Intel NASPT Test Results
NASPT brings an important perspective to our test protocol, as it is designed to measure the performance of a NAS system, as viewed from the end user’s perspective. Benchmarks like ATTO use Direct I/O Access to accurately measure disk performance with minimal influence from the OS and the host platform. This provides important, objective data that can be used to measure raw, physical performance. While it’s critical to measure the base performance, it’s also important to quantify what you can expect using real-world applications, and that’s exactly what NASPT does. One of the disadvantages of NASPT is that it is influenced by the amount of memory installed on the client, and it was designed for systems that had 2-4 GB of RAM. Consequently, two of the tests give unrealistic results with modern systems, because they are measuring the speed of the buffer on the client, instead of the actual NAS performance. For that reason, we will completely ignore the results for “HD Video Record” and “File Copy to NAS”.
With a single, basic GbE interface in place, no individual test gets past the ~ 120 MB/s theoretical barrier, but several of them are in the 70-90 MB/s range. Some of the tests have very low transfer rates, and that’s due to the nature of the test. The Content Creation test for example, simulates a user creating a web page, accessing multiple sources for the content. The Directory Copy tests use several hundred directories and several thousand files to test a typical backup and restore scenario. That’s one of the most real-world types of test, and it’s useful for all of us to have a standard set of test data to use, because my directory of 1,000 random small files is never going to be the same as your directory of 1,000 random small files.
In keeping with the real-world scenario of this particular benchmark, I normally run these tests only on the RAID 5 configuration, since that is what most users with a mid-size NAS are going to use. It just doesn’t make sense to run realistic test scenarios on unrealistic hardware configurations. I couldn’t do that in this case, but I did run them on the single disk setup. This is the first time I’ve had a small NAS on the test bench since I started using NASPT, so I don’t have any comparative results to share. I don’t want to mix these results in with the others on a common chart, as that would not be a fair comparison. So for now, here are the summary results representing an average of five individual runs. As a whole, these are very good results. Considering that only one disk is in operation here, I’m surprised by how close the results are to all the 4-bay NAS servers that I’ve tested with RAID 5 configurations. Of course, the single GbE interface is the short pole in the tent, to borrow and butcher a common corporate phrase. Even with four disks spinning away, the 125 MB/s theoretical Ethernet limitation scales everything back. Even though they’re close, I still don’t want to mix the two sets of results together. Just know that this is very competitive performance, and if your storage requirements don’t dictate four bays worth of disks running RAID 5, you don’t need to give up throughput performance.
The Intel NASPT benchmarking tool measures tasks that almost any NAS user would be interested in. Whether they have one disk, two, four, or eight, the workload doesn’t change much. The workload changes when the environment does; a NAS in a LAN room of a SMB probably handles different scenarios than one that sits on a shelf in someone’s home, next to the cable modem and 802.11b/g/n Wi-Fi router. The results above are quite respectable for a home-based NAS server, which is almost always going to be limited by the bandwidth of a single GbE connection. When I first thought about running the NASPT tests, I told myself, “What’s the point, you can’t compare them to the NAS servers you’ve already tested.” After seeing the results, I now see that this benchmark still represents a valid test. The results are definitely more realistic than let’s say, running Crysis on an 8400GT video card….oh wait, that crashed.
Let’s look at a couple more benchmarks that test both sequential and random access performance: ATTO Disk Benchmark and CrystalDiskMark.