In this block we will tell you the difference between a router and a switch. They look similar – a box with lots of inputs and outputs, but their operation is based on different principles.
- Router
- Operating Principle
- Routing Table
- Use
- Applications
- Is there a difference between a router and a router
- What should I choose?
- Lesson four: The internal structure of the router must have an even distribution of bandwidth, because we cannot predict the distribution of traffic.
- Lesson Five: Routers should not have a single point of failure
- Router
- Firewall
- Advantages and disadvantages of a router
- Which router to choose
Router
router (prof. jarg. router, router (from Eng. router /ˈɹu:tə(ɹ)/ or /ˈɹaʊtəɹ/ [1] , /ˈɹaʊtɚ/ ) or router (A pronunciation of the word English. router (as transliterated)) – router is a specialized network computer, which has at least two network interfaces and forwards data packets between different network segments, making forwarding decisions based on information about the network topology and certain rules set by the administrator.
Routers are divided into software and hardware. A router works at a higher "network" layer 3 of the OSI network model than a switch and a network bridge, which work at layer 2 and layer 1 of the OSI model, respectively.
Operating Principle
Normally the router uses the destination address specified in the packet data and determines from the routing table the path to send the data. If no route is specified for the address in the routing table, the packet is discarded.
There are other ways to determine the route for forwarding packets when, for example, using the sender's address, the upper layer protocols used and other information contained in the network layer packet headers. Often routers can translate sender and receiver addresses, filter the data flow in transit based on certain rules in order to restrict access, encrypt/decrypt the data transmitted, etc.
Routing Table
The routing table is the information on the basis of which the router decides whether to forward packets or not. The table consists of some number of entries – routes, each of which contains the recipient's network address, the address of the next node to which packets should be transmitted and some record weight – a metric. The metrics of the entries in the table play a role in calculating the shortest routes to different recipients. Depending on the router model and routing protocols used, the table may contain some additional service information. For example:
192.168.64.0/16 [110/49] via 192.168.1.2, 00:34:34, FastEthernet0/0.1 where 192.168.64.0/16 is destination network, 110/- administrative distance /49 - route metric, 192.168.1.2 is address of next router to transmit packets for network 192. 168.64.0/16, 00:34:34 - time when this route was known, FastEthernet0/0.1 - interface of the router through which to reach the "neighbor" 192.168.1.2.
The routing table can be created in two ways:
- static routing – where entries in the table are entered and changed manually. This method requires administrator intervention every time there is a change in network topology. On the other hand, it is the most stable and requires a minimum of router hardware resources to maintain the table.
- dynamic routing – When entries in the table are automatically updated using one or more routing protocols – OSPF, IS-IS, BGP, etc. In addition, the router builds a table of optimal paths to destination networks based on various criteria – the number of intermediate nodes, bandwidth, data transmission delays, etc. The criteria for calculating optimal routes most often depend on the routing protocol and are also set by the configuration of the router. This method of table construction allows you to automatically keep the routing table up to date and calculate optimal routes based on the current network topology. However, dynamic routing places additional stress on devices, and high network instability can lead to situations where routers fail to synchronize their tables, resulting in inconsistent information about the network topology in different parts of the network and loss of transmitted data.
Use
Routers help to reduce network load by dividing the network into collision domains and broadcast domains, as well as by packet filtering. They are mostly used to join networks of different types, often incompatible architectures and protocols, for example to join local networks of WAN-connections using protocols xDSL, PPP, Frame relay, etc. Often the router is used to provide access from the local network to the global Internet, acting as a firewall as well.
A router can be either a specialized (hardware) device (a typical Juniper representative) or an ordinary computer that acts as a router. There are several software packages (mostly based on Linux kernel) that can turn a PC into a high-performance and multi-function router, for example GNU Zebra.
Applications
Routers help reduce network load by dividing networks into collision and broadcast domains and by filtering packets. They are mostly used to interconnect different types of networks, often incompatible architectures and protocols, such as local area networks with WAN connections using xDSL, PPP, Frame relay, etc. Often the router is used to provide access from the local network to the global Internet, acting as a firewall as well.
A router can be either a specialized (hardware) device (a typical Juniper representative) or an ordinary computer that acts as a router. There are several software packages (mostly based on Linux kernel) which can turn a PC into a high-performance and multifunction router, for example GNU Zebra.
Is there a difference between a router and a router
Let me tell you straight away, I don't want to pound water in a bowl. There is no difference at all between the two. These two names are synonymous and are used depending on the speaker's preferences. The word "router" is just a transliteration of the English word router, which is translated into Russian as "router". If you think about it a bit, it is not difficult to give several similar examples.
The confusion of terms is most likely due to homegrown "experts". Some argue that a router's mandatory attribute is the ability to connect via Wi-Fi, while a router allows you to connect only by wires. Others boast that a router is such a small home box, while a router is serious equipment of the ISP level. Well, others are confused about the terms, saying that the difference is the number of ports for connecting different peripherals. There are those who confuse everything with everything else. Be that as it may, if you have a need to understand the subject, you should either trust a really savvy specialist, or find the right information on a reputable resource on the Internet, the benefit of this is not difficult.
What should I choose?
Let's get some clarity into all this confusion. In principle, any router is a small, highly specialized computer whose only purpose is to transmit data packets between different networks. The fact that we are talking about synonyms, we have already determined, it is worth mentioning some functional differences between different devices, focused on home use, leaving more serious solutions for system administrators. Routers may differ:
- the presence and features of Wi-Fi modules.
- The number and speed of ports for connecting network wires.
- installed USB-connectors (for example, for printer sharing).
- The presence of GSM-receiver.
- Support for different standards for connecting to the provider's network.
- operation stability (very much depends on the manufacturer and the specific model).
- other differences (there are indeed many, the most obvious ones have been listed above).
When choosing a particular model, first of all proceed from the availability of the options you need. Do not listen to the advice "take the first one you find" because "they are all the same" – it is fundamentally wrong. If you are interested in a model, it is worth before buying to read reviews and feedback about it, because advertising is one thing, but the experience of using it in real life can make changes.
Lesson four: The internal structure of the router must have an even distribution of bandwidth, because we cannot predict the distribution of traffic.
A completely different approach was to create a full NIC-NIC link network and distribute cells across all NICs:
Figure 6 – Full link structure with cell allocation to all NICs.
Despite learning the previous lessons, new problems were discovered. In this architecture, everything worked well enough until the board needed to be removed for repair. Because each NIC contained cells for all the packets in the system, no packets could be recreated when the board was removed, resulting in brief but painful downtime.
Lesson Five: Routers should not have a single point of failure
We even took this architecture and turned it upside down:
Figure 7 – Here, all packets go to the central memory and then to the output NIC.
This system worked quite well, but the problem was memory scaling. You could just add multiple controllers and memory banks, but at some point, the total bandwidth proved too difficult to physically design. Faced with practical physical limitations, we were forced to think in other directions.
Our inspiration came from the telephone network. A long time ago, Charles Close figured out that scalable switches could be made by creating networks of smaller switches. As it turns out, all the wondrous properties we need are present in the Kloz network:
- Power grows with scale.
- Does not have a single point of failure.
- Maintains sufficient redundancy for fault tolerance.
- Handles congestion by distributing the load throughout the structure.
Unfortunately, even this architecture has its problems. The format of the cells used between switches is proprietary and belongs to the chip vendor, making it dependent on chipsets. Dependence on a chip vendor is not much better than dependence on a single router vendor, the problems are the same: tying pricing and device availability to a single source. Hardware upgrades are challenging because the new cell switch must simultaneously support legacy links and cell formats to maintain interoperability, as well as all the link speeds and cell formats of the new hardware.
Router
The very name router encodes the purpose of the router.
In the classical (academic) view, a router is needed to translate packets between separate IP networks. It solves the problem of joining separate LANs and prevents the growth of broadcast traffic in one large local network by segmenting it. Of course to redirect traffic correctly you need to know where to send it, that means to build route (the author thanks Captain Hindsight for the exact wording).
Modern router models operate above layer 3 of the OSI model. In addition to broadcasting IP packets from one network to another, these devices often have traffic management features, such as the ability to close/open TCP or UPD ports, perform Port Address Translation, PAT (sometimes called Destination NAT, DNAT), and so on. Also, some protocols require the router to be able to work as an Application-level gateway, ALG, to enable protocols such as: PPTP, IPsec, RTSP, SIP, H.323, SMTP, DNS, TFTP.
A router can be both an old computer with a configured routing table and a specialized network device that only analyzes simple conditions like ACL lists and forwards packets from one network to another.
In particular, routers as separate devices are used when not only logical (VLAN) but also physical division into subnets is required. For example, you need to separate the campus network, where students live, from the university network, where the research takes place.
Modern LANs often use L3 switches instead of routers as separate devices to manage VLANs and thus separate subnets.
Case in point. A network of a small enterprise, where Cisco Catalyst 3750 was used as the core of the network. According to the security requirements, the access level switches were configured according to the principle: one switch – one subnet – one VLAN. To make the project cheaper, we chose 3Com switches. Simply put, each 3Com was connected strictly in one VLAN and one subnet, and packets between subnets went through Catalyst.
Firewall
The usual set of built-in ITU (firewall) features: antivirus, IDP, application patrol – allows you to check traffic up to OSI layer 7. In addition, there are other control features that conventional routers don't have.
Of course, many firewalls have the standard "gentleman's kit" of a typical router. But the "strength" of the ITU is determined by the availability of filtering and traffic management functions, as well as by the enhanced hardware to implement these tasks.
It is worth noting that the set of filtering capabilities of a particular ITU device does not mean at all: "The more features we were able to 'tweak', the 'better' the firewall". The main mistake would be to focus on the length of the list of all sorts of "features" when buying, without taking into account the specific purpose, design features, performance parameters and other factors. Everything must be strictly dosed and balanced without any bias towards "super security" or "super economy".
And here the network administrator faces his first problem. If for SOHO segment it is not so difficult to formulate a typical set of requirements, then for enterprise segment it requires additional preparation. To better meet the needs of the business, there are different devices – each for its own niche. For example, for a VPN Gateway, the set of security features, of course, plays a big role, but the main task is still the creation and maintenance of VPN channels. An example of such a device is the ZyWALL VPN1000
But for Secure Gateway, all sorts of filters, "Sandbox" and other types of checks are in the first place. The ZyWALL ATP800 is an example of such a specialized device for a higher level of protection.
Advantages and disadvantages of a router
The distribution of Internet from the router to the stationary computers, smart TVs, phones and other devices goes through the wire or via Wi-Fi. This way users create a common network for easy sharing of information with each other and access to the World Wide Web.
A properly configured router has the following advantages:
- The network signal distribution to multiple devices at the same time;
- connection of gadgets by wired and wireless means;
- stability and speed of information transfer;
- no data transmission losses.
The disadvantages of the router include the need for constant power from the mains. In addition, the router should be placed so that all devices fall within the Wi-Fi coverage area. If the house or apartment has a large area, so that the signal from the router does not reach the far corners, you need to use a repeater or other ways to increase the range of the router.
Which router to choose
Each router is a minicomputer of narrow specialization, whose main task is to receive and redistribute data between devices in the network. Different models have features that affect the choice of users. When buying a router for your home, you should pay attention to the following features:
- Availability of Wi-Fi module and its technical capabilities;
- The number of ports for connecting network cables and their speed;
- availability of USB-connectors (to connect a modem or share access to the MFP);
- stability of operation (the specific model and manufacturer play an important role);
- support of different standards for connecting the provider services;
- Availability of GSM-receiver.
There are many differences, so before buying a specific model, you need to make a list of useful options, watch reviews, read reviews, get more information. Do not blindly trust advertising – real experience will more accurately show which device you need to buy.
Read More: