
**Created:** [2월 26, 2017, 11:31오후 UTC](https://meta.discourse.org/t/how-much-of-a-performance-booster-is-2gb-vs-1gb-ram/58000 "2017-02-26T23:31:32Z")
**Posts on this page:** 7
**Page:** 1

<div class="post-metadata">

### Author: ![tophee](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/tophee/32/73406_2.png) [@tophee](https://meta.discourse.org/u/tophee)
#### Post date: [2월 26, 2017, 11:31오후 UTC](https://meta.discourse.org/t/how-much-of-a-performance-booster-is-2gb-vs-1gb-ram/58000/1 "2017-02-26T23:31:32Z")

</div>

I noticed that at leastr some of the pages in my (newly installed) discourse forum are quite slow to load:  
 ![](https://global.discourse-cdn.com/meta/original/3X/6/8/68145902e1417e2a083365e2d38d0ecdfc0f8529.png)  
Currently I’m on a 1GB digital ocean droplet and have a WP site on the same droplet, but with no traffic yet, so I am wondering whether upgrading to 2GB RAM would noticeably improve anything?

---

<div class="post-metadata">

### Author: ![codinghorror](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/codinghorror/32/110067_2.png) [@codinghorror](https://meta.discourse.org/u/codinghorror)
#### Post date: [2월 26, 2017, 11:37오후 UTC](https://meta.discourse.org/t/how-much-of-a-performance-booster-is-2gb-vs-1gb-ram/58000/2 "2017-02-26T23:37:55Z")

</div>

Not much unless you have memory pressure. Memory pressure would come from a large database or many active simultaneous users.

---

<div class="post-metadata">

### Author: ![Lutz](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/lutz/32/65103_2.png) [@Lutz](https://meta.discourse.org/u/Lutz)
#### Post date: [2월 27, 2017, 12:05오전 UTC](https://meta.discourse.org/t/how-much-of-a-performance-booster-is-2gb-vs-1gb-ram/58000/3 "2017-02-27T00:05:13Z")

</div>

This means you have PHP, MySQL/MariaDB, nginx/Apache as reverse proxy and Wordpress, possibly with plugins, on a 1 GB droplet together with Discourse? Or is this all shared or provided by other droplets?

If everything is hosted on a single System, you should really check your memory usage under some load.

---

<div class="post-metadata">

### Author: ![sam](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/sam/32/102149_2.png) [@sam](https://meta.discourse.org/u/sam)
#### Post date: [2월 27, 2017, 12:17오전 UTC](https://meta.discourse.org/t/how-much-of-a-performance-booster-is-2gb-vs-1gb-ram/58000/4 "2017-02-27T00:17:41Z")

</div>

Note, that mini profiler time is showing 14 reqs took 2.5 secs, if you click it expands to add individual reqs

---

<div class="post-metadata">

### Author: ![tophee](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/tophee/32/73406_2.png) [@tophee](https://meta.discourse.org/u/tophee)
#### Post date: [2월 27, 2017, 12:22오전 UTC](https://meta.discourse.org/t/how-much-of-a-performance-booster-is-2gb-vs-1gb-ram/58000/5 "2017-02-27T00:22:01Z")

</div>

> [@Lutz](#):
>
> This means you have PHP, MySQL/MariaDB, nginx/Apache as reverse proxy and Wordpress, possibly with plugins, on a 1 GB droplet together with Discourse? Or is this all shared or provided by other droplets?

Yes, all that on one droplet together with discourse. So here is my memory check.

```
              total used free shared buff/cache available
Mem: 992 665 88 47 238 116
Swap: 2047 501 1546

```

And that is without any load (meaning users/page views).  
I guess that means I do have some “memory pressure”, right?

---

<div class="post-metadata">

### Author: ![mpalmer](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/mpalmer/32/45740_2.png) [@mpalmer](https://meta.discourse.org/u/mpalmer)
#### Post date: [2월 27, 2017, 2:06오전 UTC](https://meta.discourse.org/t/how-much-of-a-performance-booster-is-2gb-vs-1gb-ram/58000/6 "2017-02-27T02:06:50Z")

</div>

> [@tophee](#):
>
> I guess that means I do have some “memory pressure”, right?

Maybe, maybe not. Since you’ve got a reasonable amount of memory available and used by cache, things aren’t as dire as they might appear. Without traffic, it’s nearly impossible to tell whether you’ll have problems. Throw some requests at the system using a load tester, and see what happens. I wrote something recently on [how to tune for memory usage](https://meta.discourse.org/t/optimizing-the-number-of-unicorns-and-buffer-size/51514/5) and the metrics to look at, which will give you a much better idea of how swamped your system is than just looking at `free`.

---

<div class="post-metadata">

### Author: ![tophee](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/tophee/32/73406_2.png) [@tophee](https://meta.discourse.org/u/tophee)
#### Post date: [3월 3, 2017, 11:36오전 UTC](https://meta.discourse.org/t/how-much-of-a-performance-booster-is-2gb-vs-1gb-ram/58000/7 "2017-03-03T11:36:12Z")

</div>

This question has unintentionally resurfaced in another topic and, after some more testing, the answer there is pretty clear: for running Wordpress and Discourse on the same server, 1GB is not enough:

> [@Is 1GB RAM enough for both WordPress and Discourse?](https://meta.discourse.org/t/is-1gb-ram-enough-for-both-wordpress-and-discourse/58141):
>
> I noticed a couple of warnings when rebuilding my container: WARNING: The TCP backlog setting of 511 cannot be enforced because /proc/sys/net/core/somaxconn is set to the lower value of 128. WARNING overcommit\_memory is set to 0! Background save may fail under low memory condition. To fix this issue add ‘vm.overcommit\_memory = 1’ to /etc/sysctl.conf and then reboot or run the command sysctl vm.overcommit\_memory=1’ for this to take effect WARNING you have Transparent Huge Pages (THP) support en…
