Four-Corner Performance Testing
All of the systems under test use traditional SMB/CIFS file transfers. SMB is the easiest NAS protocol because it simplifies file access to where you can simply type in "\\name_of_nas\folder_name" in file explorer to access the data on the system. We test and highlight the performance of SMB encrypted folders and iSCSI volumes on applicable models, but not all systems support the features.
Our test system is a modified Quanta MESOS CB220 Cluster-In-A-Box multi-node server. We upgraded the system with two Intel E5-2670 V1 processors and 128GB of system memory per node. An Intel X520 network interface card with two 10GbE ports resides in each node, as well. The system allows us to deploy four test nodes that execute custom software scripts in industry standard test software to measure our base performance metrics.
The test system and devices under test connect via a Netgear M6100 fully-modular switch, which provides ample 10GbE and 1GbE ports to sustain our large testing environment. This round of NAS testing is the second time we've used this switch. We discovered an issue with our previous switch and some NAS products that significantly reduced file transfer performance. The Netgear M6100 doesn't give us the same performance issues, and it allows us to expand our test environment while still running a dedicated network.
The feature-rich QNAP TS-453A is the only one of the four NAS in the test pool that we have covered in a dedicated review. It, as well as the Synology DS416J and Netgear ReadyNAS 214, are four-bay systems with prices that range between $289 and $599. The Synology DS416j and Netgear ReadyNAS 214 reviews are still in flight.
At low queue depths, the Drobo 5N transfers data at a lower rate than the other systems on the comparison list. The 5N comes alive with eight outstanding IO and matches the other systems at most block sizes during the read-centric workload.
Most file transfer to your NAS are sequential, unless you use it for hosting virtual machines or other advanced workloads. Again, we see that low queue depth sequential transfers, this time to the NAS, are slower on the 5N compared to many of the other systems. At higher queue depths, the 5N closes the gap and even outperforms some of the other systems.
Sequential Mixed Workloads
Nearly all consumer-focused storage products lose performance when you mix data writes with data reads. Mixed workloads cause severe performance degradation that the mechanical hard disk drives amplify due to the way they work. An equal mix of reads and writes is the most troublesome for hard disk drives. It's easy to replicate this state by simply reading one file from the NAS while writing another. The impact is even more severe if you increase the workload with multiple simultaneous transfers to and from the system.
Some network-attached storage systems use the system memory to buffer incoming data, a process that eases the bathtub curve and increases performance. The Drobo 5N with 2GB of system memory doesn't appear to use the same algorithm, or the buffer set aside for the cache fills faster than our test measures.
Marvell SoC processors are not known for exceptional random performance. Adding liberal doses of system memory helps to increase performance, but with only 2GB, the Drobo 5N has trouble reading back random data to the host computer. If you run applications from the 5N, the responsiveness will feel slower than if the data came from an internal hard drive.
The Drobo 5N fares much better than the other systems during the random write workload. The line showing performance isn't as smooth as the others, largely due to the impact of the cache.
Random Mixed Workloads
The high random write performance doesn't save the Drobo 5N from falling short with random mixed workloads. The tests in this section will carry over to the application workloads on the next page, which measure performance when we have software running from the NAS.