Some changes

As you'll see, we've started to add comparisons in the graphs, this time we've only added the Ugreen DXP4800 Plus, which we chose due to it having a 2.5 Gbps and a 10 Gbps Ethernet interface. We will be adding a second NAS to these comparisons, but we simply didn't have time to finish testing it ahead of the NASync iDX6011 Pro deadline. We've also changed around the graphs a little bit, which will hopefully make it easier to read them. Feedback is welcome on these changes, and we'll be making adjustments over time this year to try and present the data in a clearer way than we have in the past.

CrystalDiskMark

RAID 5

RAID 1

RAID 0

SSD RAID 0

Thunderbolt 4 Networking SSD RAID 0

CrystalDiskMark is a simple drive benchmark that gives us a quick and easy baseline for the drive performance. We ran the benchmark using RAID 0, 1, and 5, but we also ran it on a pair of NVMe SSDs in RAID 0 to see if that made any difference. It seems like the 64 GB of RAM in the NASync iDX6011 Pro makes sure there are no bottlenecks for the sequential drive tests, but the random 4K test with low queue depths performed a bit unexpectedly, with RAID 5 and the SSDs being the slower ones, oddly enough. We also ran a test using Thunderbolt networking, but the results are not quite what we expected. Sequential write speeds are as you can see, very fast, at twice the speed compared to 10 Gbps Ethernet, but the read speeds were actually slower than over Ethernet. The random 4K results are also terrible, for lack of a better word, considering we tested against two SSDs in RAID 0. We should point out that Windows 11 reports the network interface as a "USB4 P2P Network Adapter," but that shouldn't make any difference, as the Acer laptop used is all Intel based.

iPerf3

To make sure that the networking interface is working as intended, we ran a quick test with iPerf3. Nothing really stood out here, as we expected.

NAS Performance Tester

RAID 5

RAID 1

RAID 0

SSD RAID 0

This test is simply called the NAS performance tester, and this is where we normally get to the first proper drive benchmarks. The NAS performance tester is configured to use an 8 GB dataset—which is the largest dataset available in the benchmark—that is deleted and re-created during each iteration of the test. The test runs five iterations for the read and write operations. The reason the test data is deleted between each test run is to prevent any kind of caching from taking place. However, in the case of the iDX6011, with its 64 GB of RAM, it seems like this test went straight to cache. Admittedly the data is still written to the drives, but there's lag between the test and the drive utilization, something we haven't seen on any NAS we've tested to date.


This is quite clear in the Task Manager, although we can't show you the network interface and drive utilization in parallel, we had to wait to run the next test until the drives had finished writing the data, to try and get somewhat accurate test results. In real-world usage, this is actually a good thing in most scenarios, since the cache will allow for faster copying of data to the NAS. That said, it might be worth investing in a UPS for the NASync iDX6011 Pro, since in the case of a power outage, you'd end up losing the data sitting in the cache, which could be a lot more data than what you'd lose when writing to mechanical drives or SSDs.