 |
|
|
by Steve Broadhead Table of Contents Ethernet
switches:� simple purveyors of 'commodity' bandwidth or intelligent
networking devices? Ethernet switches: simple purveyors of �commodity� bandwidth or intelligent networking devices? There was a time when Ethernet switching was declared as groundbreaking technology. Ethernet reinvented - faster and without the legendary collision flaws where packets blindly fizzed around the network in search of their real destination, suffering several head-on collisions in the process. Call it a high-speed, one-way system if you will, but full-duplex Ethernet switching, maximising bandwidth by removing the possibility of collisions, really did resolve a huge traffic problem on Ethernet networks. As such it merited the term �groundbreaking�, even if �switching� was itself derived from multiport bridges invented in the late �80�s but that�s another, and largely irrelevant, story. The point is, what was advanced technology a few years ago is now so commonplace as to be considered commodity. One look at the price, per port, of Ethernet switches five years ago compared with now, shows just how commoditised the switch market has become. What was formerly the price per port can, in some instances, now buy you a large percentage of the whole switch! But is it really as simple as just buying bandwidth as cheaply as possible, or is there more to consider when looking at an Ethernet switch? Well yes there is. The old classic �throw more bandwidth at a problem and it�ll go away� never has worked in the medium � let along long � term and certainly will be less and less effective as the applications themselves are getting smarter.� Applications Are The Priority For too long networking was black art, a science dedicated to hiding the reality � that networking hardware is essentially plumbing, albeit complex plumbing at that. Even domestic plumbing would be of no use if �applications� such as the delivery of hot and cold water, central heating and air-conditioning around the home and office were not required. And so it is with the networking variety. There is no use in �plumbing in� gigabytes of networking capacity unless there are applications in demand that take advantage of this capacity. In the same way that you don�t base a central heating system for a small house on a boiler designed to provide heating to an office-block just to guarantee that it works, similarly networking should be based around intelligent use of what you have and what you need, not over-engineering a solution at great expense just to be on the safe side. For this reason, the IEEE committee introduced 802.1p, a standards-based method of assigning priority to applications (based on a value between 0-7) as they pass through the network. In this way it is entirely feasible to run real-time, high-priority applications alongside bandwidth-hogging but low-priority jobs (such as file transfer and large print jobs) without dedicating expensive bandwidth to each and every user/application combination on a permanent basis. 3Com�s 4400, reviewed here, supports this standard. Quality of service was once something you only associated with the reseller providing the kit. At best networks had two inherent service levels � up or down. But now even the PC clients are capable of creating policies that define the required amount of bandwidth for a given application and the devices on the network must be able to support this type of requirement.� Bandwidth itself, then, is not enough. In this report therefore we�re focusing on what 3Com�s SuperStack 4400 has to offer beyond basic, backplane bandwidth, though we have also given it lengthy soak tests and liberally thrown gigabytes of data at it, to see how it copes with the classic network overload scenario. The other point we have borne in mind during the testing is just how easy is it for a relatively inexperienced computer user to take advantage of what performance enhancement and control features the 4400 has to offer? After all, there is no point in packing a product with an ultra-high feature specification if 90% of the target users are not going to be able to access those features. We�ve experienced enough of those days back in the early �90�s.� Networking is no longer black art but day-to-day computing, delivering essential services to users, whether they be simple or complex requirements. To the user, they are all important needs, �must haves� even. So an Ethernet switch has to be able to mirror this level of requirement now, both in terms of its feature set and the ease of access to those features. The question is, then, can 3Com deliver what is possibly a tricky combination of rich feature set and ease of use in its third generation of stackables, represented here by the 4400 Ethernet switch? Read on to find out� Superstack 3 switch 4400 reviewed The 4400 is, put simply, a stackable 24-port 10/100 (auto-negotiating, half/full duplex) Ethernet switch in 19� rack-mount format. The design is very neat, taking up just one �unit� (4.4cm) of space in the wiring closet. Stacks of up to eight 4400�s can be created, connecting via a dedicated backplane module. A second module slot supports optional 100BASE-TX or Gigabit Ethernet (UTP) uplinks, providing a maximum of 200 switch ports per stack and up to 8000 physical addresses. The complete stack can be managed as a single entity with a single IP address. It also enables switches within the stack to be hot-swapped and/or to be configured as resilient connections. A second, redundant PSU can be attached via a dedicated external connection. Support for Spanning Tree rounds of the resilient features of the 4400 Maximum packet forwarding rate is claimed to be 6.6million packets per second according to 3Com � wire speed on all ports basically. More interesting, however, is support for layer 4 networking prioritisation. The 4400 adds to basic IEEE 802.1p traffic prioritisation support by providing an easy way of creating application-based prioritisation using wizard-based configuration (more on this later). As such it offers much more than just basic, low-cost Ethernet in a compact format. Indeed, 3Com offers another model in the range � the 4300 � which is designed to do just this.
The 4400 also directly supports 3Com�s NBX LAN telephony solution with integrated prioritisation support for this application if you have it on your network. In addition to 802.1p support, the 400 has a number of integrated features designed to increase performance across the network, as follows: Aggregated Links The 4400 provides support for up to four Ethernet ports to be aggregated to create a �fatpipe� between switch units. In full-duplex mode this means that an 800Mbps connection can be created. As well as providing dedicated, high-bandwidth links between specific switches, they also provide redundancy so if one link is broken the other links share the burden of moving traffic that was intended to be over the link that is now down. Broadcast Storm Control Within the 4400 broadcast storms are controlled by using a system that monitors the level of broadcast traffic on a particular port. If a pre-defined threshold is exceeded, in terms of frames per second, broadcast traffic on that port is blocked until it drops below the defined level. One example of how this can prevent a network becoming overloaded is in the event of a problem with a NIC or an IP address mis-configuration. Multicast Filtering With multicast filtering you can restrict the amount of traffic being broadcast by creating a subset of client nodes which specifically receive a particular broadcast. An example of this use would be in a video- conference situation, where a general broadcast would seriously impact on network bandwidth availability. To automate multicast filtering, the 4400 supports the industry-standard IGMP filtering using �snooping� where it taps into conversations between a endstation and an IGMP-enabled device to detect whether that port needs to join a multicast group or not. Smart Autosensing While all the 10/100Mbps ports on the 4400 are configured to support speed auto-detection as standard, 3Com has gone one step further here and added smart autosensing. After initially negotiating with the end port � for example a PC or server � to detect the correct speed and duplex mode, the link is then monitored. If a high packet failure or error rate is detected, or if the initial detection is deemed to be incorrect, the link speed will be adjusted accordingly. VLAN Support In line with most switches nowadays, the 4400 supports VLANs (Virtual LANs) and specifically IEEE 802.1q VLAN tagging. This standard allows traffic from multiple VLANs to be carried across a single, physical link and provides interoperability in multi-vendor networks. VLANs are especially useful at controlling broadcast traffic, as they limit the number of nodes within a virtual segment of the network, so in conjunction with the broadcast storm control (see above) create a powerful tool for preventing excess broadcast traffic from slowing the network. Configuring and Managing the 4400 The 4400 is designed to be used straight out of the box with a minimum of � or even no � configuration, but in order to take advantage of its feature set, a number of different options are provided. As standard the product ships with integrated, web-based management, a console based alternative and 3Com�s Network Supervisor management software on CD in version 3.0. Both SNMP and RMON (first four groups only) are supported by the switch. In some cases actions can be carried out on any of the three �platforms� but equally sometimes a specific tool has to be used to perform a particular task. During general use, if the switch is visible, a number of LEDs on the front panel indicate port status, unit status and module status for any modules you have installed. These simply slide into a slot at the rear of the switch with hand-tightening screws � no screwdrivers required! The LEDs indicate traffic being transmitted/received, the duplex mode in operation and the speed of the link in normal operation. A separate batch of LEDs shows the general status of the switch, in normal operation, during initialisation or software download mode, and whether it has failed its POST (Power On Self Test). The first configuration step most users will want to carry out is to assign an IP address to the switch. The easiest way to do this is to use console-based access, via a PC and a serial connection into the device. Terminal configuration settings are provided in the manual. On connecting, a simple menu is presented, from which you can control all aspects of the 4400. For anyone familiar with 3Com products over the years, the menu structure and operation is essentially unchanged, but with some additional commands and command groups. General options are: bridge, features, logout, physical interface, protocol, security, system and traffic management. Beneath each of these options lie one or more sub-menus. For example, within the bridge sub-menu you can configure broadcast storm control, link aggregation, multicast filters, resilient links and VLANs, as well as administering the address database, bridge ports and enable/disable Spanning Tree. In all cases there are shortcuts for typing in the commands such as �br� for bridge� so, though text-based, navigation around the menu system is fast and the preferred choice still of systems engineers � the guys most likely to be configuring the switch, at least initially.
While several of these menu options are replicated in the web-based manager, there are still many � such as setting up VLANs � that must be done at the console level. This is not uncommon among switch vendors but does render the Web-based management somewhat redundant unless you really prefer to work with a browser. One reason why you might want to use the web-based manager is that it enables at least an element of point �n� click, mouse-based configuration. The switch is presented in graphical format onscreen with a menu on the left-hand side of the browser. So, for example, to configure the basic setup of a port you simply click on the graphic of it, choose the option you want from the drop-down menu that appears and type in the required information. Using the �Explorer� type menu on the left involves clicking on an option to expand it and reveal more sub-menu options which are then accessed by another click. Typically you are then presented with a text box or check list, depending on the nature of the operation. However, the overall appearance of the browser-based manager does leave a little to be desired even if it is as good as that provided by most of the opposition. The main problem is that it does not allow you to configure the full range of 4400 options, notably the QoS parameters, though Network Supervisor (see below) does this job instead. Managing with Network Supervisor The third management option provided with the 4400 is 3Com�s Network Supervisor 3.0, Windows-based software, effectively the next generation of Transcend for those familiar with previous 3Com offerings. Unlike the console and browser options which are geared specifically to configuring the 4400, Network Supervisor is a generic SNMP manager, with some added functionality specifically for controlling 3Com products, such as the 4400.
The first thing you have to do with Network Supervisor is to let it auto-discover the network � identifying all devices on the network and how they are inter-connected. Once the map is created you can monitor devices and links as well as accessing the 3Com switches directly via the Telnet or web-browser options we�ve already described. Network Supervisor can also be used to download software updates for the switches and upload them directly onto the devices. Discovering the network is achieved via a wizard-based routine that prompts you for some basic discovery information such as local subnet only or all subnets reachable, which SNMP community strings to search for and whether IP telephony nodes should be included. There is also an option to lightly stress monitor the network from the moment the map is created. In most cases, simply clicking through the screens and accepting the default options would suffice. The discovery process then creates a network map, that is a visual representation of your network. This is a quick process (a few minutes maximum) for a relatively small network but could take some amount of time on a larger network, though it is limited to 1500 network nodes (plus 1500 IP telephony nodes). To keep your map up to date, you can manually add and remove network objects, or rediscover it to update the map automatically with any changes that have occurred from an option within Tools on the menu bar that sits at the top of the screen. Within a map you can also group devices, so create a realistic facsimile of a departmental layout, say, in a real office. An online Help system is always available and contains a glossary of networking terms, useful for less experienced users. Once the network is discovered, the screen splits with network node addresses (these may be the device name, DNS name, IP address or MAC address, for example, depending on what Network Supervisor was able to discover about that particular device) on the left and the network map on the right. Right-clicking on a device name/address reveals a drop-down menu where the number of options vary depending on the device. For example, if it is a 3Com switch such as the 4400, there are options to Telnet in the device or web manage it, as well as a number of standard options such as showing the monitored state of the device, an option to set alerts and viewing any �events�, such as thresholds being exceeded.
From the network map, you can select any device or link you wish to monitor. Each device or link has various monitors, each tracking a different aspect of the object's performance both internally and on the network. By then right-clicking on the device Network Supervisor displays stress bars that can be updated regularly. Each monitor has a threshold value in Network Supervisor. If the monitor's value exceeds the threshold, it indicates that monitor has detected an abnormal state. The default setting for the threshold should be suitable in most networks but you can change this, if required, using the Threshold Settings dialog box. When a monitor then exceeds its threshold, an event is generated, though this option can be disabled for a specific device or link. These events are then logged in an event list that can be viewed as explained earlier. It may also trigger an alert. Alerts can be set up in three ways: as an email, a pager call or a screen popup message. Configuring QoS Parameters on the 4400 QoS or Quality of Service is defined by 3Com as �advanced prioritisation� which is as sensible a term as we can think of ourselves. Essentially it moves basic traffic prioritisation on a step by allowing you to create specific service levels and attaching them to particular applications. When you are trying to run a number of very different applications � in terms of style, bandwidth demands and level of importance � simultaneously, applying QoS is the best method to control traffic levels on the network and ensure those applications and services reach the users. Within the 4400, QoS is defined in a number of different ways that combine to create a QoS policy to apply to a specific traffic type, protocol type or IP/MAC address. These are as follows: Classifier This is used to classify traffic on the network by variables such as protocol, application, source and destination. The 4400 then groups classified traffic in order to schedule it with an appropriate service level (see below). DiffServ Code Point (DSCP) The DiffServ or Differential Service is the traffic prioritisation bits within the header of an IP packet that certain applications and devices encode to identify the level of service that packet requires as it travels across the network. Policy As with any policy this is a set of rules that � in this case�� are applied to a network so that it delivers the most critical applications and services ahead of the less important applications. QoS Profile This consists of multiple sets of rules � a classifier plus service level combinations � which are then assigned to a port or range of ports. Rules These comprise a service level and a classifier which define how a 4400 will manage certain traffic types and are always associated with a QoS profile, as described above. Service Level This defines the priority given to a set of classified traffic. There are two very distinct ways of applying QoS to traffic on the 4400. Both produce the same end results and utilise exactly the same code to do so but have a very different appearance during configuration. One option is to use the Console-based management under the TrafficManagement then QoS menu/submenu options.� Here you have to define a service level to determine the priority to be applied to a traffic type, then apply a classifier to determine how the incoming traffic will be classified by the 4400. Then you need to create a QoS profile which associates a service level and classifier combination, then apply a QoS profile to a port or range of ports. Alternatively you can run the Prioritise Network Traffic wizard from within Network Supervisor.
The wizard runs you through a series of screens, identifying what traffic/application types � such as HTTP, SMTP, telnet, SNMP, Lotus Notes, FTP etc � you want to prioritise and which IP addresses on the network you want to prioritise � maybe a web server, for example. It then creates the various policies, service levels and so on, in order the deliver the required QoS parameters you�ve declared. On checking the Console-based management afterwards, within the QoS submenu you can then see exactly what the wizard created, if you really need to see it to believe it!
On completion, the prioritisation wizard also creates a report so you can double check on the settings and make sure you got your priorities right! The important point here, of course, it that it brings QoS to those who do not have networking experience and therefore would find the console-based method of setting it up challenging but understand the importance of delivering applications to the desktop. Overall, the range of options for managing and configuring the 4400 seems to be designed to suit most skill levels, with just the browser-based configurator really lacking in some functionality Performance and Resilience testing In order to fully test the capabilities and resilience of the 4400 we took a stack of two units (48 ports) and ran them without interruption in the offices for five weeks. Other than confirming the reliability of the units under heavy use and qualifying basic throughput, we wanted to see how well the integrated traffic prioritisation works within the 4400�s. So during this period we ran a number of different benchmarks and traffic generation devices against the switch stack in a number of different configurations. The basic labs setup we used was a combination of six web servers � NT4/IIS4.0 and 20 physical PC clients (NT4) running multiple virtual PC clients � up to 600 in total. With these we ran a series of benchmarks developed by Microsoft � the WCAT suite � for web server benchmarking. The benchmarks enabled you to specify traffic and content types so are ideal for testing how well traffic prioritisation on application type works. In addition we used Spirent Communication�s SmartBits 6000 chassis to generate heavy traffic and additional benchmarking. The chassis was configured with four 10/100 Ethernet modules which we connected directly to the 4400 stack and ran benchmarks across the two switches, ensuring the traffic had to pass across the stack�s backplane during all the benchmarks. We also used Chevin Ltd�s LAN Loader utility (part of its� TeVista network management software suite and a utility which has brought Layer 3 and 4 switches to a standstill in the past) and Socrates, a public-domain denial of service attack simulator to further attempt to overload the 4400 stack. The first point to make is that, despite attempts to kill the 4400�s we never had to reboot or reconfigure (from default) the units once during the time they were in our labs � impressive stuff for a brand, spanking new product with only a few kilobytes under its belt.... With the SmartBits we looked at how the switches would behave over elongated periods of very heavy traffic.
To achieve this we ran
SmartBits� SmartApplications Throughput test. We configured 24 SmartBits
ports in 12 pairs � each pair using a port on each of the two 4400�s
in the stack � and set up a test to run overnight sending a range of
frame sizes (64 bytes � 1344 bytes) between the port pairs and therefore
straight across the 4400 stack backplane. This means that up to 2.4Gbps of
traffic was attempting to pass across the stack. Overall, the throughput
result averaging around 82% of traffic passed is impressive. We repeated
the test just using the 64byte packet size as this produced the lowest
figure and on the second run it too averaged 82% passed traffic so maybe
there was an unidentified problem with the benchmark setup on the first
run. We then ran a WCAT test which sent a mix of different http requests to a bank of six IIS 4.0 web servers, so the WCAT traffic was effectively prioritised. 20 PC clients each ran 30 virtual clients creating a total of 600 virtual clients. We initially ran this test in isolation, then added firstly the SmartBits throughput test (as defined earlier), then combined background file transfer from a bank of three Windows 2000 clients � aimed at the same bank of web servers � then combined all three types of traffic. Finally we added Socrates and Chevin�s LAN Loader to the mix�
Scenario 1 is WCAT only. Scenario 2 is WCAT plus SmartBits Throughput Scenario 3 is WCAT plus File Transfer Scenario 4 is WCAT plus SmartBits Throughput and File Transfer Scenario 5 is as above plus Socrates and Chevin LAN Loader As you can see, the main hit came when, interestingly, generic file transfer � effectively a simulation of backup � was running alongside the prioritised WCAT traffic. However, throughput overall remained extremely high considering that the switch was seriously over-saturated in terms of the data being thrown at it. In the end it seems that the �traffic generators� fought with each other and the WCAT traffic was able to get through relatively easily. Overall we were impressed with the resilience of the switch stack � the important point being that it continued to service applications at a very healthy level of performance even when being bombarded with data from all angles. There is always talk of mission critical applications and high-priority services and much of this talk is associated with the big financials and corporates. But the truth is that to any company of any size, their particular application and service requirements are just as critical to themselves. So it is good to see features such as traffic prioritisation and resilience being brought down to the level where just about any company can afford to network intelligently and really attempt to maximise their IT investment. No one can afford nowadays to simply blindly invest in networking hardware and expect it to be self-justifying. The truth is that IT expenditure always should have been accountable but often hasn�t been. However, at �50 a port for intelligent networking there�s a fair argument to be made for justification of purchase in this instance. As such we have no hesitation in awarding the SuperStack 3 4400 Ethernet switch our NSS Approved award. Bandwidth or intelligence? Intelligence wins every time�
Contact:
3Com UK ��� Appendix A: Spirent Communications Smartbits 6000B Spirent Communications� SmartBits 6000B (SMB-6000B) chassis is claimed to be the industry �s highest-port-density network performance analysis test system. Each chassis can support up to 24 Gigabit Ethernet ports, 96 10/100 Mbps Ethernet ports, 12 POS (Packet over SONET) ports, 12 SmartMetrics (more on this later) Gigabit Ethernet ports, or a mixture of these port types. SmartBits 6000B �s can be daisy-chained together to achieve very high port density, enabling users to perform automated large scale testing in Quality Control and high-volume production environments. The SMB-6000B is controlled by a PC through a 10/100 Mbps Ethernet connection and uses a Windows-based interface. On this platform, Spirent has made a wide range of applications available to use with the 6000B covering everything from classic throughput testing to complex QoS/CoS tests. The latter is achieved with the SmartMetrics modules, offering a layered approach to performance, designed specifically to address the often-complex set of optimization and prioritization methods at all network layers. It offers the ability to measure: Virtual Local Area Network (VLAN), IP Type of Service (TOS), DiffServ, Multi-protocol label switching (MPLS), voice, and multimedia traffic flows. Calls and connections (signaling functions). Network applications (management capabilities). Transition of the data between different technologies. SmartMetrics gives you the ability to measure and analyse every aspect of your network, from the performance of each network port, to the performance of millions of IP flows, to the effect of opening and closing thousands of TCP or Multicast sessions. The optimization of traffic can take place at layer 2, layer 3, layer 4, and even up to layer 7. IEEE 802.1p prioritizes traffic at layer 2 (Datalink). DiffServ and TOS optimize and MPLS and RSVP manage resources at layer 3 (Network). A multitude of new QoS products, such as server load balancers and traffic shaping/access control services, optimize traffic based on criteria including TCP or User Datagram Protocol (UDP) port number at layer 4 (transport).� Even higher-layer criteria such as Uniform Resource Locator (URL) or application type can be used so it is vital to be able to measure performance in these areas. SmartMetrics addresses these needs by measuring how well devices and networks optimize, prioritize, and segment traffic using an expanded set of metrics, based on per-flow, connection, network application and access device and cross-technology. Spirent�s complete library of existing test applications as follows: SmartFlow Tests line rate QoS. Enables both forwarding and policy tests. Analyses each incoming stream to test a device's (or network's) ability to forward very large numbers of flows. Analyses the device's ability to correctly handle policies implemented in the network or device under test. Script Automation Interface SAI is a platform and language-independent Script Automation Interface tool designed to simplify the testing of Frame Loss, Latency, Throughput and Jitter. Its easy to use, text based layout makes it the ideal tool for users looking for simple, quick tests that are repeatable, scalable and allow for performance testing with very little knowledge about the application. SAI is a stand-alone application that it is fully supported by the SmartBits scripting tool ScriptCenter (see below) and can be invoked by other scripting languages like Perl.� DOCSIS Acceptance Test Suite Helps vendors verify the forwarding capability of a Cable Modem (CM) or Cable Modem Termination System (CMTS) and verify products for DOCSIS and EuroDOCSIS certification. DOCSIS ATS is based on the SmartBits� network performance analysis platform. AST II AST II is Spirent Communications' second generation of Advanced Switch Tests. Based on RFC 2285 and the IETF Switch Methodology Draft, AST II provides the TRUE first-level benchmark for all switches. AST II tests include Forwarding, Congestion Control, Address Learning Rate, Address Caching, Error Filtering, Broadcast Forwarding, Broadcast Latency, and Forward Pressure.� ScriptCenter A platform-independent Tcl/Tk visual scripting tool designed to reduce overall script development time. Enables the quick setup of tests or the use of predefined scripts provided with the software. Script code is automatically generated, based on GUI input at three different levels. Runs in Windows�, UNIX�, or Linux environments. Spirent claims it is suitable for either the non-programmer or the experienced programmer.� SmartApplications Provides automated performance analysis for bridges, switches, and routers per RFC 1242, Benchmarking Terminology for Network Interconnection Devices and RFC 2544, Benchmarking Methodology for Network Interconnect Devices. Tests are available for Ethernet, Token Ring, ATM, and Frame Relay.� SmartCableModem Test Provides cable modem performance testing and quality analysis. Designed for cable modem manufacturers, CMTS (Head End) manufacturers, MSOs, and system integrators. Used to evaluate the key performance parameters of cable modems, head ends, and cable plants under typical or extreme traffic load conditions; to perform comparative analysis of cable modems and cable plants; and to re-qualify cable plants after firmware upgrades.� SmartMulticastIP Measures IP multicast performance of routers and switches. Designed for network managers, network equipment manufacturers, ISPs and carriers. Used to perform a comparative analysis of IP multicast devices, to evaluate key performance parameters of IP multicast devices under typical or extreme traffic load conditions, and to re-qualify IP multicast devices after firmware upgrades. SmartLib Programming Library Helps developers create custom applications for testing networks and network devices using the SMB-200, SMB-2000, SMB-600, and SMB-6000B chassis. Use SmartLib to automate complex suites of tests, create simplified GUIs specifically tailored for a production line, and customize tests for unique network components.� TeraRouting Tester Spirent believes this is a highly scalable router performance tester. Provides the first integrated control and data plane routing test that includes system-level configuration and analysis of routed networks. Allows network service providers and network equipment manufacturers to precisely measure the performance of QoS-enabled BGP4 routers in a variety of scenarios. Uses a deterministic and repeatable test methodology.� MPLS Performance Suite Provides tools for assessing the viability, performance impacts, and scalability of MPLS solutions. Enables IP packet forwarding that supports sophisticated packet classification and high-rate data forwarding. Tests performance, load balancing, QoS, and fast failover. Used to accurately recreate the complexity and dynamic nature of MPLS traffic and traffic management prior to deployment on live networks.� SmartSignaling Measures the capabilities of ATM switch devices and ATM/LAN edge devices to accept calls and to set up and tear down switched virtual circuits. Its two pre-programmed tests test call capacity and peak call rate. Supports all SmartBits� ATM SmartCards.� SmartTCP Tests load balancer performance. Designed for network managers and network equipment managers, ISPs, and web hosting firms. Tests measure the TCP session performance of server load balancer devices that make forwarding decisions based on Layer 4-7 information. SmartTCP benchmarks both the rate and connection capacities of the device under test to establish, maintain, and tear down TCP sessions.� SmartVoIPQoS Voice over IP network testing. Allows the user to fully stress a network and analyse the network's ability to deliver both voice and data. Designed to evaluate, qualify, and perform comparative analysis of QoS-enabled routers and IP PBXs, and to qualify a network for delivering VoIP services. All results are formatted to speed user interpretation. SmartWindows SmartBits� virtual front panel. Allows the user to access SmartBits� equipment with greater test control than a pre-programmed application. Used to verify design, improve product quality, perform low-volume production and repair testing, and perform competitive marketing analysis. Within SmartWindow, simply select a protocol, set class of service parameters, and then test any of the following: NIC cards, servers, bridges, cable modems, xDSL modems, switches, routers, VLANs, firewalls, live networks, or multimedia scenarios.� SmartxDSL Provides xDSL performance testing and quality analysis. Designed for use with xDSL modems by DSLAM and solutions manufacturers, carriers, and system integrators. Allows users to stress an xDSL network to its full capacity. Provides stability, frame loss, and latency measurements at full wire rate. Minimizes the time required to configure tests and analyse results. VAST Tests and verifies the performance of next-generation Ethernet switches, routers, and VLANs. VAST provides a 10/100 Mbps, full/half duplex, stream-based set of tests. Tests measure throughput and packet loss for different traffic patterns.� WebSuite/Firewall Designed to simulate real-world traffic loads in order to support the testing of next-generation content delivery and network equipment. Gauges the performance of a firewall's performing NAT (Network Address Translation). Determines maximum application transaction capacity. Measures application throughput with TCP acting as the transport agent. Evaluates a firewall's ability to deal with DoS (Denial of Service) attacks before deployment.� TeraVPN Spirent claims this is the industry's first test solution for measuring the network performance of IP Virtual Private Networks. Determines IP-VPN tunnel creation capacity using IPSec protocols. Generates UDP or HTTP traffic over each tunnel and measures data performance characteristics like packet loss, latency, and response time. Displays results in an efficient layout for quick performance benchmark comparisons. Can be used to troubleshoot interoperability issues, with bounce diagrams showing the messages being exchanged
|
 |
Send mail to webmaster
with questions or�
|
