NETGEAR GS110T Gigabit Smart Switch Review


<< PREVIOUS            NEXT >>

Network Switch Testing Methodology

To benchmark a network switch, I’m using the most straightforward and standardized test I have: Passmark Performance v7.0 Advanced Network Test. This test measures throughput between two PCs connected through the switch which is the device under test. One PC is set up as the ‘Client’ and the other is set up as the ‘Server’. Each test is run at least five times, the highest and lowest results discarded, and the remaining results are averaged to give a final result.

To establish a baseline benchmark speed, I connected two workstations (Test System #1 and #2, below) to the NETGEAR GS110T switch, and ran the PassMark Advanced Networking Test in Client-Server mode. I also used Test System #3, and discovered that its CPU couldn’t keep up with the GS110T switch, so all benchmark results displayed below were generated using the Quad-Core Intel systems. The Advanced Network Test is designed to test the data transfer rate between two computers both of which must be running the PerformanceTest application. One of the computers must act as the server and the other computer acts as a client. It connects to the server machine and sends data to it for the user-specified duration of the test. Unless at least 10 test samples have been taken, the test results are not valid, so this application takes 100 samples during each network test run. I used all default values for this benchmark, as shown in the screenshot below. The screens have different elements grayed out, depending on whether you are sitting at the Client computer or on the Server side.


This benchmark eliminates most of the variables involved in network speed testing, but not all. The PCs themselves can introduce spurious issues, such as hardware bandwidth limitations, resource conflicts, wait states, and buffer inconsistencies. The following chart shows why you have to dig a little deeper than just looking at the Average Transmission rate that is displayed on the main screen. For some reason, the start of this test was impacted, and was only running at about half speed for the first two seconds. The average value (shown in yellow) doesn’t really reflect the true capability of the network. I’m a firm believer in using real-world test scenarios as part of a comprehensive benchmarking effort, but we have to be careful not to confuse anomalies with persistent limitations. After several minutes the average might approach the real throughput level, but after 20 seconds (a looooong time in network transmissions), the calculated average result is almost 10% lower than the true number.


Another test run, with slightly different hardware, looked completely different. This kind of result showed more of what I think of as a real-world limitation, somewhere in the signal flow. In this case, the averaging function really does illustrate the true capability of the system, and 20 seconds was enough this time for the calculated average to settle down to the correct value. Note, this was a test run with the AMD Phenom II X2 system, and it was the only machine that showed this behavior.


And then there’s this almost perfect chart; there’s no doubt about what the real-world throughput of this system is. You can still see a few minor dips and peaks, but they’re only 1-2% variations.


Test System 1 (Server)

Test System 2 (Client)

Test System 3 (Client)

  • Motherboard: ASUS M4A88TD-V EVO/USB3 (1308 BIOS)
  • System Memory: 4 x 2GB OCZ, AMD Black Edition OCZ3BE1600C8LV4GK, DDR3 1600 (8-8-8-24)
  • Processor: AMD Phenom II X2 555 Black Edition Callisto 3.2GHz Socket AM3
  • CPU Cooler: Cooler master Z600
  • Video: ATI Radeon HD 5670 512MBB GDDR5 (Catalyst 8.801.0.0)
  • Drive 1:Corsair F100 SSD 100GB (CSSD-F100GB2)
  • Drive 2: Western Digital Velociraptor 300GB Drive WD3000HLFS
  • Enclosure: SilverStone Fortress FT01 SST-FT01B-W Gaming Case
  • PSU: OCZ ModXStream Pro 500 Modular ATX12V V2.2, OCZ500MXSP
  • Operating System: Windows 7 Home Premium SP1

Support Equipment

  • Intel EXPI9301 CT Gigabit Ethernet NIC, x1 PCIe 1.1, 1x CAT5
  • Intel E10G42BT, X520-T2, 10Gbps Ethernet NIC, x8 PCIe 2.0, 2x CAT6a
  • 10-Foot Category-6 Solid Copper Shielded Twisted Pair Patch Cables

Networking Comparison Products

  • NETGEAR 10-port GbE Smart Switch GS110T
  • TRENDnet 8-port GbE Switch TEG-S80g
  • TRENDnet TEW-673GRU Dual Band N300 Wireless Router
  • Linksys EA4500 Dual Band N450 Wireless Router


<< PREVIOUS            NEXT >>

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>


This site uses Akismet to reduce spam. Learn how your comment data is processed.