The test system used in this review is equipped with an Intel Core i7-6700K CPU, GIGABYTE GA-Z170X-UD3 motherboard, 32GB (16GB x 2) of Crucial Ballistix Sport LT DDR4 memory, Samsung 960 PRO 512GB SSD and a GIGABYTE GeForce GTX 1060 WINDFORCE OC 6G graphics card. For the operating system, I installed a fresh copy of Windows 10 Enterprise.
To test the performance of the ioSafe 1019+, I ran a series of benchmarks using CrystalDiskMark 3.0, ATTO Disk Benchmark 2.46, Iometer and the Intel NAS Performance Toolkit. The tests were run with the included Seagate IronWolf 4TB NAS hard drives configured as a single RAID 6 volume. The ioSafe 1019+ was connected to the computer using CAT6 ethernet cables and a gigabit Cisco switch. Unless otherwise noted, caching with the two included Samsung 512GB 970 Pro SSDs has been disabled.
CrystalDiskMark 3.0:
First, I ran a few quick tests using CrystalDiskMark. This benchmark tool measures the performance of a storage device by testing its sequential read and write speeds as well as its random read and write speeds using blocks 4KB and 512KB in size.
ioSafe doesn't really say what kind of speeds the 1019+ is capable of. Using CrystalDiskMark we can see that the NAS is able to read at 118 MB/s and write at 117 MB/s when connected to 1GbE network.
The 1019+ also performed fairly well when doing encrypted file transfers. Here too, it was able to read at 118 MB/s and write at 117 MB/s. However, its transfer rated dropped by about 40 MB/s when doing random writes at high queue depths.
ATTO Disk Benchmark 2.46:
I also used ATTO Disk Benchmark to test the 1019+'s sequential read and write speeds. The tests are run using blocks ranging in size from 0.5KB to 8192KB and the total length set to 32MB and 256MB.
The 1019+'s performance was about the same when tested with ATTO. Looking at the screenshot above, you can see that the NAS was able to read and write at 118 MB/s.
Iometer:
Next, I ran a series of tests using Iometer. This tool can be configured to benchmark a number of things. In this case, I used it to measure the 1019+'s read and write speeds and the number of operations per second using blocks ranging from 512B to 2MB in size. The tests were run using random bytes and a queue depth of 3.
The 1019+ performed relatively well when doing sequential reads and writes with Iometer. In both cases, the NAS was able to reach speeds as high as 112 MB/s.
The 1019+ did relatively well when doing random reads and writes. With the SSD cache disabled, it was able to read at 50 MB/s and write at about 28 MB/s. With the cache enabled, these numbers jumped considerably. In some cases, the random write speed increased by more than 70 MB/s.
Here too, you can see the benefit of having an SSD cache. In our tests, the number of IOPS jumped from about 218 up to 2057 when randomly writing 4K blocks. This performance advantage slowly drops as the block size gets bigger, but even with 64K blocks, the number of IOPS was still seven times higher with the cache enabled.
Intel NAS Performance Toolkit:
The Intel NAS Performance Toolkit (Intel NASPT) is a file system exerciser and analysis tool designed to enable performance comparisons between network attached storage (NAS) devices. Intel NASPT focuses on user level performance using real world workload traces gathered from typical digital home applications: HD video playback and record, data backup and restore utilities, office productivity applications, video rendering/content creation, and more.
The 1019+ reached some respectable speeds when streaming HD video and copying large files to and from the server. As with other NAS devices though, its transfer rates dropped considerably when creating content and copying directories full of small files to and from the NAS.
