Comparison MikroTik hEX RB750Gr3 vs MikroTik hEX RB750Gr2

How the router connects to the Internet or other external network.

Almost all modern routers have ethernet network connectors for this purpose, however, in addition to them, other connection options can be provided — both wired ( ADSL, SFP / SFP + optics) and wireless (mobile access via 3G / 4G modem or SIM card). Here are the features of each option:

— Ethernet. A standard LAN network cable connector (“twisted pair”) is the most popular modern wired connection format in computer networks. Widely used both in "local" and to provide access to the Internet. This standard is somewhat inferior to SFP / SFP + (see below) in terms of speed and noise immunity, but it is much cheaper. The speed of work in modern versions of Ethernet can reach 10 Gbps (see "Connection speed of WAN ports"), theoretically, a further increase in throughput is possible.

— SFP / SFP + (optics). A connector for transmitting network traffic over a fiber optic cable. The main advantage of such a cable is complete insensitivity to electromagnetic interference. And data transfer rates can reach 2.7 Gbps in the original SFP and 16 Gbps in SFP+. At the same time, maintaining this standard is not cheap, and the benefits mentioned are not often needed in fact. Therefore, SFP / SFP + is found mainly...in mid-range and top-level routers.

— ADSL. Connecting to the Internet through a fixed telephone network using ADSL technology. The key advantage of this connection is the ability to use existing networks without laying additional wires; at the same time, Internet access is completely separated from telephone communication and traffic does not interfere with voice calls. On the other hand, the bandwidth of ADSL is very low by modern standards (less than 24 Mbps), moreover, the data transmission speed is noticeably lower than the reception speed. This can create problems for video communication and some other specific tasks. So nowadays ADSL is used less and less.

— 3G/4G modem (USB). Internet connection via mobile network using a separate 3G or 4G modem connected to the USB port. This feature can be useful where there is no full-fledged wired connection (for example, in rural areas), and also as a fallback option in case the main communication channel fails. And the type of network supported depends mainly on the modem used (the compatibility of the router with different models needs to be specified separately, but most often there are no problems with this). As for specific types of networks, most 3G modems work in UMTS networks (the same ones that are massively used by mobile phones); the data transfer rate in such networks can reach 75 Mbit / s (however, usually it is much lower). Less common are 3G modems for EV-DO networks based on CDMA — this standard has lower speeds (up to 14.7 Mbps) and not as extensive coverage as UMTS, however, both the equipment and the connection itself can be cheaper. And the designation "4G" means only one type of networks — LTE; it provides speeds up to 173 Mbps, but is not as widespread as 3G.

— SIM card. Another option for connecting to the Internet via mobile networks is its own SIM card slot provided in the design of the router. This option is convenient because you do not need to buy an additional device (modem) for mobile Internet — you just need to purchase an operator's SIM card. On the other hand, due to the built-in mobile communication modules, such routers themselves are more expensive than analogues for USB modems. In addition, the connectivity options in them are limited by the characteristics of the module: for example, a router for 3G networks will not be able to fully use 4G networks (whereas a USB modem can usually be changed to a more advanced one). As a result, this option is relatively rare in modern equipment.

The number of USB ports provided in the design of the router. USB is a universal interface used in computer technology to connect peripheral devices for various purposes. One of its uses is to connect a 3G modem (if provided, see “Data entry (WAN-port)”), but this is not limited to this: in fact, the options for using USB ports directly depend only on the capabilities specified in the router firmware. So, in some models, you can connect a flash drive or an external hard drive to this connector — to update the firmware or even organize an FTP server directly on the router. And for other peripherals — printers, cameras, etc. — USB connection can be used to work as network devices (however, such functions are rare in standard firmware).

The basic capabilities of the router — that is, the functions directly related to the work for the main purpose. The most common of these features are DHCP server, load balancing, link reservation, port forwarding, MAC address cloning, VPN and DDNS support. Here is a detailed description of each item:

— DHCP server. A feature that makes it easy to assign IP addresses to devices connected to the router. An IP address is necessary for correct operation in TCP / IP networks (and this is the entire Internet and the vast majority of modern “local sites”). In the presence of DHCP, this process can be carried out completely automatically, which makes life much easier for both users and administrators. However, the administrator can also set additional DHCP options — for example, specify a range of available IP addresses (to prevent errors) or limit the time of using one address. If necessary, you can even manually enter a specific address for each device on the network, without automatically adding new devices — DHCP also simplifies this procedure, as it allows you to carry out all operations on the router without delving into the settings of each subscriber device.

— Load balancing. A feature found on mo...dels that have two or more channels to connect to the Internet (and other external networks); most often these are two or more WAN ports, but there is another option — one port, supplemented by support for 3G / 4G mobile networks. Anyway, the idea of balancing is to use several channels for external connection at the same time, dividing the load between them in one way or another. This allows you to increase the efficiency of the connection, achieving maximum data transfer speeds and at the same time avoiding unnecessary congestion. For example, a channel for online games can be separated from the rest of the connection, minimizing lags and reducing the likelihood of failures. As for load distribution, it can be either automatic (when the router itself determines the optimal channel for each device, depending on the current traffic consumption), or manual (when specific channels are prescribed for different network devices, applications, or even types of traffic).

— Channel reservation. Another feature related to the simultaneous use of several channels to connect to the Internet (or other external network). In redundancy mode, the router constantly uses the main channel (or several channels) for external connection, and in case of failures on this channel, it automatically switches to the spare (spare). This relieves the administrator of the need to manually organize the connection when the main connection fails; and the spare channel works only when it is indispensable, which in some cases allows you to avoid unnecessary costs. A typical example of working with redundancy in everyday life is the use of a wired Internet connection as the main channel and a 3G / 4G modem as a spare; although, of course, other, more specific options are possible.

— Port forwarding. The ability to redirect traffic from the router's own ports to the address of a specific computer (or other device) on the local network. When working in this mode, such a computer “outside” will look like it is connected to the Internet directly, without a router. This mode may be needed to use some specific functions — for example, working in HTTP server mode or participating in P2P networks.

— Cloning the MAC address. The ability to copy the MAC address of one of the devices connected to it to the router — so that when accessing the router, it is the address of this device that is visible, and not the router itself. The MAC address is a unique identifier assigned to each device with a WAN port. And the need to clone this identifier arises from the fact that some Internet providers use not only a login / password, but also the MAC address of a specific computer connected directly to the network to authenticate users. If, however, such a computer is supplemented with a router, then the provider's equipment will see a new, unfamiliar device, and will not give access to the network. Cloning the MAC address allows you to fix this situation as quickly and simply as possible.

— VPN support. Support for the VPN function of the router — virtual private networks. One of the key principles behind this feature is the transmission of encrypted data over open networks, primarily the Internet. A VPN is used mainly in two formats:

• Creating virtual networks based on an Internet connection. Thus, it is possible, for example, to combine branches of one company located in different cities or even countries into one logical network. At the same time, thanks to traffic encryption, the entire network remains closed to outsiders, although data is transmitted over an open channel. For this format, devices of the Firewall type (see "Type") are mainly used, while such a device actually plays the role of a VPN server.
• Internet connection through an external VPN server. The functions of such a server are in many ways similar to a proxy: it serves as an "intermediary" in the exchange of traffic and replaces the user's IP address with its own address. The latter, in particular, allows you to bypass regional restrictions: nowadays, servers with IP addresses related to almost any country in the world are available. However, a VPN server, unlike a proxy, additionally encrypts the traffic transmitted to the user — this, again, has a positive effect on security and privacy. This mode is also available in regular routers.

Note that the connection to the VPN server can also be “raised” on individual devices on the network (for example, through tools in some Internet browsers). However, using this feature on a router is often more convenient: you only need to set up a VPN once, you don’t have to fiddle with options for each individual subscriber, and any network device can use such a connection (including those that do not have their own VPN tools) . On the other hand, the connection speed when working through a VPN can drop noticeably, and enabling and disabling this feature on a router is usually more difficult than on user devices.

— DDNS. Short for Dynamic DNS — "dynamic DNS". This feature allows you to assign a permanent domain name to a device with a dynamic IP address. A domain name is the name of a device on the local network or the address of a site on the Internet (for example, m.ua or e-katalog.ru). An IP address is service information in the form of a digital code; it is thanks to her that network equipment can find the desired device and issue the required data from it. Actually, IP is the primary network "coordinates"; however, remembering addresses as a sequence of numbers is quite difficult, so domain names appeared — they are much more convenient for a person. Both on the Internet and in local networks, the connection between a domain name and an IP address is responsible for the so-called DNS servers: for each domain in the database of such a server, its own IP is registered. However, for technical reasons, situations often arise when the router has to use a dynamic (changing) IP; accordingly, in order for information to be constantly available on the same domain name, it is necessary to update the data on the DNS server with each IP change. It is this update that the DDNS function provides.

The model of the processor installed in the device. The processor is responsible for processing network traffic and running software. Knowing its name, you can get more detailed data on the speed capabilities of the equipment and understand how much such a powerful or, on the contrary, mediocre element is needed on board. In new models of Wi-Fi equipment, coprocessors or so-called NPU modules are often installed, which relieve the load from the main processor.

The number of cores in the processor installed in the device. The core in this case refers to the part of the processor that executes one thread of instructions. Accordingly, the presence of multiple cores allows you to work with multiple threads simultaneously, which has a positive effect on performance.

The number of cycles per second that the processor produces in its normal operating mode. A clock is a single electrical impulse used to process data and synchronize the processor with the rest of the computer system. Different operations may require fractions of a clock or several clocks, but anyway, the clock frequency is one of the main parameters characterizing the performance and speed of the processor — all other things being equal, a processor with a higher clock frequency will work faster and better cope with significant loads.

The amount of random access memory (RAM) provided in the device. The amount of "RAM" is one of the indicators of the power of the device: the larger it is, the higher the speed and the better the device will cope with "heavy" tasks.

The amount of memory allocated for the operation of the operating system on board the device. It stores the OS and the control programme. Note that Flash memory is not available for use by the end user.

The standard for the PoE input provided in the device.

By itself, PoE (Power over Ethernet) technology makes it possible to transfer not only data over an Ethernet network cable, but also energy to power network devices. And the presence of a PoE input allows the router itself to receive power in a similar way. Note that there are special devices - the so-called PoE injectors - that allow you to add power to a regular network signal (that is, add PoE support to equipment that does not initially have such a function).

As for the PoE standards, they determine both the power supply and the main possibilities for coordinating the power source with the consumer - both must support the same standard, otherwise normal operation will be impossible. At the same time, formats that are marked like “802.3*” are called active; their common feature is that when a load is connected, the power source first “interrogates” it, checking whether the powered device complies with the requirements of the corresponding standard, and if so, what kind of power should be supplied to it. There is no such feature in the passive standard. And here is a more detailed description of specific options:

— 802.3at. A standard originally released back in 2009 and known as PoE +, or PoE type 2. The standard power received at this input is 25.5 W, with a voltage of 42.5 to 57 V and a pair power of up to 600 mA.

— 802.3af/at. This...marking means that the PoE input supports both the 802.3at standard described above and the earlier 802.3af (PoE type 1). The second format is noticeably more modest in terms of capabilities: it provides power at the power input up to 13 W, input voltage 37 - 57 V and power in a pair of power wires up to 350 mA. Despite their "venerable age", many devices with 802.3af outputs are still in use today; so for the power input of the router, compatibility with this standard may not be superfluous. We only note that 802.3af covers as many as four so-called power classes (from 0 to 3), which differ in the specific number of watts at the output and input. So when connecting power from a device with this PoE standard, it does not hurt to further clarify compatibility by power class.

— Passive. The most simple and inexpensive standard, designed to be used mainly in entry-level equipment (since the implementation of active PoE standards is generally expensive). As mentioned above, the key difference from the formats described above is that the power supply supplies energy "as is" - with a strictly fixed voltage and power, without checking the characteristics of the load and without adjusting to it. This is what ensures low price and availability. On the other hand, when using a passive PoE input, care must be taken to ensure that the voltage and power of the power supply match the characteristics of the router; and such coordination can be quite difficult in light of the fact that the passive standard does not have strictly defined standards even for voltage, not to mention power. At the same time, the inconsistency leads to the fact that in the best case (if the output voltage / power is lower than tech required for the load), the power simply will not work, and in the worst case (with excess voltage / power), there is a high probability of overloads, overheating, and even breakdowns with fires - moreover such troubles may not occur immediately, but after a fairly considerable time. So you should pay attention to this option first of all in cases where simplicity and accessibility are more important than advanced power supply standards. At the same time, we note that some switches, which, in addition to the passive input, also have a passive PoE output, allow “cascade” connection - in the form of a serial chain of several devices powered by one external source (the main thing is that this source has enough power).

Separately, we emphasize that you should not try to connect an active power source to a passive input, and even more so vice versa. In the first case, the device simply will not pass the test that is carried out before power is applied, and the power will not turn on. And in the second case, serious failures and even accidents are possible: a passive power source supplies energy immediately, without checking the characteristics of the powered device, which creates the risk of overloads if the operating parameters do not match.