DS220+ upgrade to 18GB.

[WST16]

After my initial upgrade to 6GB with a 4GB Crucial memory that lasted for more than 7 months with no problems, I've gone mad and upgraded the 220+ to 18GB

It's been about a week now and no problems so far. I've never received any warnings even during the first boot after installation, it was immediately recognized.
I've ran the memory test and it passed (took more than 6 hours to complete).

Running DSM 6.2.4-25556 and vDSM 7.0

The model is Crucial CT16G4SFD8266.

 

[fredbert]

Was there a need or was this more ‘because I can’?

 

[WST16]

Running vDSM 7 connected to a VPN and running Emby IPTV, download station and Docker on it. On DSM, I have our main Emby instance and my Docker containers that keep on growing in addition to the usual Synology packages. It was not absolutely necessary but with everything going on it, it will surely make a difference

The 1019+ is no longer my main NAS as I've set it up and allocated it to a project and I don't want to mess with it. So the 220+ is kind of my main unit now until I upgrade it. I'd like to go with something like the 1621+ but wish for transcoding, so I might wait for the next release wave and see what comes. I'll then try to consolidate some of the low end DiskStations (120j, 118 and the 216+ on it) and find another location/use for them.

 

[fredbert]

Running vDSM 7 connected to a VPN and running Emby IPTV, download station and Docker on it

I wondered if VMM was a reason as allocating RAM is one of the two things. Only thing being that a dual core CPU may object if overly tasked, I assume, unless assignment of cores is not exclusive.

 

[WST16]

Of course you're right about the CPU limitation. So far it’s manageable. But look at the bright side, there's an argument there for a reason to upgrade

 

[fredbert]

But look at the bright side, there's an argument there for a reason to upgrade

Sneaky! Start drafting that business case

 

[equinox]

After my initial upgrade to 6GB with a 4GB Crucial memory that lasted for more than 7 months with no problems, I've gone mad and upgraded the 220+ to 18GB

It's been about a week now and no problems so far. I've never received any warnings even during the first boot after installation, it was immediately recognized.
I've ran the memory test and it passed (took more than 6 hours to complete).

Running DSM 6.2.4-25556 and vDSM 7.0

The model is Crucial CT16G4SFD8266.

I am thinking to upgrade the same RAM stick on my DS220+.
Have you noticed any issues?

 

[WST16]

So far, so good.

 

[hstriepe]

The added RAM might show up, but the Intel Celeron J4025 can only address 8GB of RAM.
Anything more is not used.
On my 920+ the 8GB run at 15% utilization unless I run VMs. The latter is why added 4GB.

 

[WST16]

I read that too. What I gathered (from various discussions) is that it's Intel's marketing segmentation and that more than 8GB can be used. The screenshots below says that it's working (DSM's perspective at least).

Of course, if one wants to adhere to the manufacturer's exact specs then shouldn't go beyond 8GB for this CPU and choose the 4GB option (for a total of 6GB).

 

[hstriepe]

Did some more reading and saw the same. At least they did not hardware lock it.
Has anybody seen any performance improvement?
Linux will use any available RAM as cache.

 

[WST16]

If you can avoid memory swaps then that's it, that's where the performance lies. If you don't have memory swaps with the standard 2GB to begin with then the gain will be quite limited to caching. If your use case is mostly sequential reading (like media streaming), no VM and no Docker containers, then –to the best of my knowledge– it's useless.
I have no memory swaps now, my containers are extra happy roaming in all that space and vDSM 7 is so fast (really tempting me to upgrade the main instance to DSM 7, but I'm still holding back)

The Achilles' heel (for the 220+) is of course the two cores (especially if you'll be running a VM).

YMMV, proceed at your own risk and all the meaningless fine print apply here

 

[equinox]

After installed Crucial CT16G4SFD8266 on my DS220+, I can see 18G RAM at Info page. However, after running a couple of minutes, everything is frozen. And received a notification that showing "checksum error". The volumn was degraded and became ReadOnly. The only thing I can do was to delete the volumn and recreate it again.

 

[Telos]

after running a couple of minutes, everything is frozen. And received a notification that showing "checksum error"

Did you run Synology's memory check prior to commissioning the stick?

 

[equinox]

No, I didn't. I didn't find Synology assistant from the package center. I am using dsm7.

 

[Rusty]

No, I didn't. I didn't find Synology assistant from the package center. I am using dsm7.

Synology assistant is not in the package center. You can download it from the Download center page at synology.com

 

[WST16]

Sorry to hear that. Could it be that it moved and popped out a little because it did not go all the way in.
Make sure it clicks when you install it, and if you can get to DSM, SSH and run a memory test immediately.

 

[equinox]

I did push it all the way in. I can see the ram size is 18gb, which means DSM was recognized it. However, after a few clicks, DSM frozen. I can't access mailplus remotely, which means the mail services was also frozen.

I have returned the ram stick and going to try a different one. Samsung 16gb M471A2K43CB1-CTD

 

[WST16]

I keep reading good things about Samsung's memory, so hopefully it'll work. Good luck with that.

Thinking about it now, the only difference between my setup and yours is that I'm still running DSM 6 (and vDSM) while you're on DSM 7. So who knows, maybe when I upgrade to DSM 7 the whole thing will come crashing down
If it happens then Synology is really becoming a PIA with their all new restrictions (disks, USB ports and maybe memory)

 

[equinox]

One possibility is the RAM stick is a faulty. I didn't test it before using.

Another possibility is DS220+ may use different onboard RAM chip based on manufactory date.