Buffalo LS220D 2-Drive NAS Review

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

Terry Kennedy

Well-Known Member
Jun 25, 2015
1,142
594
113
New York City
www.glaver.org
Our Buffalo LS220D 2-drive NAS review - this is an extremely low cost way to get two hard drives online for basic home/ SMB NAS applications.
Something looks fishy in the Buffalo LS220D 2 Drive NAS – Performance Tests chart. It looks like the "File Copy to NAS" and "HD Video Record" are seeing extreme client-side caching, as the reported values exceed 20Gbit/sec on a GigE link. While all of the other reported results on that graph appear to be within the range of possibility, the caching makes me suspicious of the accuracy of all of those results.

When I benchmark a system, I generally use IOzone, at least for local performance characterization before I move to testing with remote clients. A single run can take several days, due to the huge file sizes needed to eliminate the effect of the buffer cache and any processor caches. In the attached image, you can see the effects of the caches (mostly a 48GB buffer cache) until things level out around 500 MByte/second, which is the actual throughput to the disks. This is from a RAIDzilla II, 6 years ago. Current RAIDzilla 2.5 systems are substantially faster.

 

manxam

Active Member
Jul 25, 2015
234
50
28
May I suggest that you be a little more critical in your testing? I would be extremely underwhelmed with this compared to a equivalent prices QNAP or Synology as they have a) web based managers and b) USB 3.0

While I don't think that this is a BAD unit, I feel that you rated it way too high. This should be a 6.5/10 device and not an 8+ out of 10.

Looking forward to the other NAS reviews though.
 

William

Well-Known Member
May 7, 2015
789
252
63
66
Hey guys.

I wonder if there is something up with Intel NAS Benchmark. When starting these reviews out I checked around and saw that many have used this one so that is what I ended up deciding to use. This was actually run several times and all results seemed to be very close to the ones I posted.
I have finished up benches on a Synology unit which seems to be more inline with what I would expect now.

Another benchmark that is run by just about all reviewers is PCMark 8 Storage Tests which I think might be a better idea now to move to that before getting to far along with these.

I will say this also about scoring. I have been running two Buffalo Terra Stations here for many years without one problem, yeah they are slow compared to new units but they are solid units that work well for what they are. Like many of you here I also work with much higher end systems and devices that we might compare things two and give much lower scores for somethings because of this but for people that might end up being end users for a product like this its a good unit that's fast setting up and performs well enough for this market segment.

Here are the screens of the final benchmark runs.



 

William

Well-Known Member
May 7, 2015
789
252
63
66
Ok I was really scratching my head on this problem. I ran this bench on several different machines today and saw the same results.

Turns out the problem is not the bench but how much memory you have installed on the machine running this bench. If more than 2GB of memory installed than it caches up two benchmarks and that gives much higher results than expected. What everyone does is to not select HD Video Record and File Copy To NAS as these get cached up.

Thank you for bringing this to my attention.

I found this information here.
EonNAS 1100 NAS Network Storage Server | EonNAS 1100,EN1100MC-0032,886763000156,NAS,Network Storage,Server,Benchmarks,Performance,Review,Bruce Normann,EonNAS 1100 NAS Network Storage Server EN1100MC-0032 Benchmark Performance Review by Bruce Normann