Benchmarks - Multi-Client Test
The Intel NAS Performance Tool (NASPT) is an excellent way to determine NAS performance in a single user environment. Any review that only uses NASPT assumes that only a single computer will access the target NAS at one time. We took issue with this method of testing and spent over a year designing, building, programming and finally validating the TweakTown Multi-Client Test.
The test uses Microsoft Office data recorded to traces and played back to the NAS from up to 120 client Windows 7 installations (clients). We record total throughput of all clients and average response time per client.
Over time we'll populate the two multi-client charts with several NAS products from a span of categories. The products range from a dual Xeon server with 2x 10GbE to a 2-bay NAS with a single gigabit Ethernet connection. The products will fall into their performance categories based on performance and not marketing material or opinion.
The LaCie 5big NAS Pro dominates the other products in the same price category. We have the Thecus N8900 with an Intel X520-SR2 installed for our own reasons. It also helps to keep small performance differences from appearing larger than they really are.
LaCie is one of the few companies to list a recommended user count. Here we see their 50 user suggesting is spot on, the point just before the throughput flattens out, but latency increases.
Looking at the latency, the 5big does a really good job of keeping the latency low. The 5big is also the only product shown with Seagate HDDs and not our standard Western Digital REDs. This is a topic we'll have to cover in a future article when we put the RED head to head with Seagate's new NAS HDD.
PRICING: You can find products similar to this one for sale below.
United States: Find other tech and computer products like this over at Amazon's website.
Australia: Find other tech and computer products like this over at PLE Computer's website.
Canada: Find other tech and computer products like this over at Amazon Canada's website.