NETGEAR/Atheros: 108Mbps/802.11g Wireless Router Performance ...

607 views
537 views

Published on

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
607
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
2
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

NETGEAR/Atheros: 108Mbps/802.11g Wireless Router Performance ...

  1. 1. ® May 2004 www.veritest.com • info@veritest.com NETGEAR/Atheros: 108Mbps/802.11g Wireless Router Performance Testing Test report prepared under contract from NETGEAR Incorporated and Atheros Communications Incorporated Executive summary NETGEAR Inc. and Atheros Communications Inc. commissioned Key findings VeriTest, a division of Lionbridge The NETGEAR Wireless Router WGT624 and NETGEAR Technologies, Inc., to test the Wireless PC Card WG511T combination with “Static wireless network throughput 108Mbps” enabled for both router and PC Card generated performance of their 802.11g router the highest throughput numbers at seven of the eight specific and PC Card combination compared predetermined distances, compared to the four other wireless to similar Linksys 802.11g wireless configurations we tested. solutions. At a distance of 5 feet, the NETGEAR Wireless Router WGT624 and NETGEAR Wireless PC Card WG511T Our tests measured the wireless combination with “Static 108Mbps” generated an average network throughput performance of throughput of 62.9 Mbps. This was 68% better than its next each wireless PC Card and router closest wireless combination, NETGEAR Wireless Router solution using the system WGT624 and NETGEAR Wireless PC Card WG511T configurations described below combination with “b and g Only” enabled. running the IEEE 802.11b, 802.11g, and each vendors propitiatory (NETGEAR’s 108Mbps G™ technology and Linksys’ Speedboostertm) wireless network protocols. Our testing aimed to simulate a typical indoor office environment. To test the throughput of each configuration, we used NetIQ’s (now Ixia’s) Chariot 4.3 application to generate file-based network traffic between each external wireless solution and a wired desktop system connected through a wireless router. We installed one Chariot 4.2 endpoint on the test notebook system configured with each external wireless solution and a second endpoint on the wired desktop system. For this test, VeriTest supplied a Gateway Solo 9500 notebook system with a single Pentium-III 933MHz processor, 256MB of RAM and running Microsoft Windows 2000 Professional SP4 operating system for use as our wireless endpoint. Additionally, we provided a Dell Dimension XPS 600r desktop system with a Pentium-III 600MHz processor, 256MB of RAM and a 3Com 3c905B-TX Ethernet adapter running Microsoft Windows 2000 Professional SP4 for use as our wired desktop system. NETGEAR supplied all of the wireless routers and PC Cards used during the testing. All of the wireless devices arrived for testing in retail shrink- wrapped packaging. Atheros supplied an IBM ThinkPad T40 running Microsoft Windows XP Professional SP1 preconfigured with the Chariot 4.3 console and required licenses. Figure 1 shows the wireless configurations for the 802.11g throughput performance we tested.
  2. 2. Wireless Router Wireless PC Card Wireless Router Mode Wireless PC Card Mode NETGEAR WGT624 NETGEAR WG511T Static 108 Enabled Static 108 Enabled NETGEAR WGT625 NETGEAR WG511T b and g only Enabled b and g only Enabled NETGEAR WGT626 Linksys WPC54GS Dynamic 108 Enabled Linksys Afterburner Disabled Linksys WRT54GS Linksys WPC54GS SpeedBooster Enabled SpeedBooster Enabled Linksys WRT54GS NETGEAR WG511T SpeedBooster Enabled b and g only Enabled Figure 1. Wireless Test configurations NETGEAR’s proprietary Static 108 mode operates only on their “108Mbps G™ technology” labeled devices and is based on Atheros Super G™ technology. The unique technology helps to provide improved throughput capabilities in several ways. Super G™ is a portfolio of components that incrementally increase end user throughput. These components operate independently and engage when additional bandwidth is available and/or needed. The Super G components consist of frame bursting, fast frames, data compression, and static/dynamic turbo. Bursting allows multiple frames to be transmitted in a “burst”; increasing throughput by decreasing overhead. Fast frames provides more information “per frame” to be transmitted which also reduces overhead and increases throughput. Compression is hardware-based and provides the capability to compress data before sending data over the wireless network and then decompressing the data on the other side. This helps to reduce the amount of time required to transfer files. The turbo (or double-channel bonding) feature runs in static or dynamic 108Mbps mode on both the wireless router and the station. It operates by using two radio channels, bonding them together so they appear as a single channel to both the wireless router and station. For further details about NETGEAR’s 108Mbps G™ technology please refer to the following Web site. http://www.netgear.com/products/wireless/108_technology.php NETGEAR’s “b and g only” mode operates using the IEEE standard for 802.11b and 802.11g while including the benefits of bursting, fast frames, and compression. These industry standard wireless protocols run at 1, 2, 5.5, and 11Mbps for 802.11b and up to 54Mbps for 802.11g. These modes are available on both their wireless router and the station. The Super G™ technology runs in conjunction with other 802.11g devices with the exception of static turbo. Linksys’ proprietary SpeedBoostertm or Afterburnertm technologies operate only on their SpeedBoostertm labeled devices. This proprietary protocol also runs in conjunction with other 802.11g devices. Their unique technology helps to provide improved throughput capabilities by chaining packets together to decrease the amount of packet overhead and increase overall throughput. For further details about Linksys’ proprietary SpeedBoostertm technology, please refer to the following Web site. http://www.linksys.com/splash/speedbooster_splash.asp Please refer to the Test methodology section of this report for complete details on how we conducted the testing. Throughput Test Results The throughput testing consists of using Chariot to generate an incoming stream of network traffic from the wired desktop system through a wireless router to the client notebook system with each wireless station under test at distances of 5, 40, 60, 75, 100, 125, 150, 175, and 200 feet. At NETGEAR’s request, wee did not test the cross vendor combinations at the 5 and 40ft distances. Figure 2 shows the results of the testing for each wireless configuration we tested in megabits per second at each distance tested. For these tests higher throughput numbers are better. NETGEAR/Atheros. 108Mbps/802.11g Wireless Router Performance Testing 2
  3. 3. For these 802.11g receive tests, figure 1 shows that the NETGEAR Wireless Router WGT624 and NETGEAR Wireless PC Card WG511T combination with “Static 108Mbps” enabled for both router and PC Card generated the highest throughput numbers at all but the 175 feet test distance when compared to each of the other participating wireless configurations. In addition, the NETGEAR Wireless Router WGT624 and NETGEAR Wireless PC Card WG511T combination with “Static 108Mbps” enabled reported the highest throughput at any single test distance, namely 62.9Mbps at 5 feet. Also at a distance of 5 feet, the NETGEAR Wireless Router WGT624 and NETGEAR Wireless PC Card WG511T combination with “Static 108Mbps” enabled was 68% faster than its next closest wireless combination, the NETGEAR Wireless Router WGT624 and NETGEAR Wireless PC Card WG511T combination with “b and g Only” enabled at 37.5Mbps. Finally, the NETGEAR Wireless Router WGT624 and NETGEAR Wireless PC Card WG511T combination with “b and g Only” enabled, reported the highest throughput at a distance of 175 feet, namely 2.6Mbps. This tested 46% faster than the next closest wireless combination, the NETGEAR Wireless Router WGT624 and Linksys Wireless-G Notebook Adapter, which generated an average of 1.8Mbps at the same distance. It is worth noting that 175 feet is the maximum distance at which any of the wireless combinations was able to complete the three 1-minute test iterations. Receive Throughput Performance Chart 70.0 60.0 NETGEAR Router & NETGEAR PC Card (Static 108Mbps) Throughput (megabits per second) 50.0 NETGEAR Router & NETGEAR PC Card (“g and b” only) 40.0 NETGEAR Router & Linksys PC Card ("Auto 108Mbps", disable Afterburner) 30.0 Linksys Router & NETGEAR PC Card (SpeedBooster On, “g and b” only) 20.0 Linksys Router & Linksys PC Card (SpeedBooster On) 10.0 0.0 5-ft 40-ft 60-ft 75-ft 100-ft 125-ft 150-ft 175-ft Distance (feet) Figure 2. Wireless Throughout Receive Chart NETGEAR/Atheros. 108Mbps/802.11g Wireless Router Performance Testing 3
  4. 4. Testing methodology NETGEAR Inc. and Atheros Communications Inc. commissioned VeriTest, a division of Lionbridge Technologies, Inc., to test the wireless network throughput performance of their 802.11g router and PC Card as compared to comparable Linksys 802.11g wireless solutions. Testbed wireless configurations Specifically, we tested the following wireless configurations listed in figure 3 for the 802.11g throughput performance. Wireless Router Wireless PC Card Wireless Router Mode Wireless PC Card Mode NETGEAR WGT624 NETGEAR WG511T Static 108 Enabled Static 108 Enabled NETGEAR WGT625 NETGEAR WG511T b and g only Enabled b and g only Enabled NETGEAR WGT626 Linksys WPC54GS Dynamic 108 Enabled Linksys Afterburner Disabled Linksys WRT54GS Linksys WPC54GS SpeedBooster Enabled SpeedBooster Enabled Linksys WRT54GS NETGEAR WG511T SpeedBooster Enabled b and g only Enabled Figure 3. Wireless Test configurations Testbed configuration The throughput testing consisted of sending a stream of data from a wired desktop system through a wireless router to a notebook system running on of the Wireless PC Cards under test. We tested for performance using an indoor office environment. We used NetIQ’s (now Ixia’s) Chariot 4.3, supplied by Atheros, to generate network traffic between the 802.11g wireless stations and a wired Ethernet desktop end point system, which bridged through each wireless router. Chariot is a network traffic generation product commonly used to measure network throughput performance by sending a continuous stream of data between two endpoints. The Chariot controller system monitors and collects the reported information for each of the test iterations. For the Chariot controller system, we installed Windows 2000 Server SP3 on an IBM ThinkPad T40 notebook system configured with an Intel P4-1.8GHz processor, 512 MB RAM, and a single internal Intel Pro/100 NIC. For the Desktop endpoint system, we installed Windows 2000 Professional SP4 on a Dell Dimension XBR 600r desktop system configured with an 600MHz Intel Pentium III processor and 256MB RAM, and a single 3Com 3c905B-TX Integrated Fast Ethernet controller. We connected the Chariot controller system to the wireless router under test and the desktop endpoint system via the built-in LAN 100Mbps full duplex switch. Each external wireless solution connected to the DSS (Distributed Service Set) using its wireless connection to the router. Figure 4 shows the basic components and network configuration we used for the tests in this report. We listed detailed information about each of the client notebook system, Chariot controller system, Chariot desktop endpoint system, wireless routers, and wireless PC Cards in Appendix A of this report. NETGEAR/Atheros. 108Mbps/802.11g Wireless Router Performance Testing 4
  5. 5. Distance from Wireless Router 5, 40, 60, 75, 100, 125, 150, 175, 200 feet Wireless Chariot End Point 1 Router Gateway Solo 9500 PIII/900MHz 256MB Chariot Controller Windows 2000 100Mbps Chariot End Point 2 Professional SP4 IBM ThinkPad T40 Dell Dimension XPS T600 Full Duplex P-Mobile 1.6GHz PIII/733MHz 256MB Connections Intel Pro/100+ VE 3Com 3c905B-TX Windows XP Pro SP1a Windows 2000 Professional SP4 Figure 4. Wireless test bed setup For the throughput performance tests, we did not use encryption. We configured fixed IP addresses for the Chariot controller system, the desktop endpoint system, each wireless solution and wireless router. Figure 5 shows the setting we used for our tests. Test System Internet Protocol address Chariot Controller System 192.168.0.200/24 Wireless solution endpoint 1 192.168.0.35/24 Desktop endpoint 2 192.168.0.201/24 Wireless Router 192.168.0.1/24 Wireless Network Settings SSID NETGEAR LinksysVT SSID Linksys NETGEARVT Figure 5. Distributed Service Set Network Settings For the indoor office environment tests, we tested the wireless station at distances up to 200 feet from each wireless router. We measured the distances in a straight line from the router to each location where we tested each wireless station. Figure 6 shows a diagram of the indoor office environment we used for testing. The interior wall construction consists of dry wall and metal studs. We used the following fixed distances with the wireless station pointing directly towards the wireless router. • 5ft, 40ft, 60ft, 75ft, 100ft, 125ft, 150ft, 175ft, and 200ft For the 5 feet and 40 feet tests, we moved the wireless routers closer into the lab to include at least one wall at 40 feet. Figure 6. indoor office environment NETGEAR/Atheros. 108Mbps/802.11g Wireless Router Performance Testing 5
  6. 6. Throughput performance test configuration To perform the throughput performance tests we used a modified Chariot 4.3 script provided by Atheros that is based upon file send long (FILESNDL.SCR) script. Figure 7 shows the parameters changed and their corresponding default setting. This script emulates sending a file from a wired Ethernet desktop system through a wireless router to a laptop configured with the wireless station. Provided that the wireless devices can handle data compression, setting the send_datatype parameter to a value of news.cmp was chosen to reflect more typical data (or semi-compressible data). For this testing, the NETGEAR devices are the only combinations that offer this technology feature and must be used together. If one of the devices participating in any of the test combinations we tested does not support compression, then the data is handled as a standard data type with no compression. The send_data_type setting sets the data type as text in the encapsulated package versus a NOCOMPRESS setting that sets the data type as non-compressible raw data. Because the data is compressible, the file_size parameter can be increased. In this case, Atheros chose 1,000,000 bytes. The buffer settings are set to the maximum threshold the wireless hardware can handle, namely 65,535 bytes. Atheros Default File Variable Name Values Send Long v.4.1 Description initial_delay 0 0 Pause before the first transaction number_of_timing_records 100 100 How many timing records to generate transactions_per_record 10 1 Transactions per timing record file_size 1000000 100000 How many bytes in the transferred file send_buffer_size 65535 Default value How many bytes of data in each SEND receive_buffer_size 65535 Default value How many bytes of data in each RECEIVE transaction_delay 0 0 Milliseconds to pause send_datatype news.cmp NOCOMPRESS What type of data to send send_data_rate UNLIMITED UNLIMITED How fast to send data destination_port AUTO AUTO What port to use for Endpoint 2 close_type Reset Reset How connections are terminated source_port AUTO AUTO What port to use for Endpoint 1 Figure 7. Modified FILESENDLONG script changes Chariot Controller configuration Chariot configuration for each of the three iterations for each test script (send/receive) at each distance: • Run for a fixed duration: 1 min. • Reporting type: Batch • Automatically poll endpoints: No We computed an average for the three iterations and report our results in megabits per second at each distance we tested. NETGEAR/Atheros. 108Mbps/802.11g Wireless Router Performance Testing 6
  7. 7. Test results This section provides the details of our tests comparing the wireless network performance of NETGEAR and Linksys combination wireless product. Please refer to the Test methodology section of this report for complete details on how we conducted the testing. Receive Data Throughput Test Results The throughput testing consists of using Chariot to generate an incoming stream of network traffic from the wired desktop system through a wireless router to the client notebook system with each wireless station under test at distances of 5, 40, 60, 75, 100, 125, 150, 175, and 200 feet. At NETGEAR’s request, the cross vendor combinations were not tested at the 5 and 40ft distances. Figure 1 shows the results of the testing for each wireless configuration we tested in megabits per second at each distance tested. For these tests higher throughput numbers are better. For these 802.11g receive tests, figures 8 and 9 show that the NETGEAR Wireless Router WGT624 and NETGEAR Wireless PC Card WG511T combination with “Static 108Mbps” enabled for both router and PC Card generated the highest throughput numbers at all but the 175 feet test distance when compared to each of the other participating wireless configurations. In addition, the NETGEAR Wireless Router WGT624 and NETGEAR Wireless PC Card WG511T combination with “Static 108Mbps” enabled reported the highest throughput at any single test distance, namely 62.9Mbps at 5ft. At a distance of 5 feet, the NETGEAR Wireless Router WGT624 and NETGEAR Wireless PC Card WG511T combination with “Static 108Mbps” enabled tested 68% faster than its next closest wireless combination, the NETGEAR Wireless Router WGT624 and NETGEAR Wireless PC Card WG511T combination with “b and g Only” enabled at 37.5Mbps. Finally, the NETGEAR Wireless Router WGT624 and NETGEAR Wireless PC Card WG511T combination with “b and g Only” enabled, reported the highest throughput at a distance of 175 feet, namely 2.6Mbps. This was 46% faster than the next closest wireless combination, the NETGEAR Wireless Router WGT624 and Linksys Wireless-G Notebook Adapter, which generated an average of 1.8Mbps at the same distance. It is worth noting that 175 feet is the maximum distance at which any of the wireless combinations was able to complete the three 1-minute test iterations. Linksys Router NETGEAR Router Linksys Router NETGEAR PC Card NETGEAR Router NETGEAR Router Linksys PC Card Linksys PC Card SpeedBooster On NETGEAR PC Card NETGEAR PC Card Auto 108Mbps" Distance SpeedBooster On b and g Only Static 108Mbps “g and b” only Afterburner Off 5-ft 34.8 N/A 62.9 37.5 N/A 40-ft 33.3 N/A 62.1 34.4 N/A 60-ft 31.6 23.8 61.4 32.9 25.0 75-ft 27.4 20.5 58.9 32.4 14.4 100-ft 3.7 13.0 42.4 10.7 11.8 125-ft 0.5 3.5 23.3 5.6 4.0 150-ft 0.0 0.0 8.7 3.0 3.1 175-ft 0.0 0.0 0.0 2.6 1.8 Figure 8. Wireless Throughput Receive Results Table in megabits per second NETGEAR/Atheros. 108Mbps/802.11g Wireless Router Performance Testing 7
  8. 8. Receive Throughput Performance Chart 70.0 60.0 NETGEAR Router & NETGEAR PC Card (Static 108Mbps) Throughput (megabits per second) 50.0 NETGEAR Router & NETGEAR PC Card (“g and b” only) 40.0 NETGEAR Router & Linksys PC Card ("Auto 108Mbps", disable Afterburner) 30.0 Linksys Router & NETGEAR PC Card (SpeedBooster On, “g and b” only) 20.0 Linksys Router & Linksys PC Card (SpeedBooster On) 10.0 0.0 5-ft 40-ft 60-ft 75-ft 100-ft 125-ft 150-ft 175-ft Distance (feet) Figure 9. Wireless Throughout Receive Chart NETGEAR/Atheros. 108Mbps/802.11g Wireless Router Performance Testing 8
  9. 9. Appendix A. Wireless Driver and Utility Disclosures Device Driver/Firmware Rev. Date Version Wireless Client Utility NETGEAR Wireless Router WGT624 V3 4.0.4 2/13/04 V3 Web Interface NETGEAR WG511T 3.0.1.6 3/08/04 3.6.0 Smart Wireless Utility Linksys Wireless Router WRT54GS 3.50.21 2/19/04 N/A Web Interface Linksys Wireless-G Notebook Adapter Wireless Network WPC54GS 3.50.21.11 2/26/04 V2.07.1 Monitor B. System Disclosures IBM ThinkPad T40 – Chariot Controller System Processor / Speed / # of CPUs P4M / 1.6GHz / 1 System RAM / Type / # of Slots 256MB Ethernet Adapter Wired: Intel Pro 100 Operating systems Windows XP Professional SP1a Gateway Solo 9500 – Chariot Client System (End Point 1) Processor / Speed / # of CPUs PIII / 900MHz / 1 System RAM / Type / # of Slots 256MB Ethernet Adapter Wired. Intel Pro 100 VM – not used Operating systems Windows 2000 Professional SP4 Dell XPS 600r – Chariot Desktop System (End Point 2) Processor / Speed / # of CPUs PIII / 600MHz / 1 System RAM / Type / # of Slots 256MB Ethernet Adapter Wired. 3Com 3c905-TX Operating systems Windows 2000 Professional SP4 NETGEAR/Atheros. 108Mbps/802.11g Wireless Router Performance Testing 9
  10. 10. VeriTest (www.veritest.com), the testing division of Lionbridge Technologies, Inc., provides outsourced testing solutions that maximize revenue and reduce costs for our clients. For companies who use high-tech products as well as those who produce them, smoothly functioning technology is essential to business success. VeriTest helps our clients identify and correct technology problems in their products and in their line of business applications by providing the widest range of testing services available. VeriTest created the suite of industry-standard benchmark software that includes WebBench, NetBench, Winstone, and WinBench. We've distributed over 20 million copies of these tools, which are in use at every one of the 2001 Fortune 100 companies. Our Internet BenchMark service provides the definitive ratings for Internet Service Providers in the US, Canada, and the UK. Under our former names of ZD Labs and eTesting Labs, and as part of VeriTest since July of 2002, we have delivered rigorous, objective, independent testing and analysis for over a decade. With the most knowledgeable staff in the business, testing facilities around the world, and almost 1,600 dedicated network PCs, VeriTest offers our clients the expertise and equipment necessary to meet all their testing needs. For more information email us at info@veritest.com or call us at 919-380-2800. Disclaimer of Warranties; Limitation of Liability: VERITEST HAS MADE REASONABLE EFFORTS TO ENSURE THE ACCURACY AND VALIDITY OF ITS TESTING, HOWEVER, VERITEST SPECIFICALLY DISCLAIMS ANY WARRANTY, EXPRESSED OR IMPLIED, RELATING TO THE TEST RESULTS AND ANALYSIS, THEIR ACCURACY, COMPLETENESS OR QUALITY, INCLUDING ANY IMPLIED WARRANTY OF FITNESS FOR ANY PARTICULAR PURPOSE. ALL PERSONS OR ENTITIES RELYING ON THE RESULTS OF ANY TESTING DO SO AT THEIR OWN RISK, AND AGREE THAT VERITEST, ITS EMPLOYEES AND ITS SUBCONTRACTORS SHALL HAVE NO LIABILITY WHATSOEVER FROM ANY CLAIM OF LOSS OR DAMAGE ON ACCOUNT OF ANY ALLEGED ERROR OR DEFECT IN ANY TESTING PROCEDURE OR RESULT. IN NO EVENT SHALL VERITEST BE LIABLE FOR INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES IN CONNECTION WITH ITS TESTING, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. IN NO EVENT SHALL VERITEST'S LIABILITY, INCLUDING FOR DIRECT DAMAGES, EXCEED THE AMOUNTS PAID IN CONNECTION WITH VERITEST'S TESTING. CUSTOMER’S SOLE AND EXCLUSIVE REMEDIES ARE AS SET FORTH HEREIN. NETGEAR/Atheros. 108Mbps/802.11g Wireless Router Performance Testing 10

×