Networking FundamentalsPart of the SolarWinds IT Management Educational SeriesV o l u me 5TCP/IP NetworkingThis paper examines TCP/IP networking with an emphasison protocol layers and the interactions of these layers. It ismeant to be an introductory level paper.
Networking Fundamentals » Volume 5, TCP/IP NetworkingPage 2Table of ContentsSection 1 —The Protocol Wars of the 90’s . . . . . . . . . . . . . . . . . 3Section 2 — Networking Models . . . . . . . . . . . . . . . . . . . . . . . . . 3The OSI Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3The TCP/IP Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Section 3 — IP Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Section 4 — IP Packets and IP Routing . . . . . . . . . . . . . . . . . . . . 7The IP Datagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7IP Routing Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Section 5 — The Transport Layer, UDP and TCP . . . . . . . . . . . 8Section 6 — Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9Related SolarWinds Products . . . . . . . . . . . . . . . . . . . . . . . . . . . 10About SolarWinds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11About the Author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11Copyright 1995–2010 SolarWinds. All rights reserved worldwide. No part of this document may be reproduced by any means nor modified, decompiled, disassembled, published or distributed, in whole or in part,or translated to any electronic medium or other means without the written consent of SolarWinds. All right, title and interest in and to the software and documentation are and shall remain the exclusive propertyof SolarWinds and its licensors. SolarWinds Orion , SolarWinds Cirrus , and SolarWinds Toolset are trademarks of SolarWinds and SolarWinds.net and the SolarWinds logo are registered trademarks ofSolarWinds All other trademarks contained in this document and in the Software are the property of their respective owners.SOLARWINDS DISCLAIMS ALL WARRANTIES, CONDITIONS OR OTHER TERMS, EXPRESS OR IMPLIED, STATUTORY OR OTHERWISE, ON SOFTWARE AND DOCUMENTATION FURNISHED HEREUNDERINCLUDING WITHOUT LIMITATION THE WARRANTIES OF DESIGN, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL SOLARWINDS, ITSSUPPLIERS OR ITS LICENSORS BE LIABLE FOR ANY DAMAGES, WHETHER ARISING IN TORT, CONTRACT OR ANY OTHER LEGAL THEORY EVEN IF SOLARWINDS HAS BEEN ADVISED OF THE POSSIBILITYOF SUCH DAMAGES.Document Revised: Oct 1, 2010
Networking Fundamentals » Volume 5, TCP/IP NetworkingPage 3S ecti o n 1S ecti o n 2The early 1990’s were a difficult time for network engineers. Not onlydid engineers need to be fluent in TCP/IP and Ethernet but there was anequal need to be knowledgeable in Apple Talk, NetBIOS/NetBEUI, IPX/SPX (Novell), DECnet, ARCNET, Token ring, FDDI, and a plethora ofother standards. It was common to see networks with mixed Token ring,Ethernet and FDDI moving traffic for DECnet, AppleTalk, IPX/SPX, andTCP/IP. Had it not been for the emergence of the “World Wide Web”, the“Information Super Highway”, and Fast Ethernet, we might be workingtoday in a very different networking environment today. The WorldWide Web (WWW) and the Information Super Highway are really onlydifferent parts of what we call the Internet today. TCP/IP has been usedfrom the inception of the Internet as a Department of Defense (DoD)and academic information exchange tool. In the mid-1990’s use of theInternet began to expand into the general public as user-friendly toolslike Internet browsers and GUI email became available. The subsequentpopularity of Internet usage resulted to the first real increase in Internettraffic since inception. The chart below (Figure 1) shows Internet usagein this timeframe as measured by Cisco Systems.The OSI model and the TCP/IP model are the prevalent methods to describethe interdependency of networking protocols. Both of these are conceptualmodels only and simply describe, not prescribe how networking can bebroken up into categories.The Protocol Wars of the 90’sFigure 1.Networking ModelsThe OSI ModelFigure 2. OSI ModelThe International Organizationfor Standardization (IOS) is aninternationally representedstandards body founded in thelate 1940’S. The IOS created theOpen Systems Interconnection(OSI) model to describe thevery complex task of movingcomputer data from somethingyou understand on a monitor,to bits on a wire, and back toreadable data on a remotesystem. Although the OSI model(see Figure 2) is the most wellknown ISO standard, the IOSis very active in other areas.Ever mounted an IOS image asa CD? This is an ISO standardtoo. The 7 layer OSI model isdescribed below.Layer 1, The Physical Layerdescribes the hardware used tomove data from one machine toanother. This includes cables, connectors, pin functions, voltage and higherlevel functions such as bit level flow control and signaling (modulation/demodulation). Data in layer 1 is bit level data.As you can see, the trend in traffic growth explodes starting about 1996.In 2010 Cisco expects the Internet to move about 21,380 Petabytes ofdata. The growth rate from 2010 forward is expected to flatten out addingabout 10 Petabytes a year. The increase in public use of the Internet wasfollowed closely by the adoption of corporate Internet usage for low costVPN tunneling, web-based commerce and web-based application providers.Now a great deal of the traffic that was once handled by private WANlinks and snail mail/phone is being sent over the Internet.Because the Internet was built around TCP/IP, there was a lot of pressure tobring the IP stack to the desktop in order to minimize protocol translationsand simplify network implementations and network trouble shooting. Ifthe end user can send Internet data using TCP/IP, the data never needsto be translated to another protocol stack.Novell added the TCP/IP stack to NetWare in 1996 and switched completelyfrom their proprietary IPX/SPX to native TCP/IP in 1998. Microsoft enteredthe server market with a server which only supported NetBEUI. NetBEUIis a non-routable protocol, severely inhibiting adoption of the new server.Microsoft then released NT server 3.1 with TCP/IP. Other vendors followedsuit and by about 1999 TCP/IP had won the protocol wars and was thenative protocol stack for most equipment from then forward.Layer 2, The Data Link Layer describes two primary functions, or sublayers.The lower sublayer, which communicates directly with the physical layeris the Media Access Control (MAC) sublayer. The MAC is responsiblefor supplying orderly access to the network media for the higher layers.This includes sending and receiving data to/from the physical layer,detecting errors from the physical layer and using MAC addressing at anoperational level. The Logical Link Control (LLC) sublayer is responsiblefor MAC address control and forming the layer 2 frame by perpendingthe upper level protocol packet with the layer 2 frame header. Data inlayer 2 is frame level data.Layer 3 is known as The Network Layer. The main responsibility of thenetwork layer is to route traffic from one network to another using networklevel addressing. In TCP/IP networking, this is the job of IP. Routing traffic atlevel 3 includes maintaining and sharing logical routing tables and applyingthe rules of the appropriate routing algorithms. The network layer alsoenacts the QoS requested by the transport layer and fragments packetswhen they are too large to be transmitted by lower level protocols. Eventhough the network layer is responsible for upper layer requests, layer 3communications are connectionless and therefore, unreliable. The OSImodel allows for several NON-TCP/IP protocols at the level includingIPX, AppleTalk, and DECnet. Data at layer 3 is packet level data. We willexamine this layer in detail in sections 3, 5, 6, 7 and 8.Layer 4, The Transport Layer, allows for connection orientedcommunications to other devices at layer 4 and reliable services to upperlayer protocols. This layer can also take advantage of using connectionlesstransport when speed is preferred over reliability. Layer 4 offers bothnative and tunneling services to carry non TCP/IP data over a TCP /IPnetwork. Layer 4 data units are called segments.
Networking Fundamentals » Volume 5, TCP/IP NetworkingLayer 5 is called The Session Layer. This layer watches over thecommunications between to devices and sets the rules for communicationwith other devices at layer 5. Layer 5 has the ability create, reset andterminate layer 5 sessions with other computers.Page 4Figure 4. 7 Layer OSI ModelLayer 6 is The Presentation Layer. Data formatting is the main function oflayer 6. This layer receives data from the session layer in various formatsand reformats the data to a form that can be used by the application layer.When receiving data from the application layer, the presentation layerformats the data for use by lower layer protocols.The dotted lines betweensame-level layers showthat there are messageswithin each layer that aremeant for the matching peerlayer on the other end of theconversation. The only wayto get those messages overto the peer layer is down thelocal stack, transmit, thenup the remote stack. Here(see Figure 5) is what thatmight look like:The Application Layer interacts directly with the user application, ifthat application implements a communication component, such aswireless networking or other LAN services. This involves synchronizinginformation exchange with the application and making the data availableto the presentation layer.Each layer has two jobs to do:1. Communicate to the peer system at the same level. In other wordscomputer 1 will send Transport layer information to the transport layerof computer 2.2. Pass information up and down the stack to the adjacent layers asrequired. Because there is no direct connection between machines atlevel 4, to get the level 4 message to the other machine, level 4 musthand the information down the stack and rely on lower layers to makesure it is delivered.To the left (see Figure 7)is how the 7 layer OSImodel might enable thiscommunication.Figure 5.Since the only real connection between these devices is at the physicallayer, the transport layer on computer 1 requests network level servicesto help move the transport segment to computer 2. Layer 3 on computer1 accepts this segment, encapsulates the layer 4 segments in a layer 3packet by adding the layer 3 header with IP addresses and passes it onto layer 2. Layer 2 accepts the layer 3 packet and appends it with thelayer 2 header giving it a MAC address that can be used at the physicallayer. The packet is then handed down to layer one for transmission onthe media. One computer 2, the bits are received from the media and theinformation is sent up to layer 2. Each layer now makes the informationready for the layer above by stripping of the header for that layer andpassing it up. Below is a hypothetical example of this interaction for twoconnected gaming computers (see Figure 3).Imagine we have two folksFigure 3.playing the very popular videogame Extreme 4-Way StopDilemma over the internet.When a player 1 chooses whichcharacter they want to be, theapplication needs to showplayer 2 that a car, or tree, hasbeen chosen. As there is nodirect connection to the remoteplayer at the application layer,the application must rely onthe protocols in the lowerstacks to get the data to thephysical layer where it can betransmitted. At the other endthe physical layer receives thebits and passes them up thestack until they are readable bythe application. The applicationthen informs player 2 that theworthy opponent has chosen“TREE” and the application marks the TREE choice a taken. Player 2 thenchooses a car and the application shows player 1 that CAR 1 has beenselected using the same methodology as above.The selection of player 1 as TREE starts in the upper left. The applicationhas the “show player 1 choice made” message now for the application layeron player 2’s machine. This avoids the unhappy chance that both playerschoose to be TREE, which severely limits the game’s action potential.Again, this is a hypothetical example of how inter-layer communicationsmight take place using the OSI model.The way that lower layers accomplish providing services for upper layersis by adding protocol headers to all data from upper (higher) layers. Thisheader addition is seen at the transport layer and below, so that is whereI’ll focus. I will also assume TCP/IP will be used.Figure 6.
Networking Fundamentals » Volume 5, TCP/IP NetworkingThe data from upper layers relies upon each lower layer to accomplishits task in moving the data to where it can be transmitted as bits. Asdata moves down to lower layers, you can see that each layer considersall of the information, including upper layer headers as data only (seeprevious page, Figure 6). Once all the data reaches layer 1, it is all bits.The opposite happens on the receiving side with each layer reading itsheader, taking the actions the header requests, stripping off the headerand passing to the next layer.The TCP/IP ModelThis model (see Figure 7) was formedFigure 7.many years before the OSI model. Asa result, the upper layer protocols werenot well defined and all tasks abovethe transport layer were groupedtogether in the application layer.The TCP model does not addressphysical layer tasks typical to the OSImodel, making it an entirely abstractmodel. Here we see a depiction ofthe TCP/IP model with some of themore popular protocols shown in theirproper layer. Although the OSI modelcame along after the TCP/IP model toaddress some oversights and changesin technologies, TCP model is usedthroughout networking and networktroubleshooting tools. Why? From thetransport layer down, there are actionsa network engineer can take to remedyfaulty communications. For example,if the engineer sees that here is a portconflict in TCP for a particular application, the engineer may be able toalter the transport port and restart the application on a non-conflictingport. On the other hand if the is an application level error, there is notmuch the engineer can do other than work with the application vendor.Most network engineering does not involve OSI levels 5-7.Now if we consider the case of Extreme 4-Way Stop Dilemma set toTPC/IP, everything would look identical except all interactions above thetransport layer would simply be application layer. It should be noted that theTCP/IP model can be thought of a supporting model for the more complexOSI model as the TCP/IP model only defines TCP/IP applications such asDNS and POP. These applications are normally used to support higherlevel applications such as email or HTTP browsing. Other than that, thelayers of the TCP/IP model are analogous in function to the same namelayers of the OSI model.Page 5S ecti o n 3IP AddressingI have always thought of IP addressing as the meat and potatoes of TCP/IP. Without a good understanding of IP addressing, many aspects ofTCP/IP networking will seem much more complicated than they needbe. As there is a New to Networking volume in the works dedicatedentirely to IP addressing, the explanations in the section will coverthe basics you need to know and will leave the details to that paper.IP version 4 (IPv4) AddressingIP address uniquely identifies a machine within a TCP/IP internet. (Noticethe use of the lower case i to specify a generic TCP/IP network, notnecessarily part of the network we all share worldwide, the ”uppercaseI Internet”. While it is also crucial that all machines on the Internet haveunique addresses, we will save that discussion for the Access Controland Address Translation section. IPv4 addresses are 32 bit designations,usually expressed in dotted decimal format such as 10.14.125.1. Althoughuncommon, the address can be displayed in hex or binary formats too. Theseare shown below (see Table 1) for this sample address. Later, we’ll seeexamples on why you would want to view IP address in different formats.Table 1010.0001110.01111101.00000001IPv4 addresses have two main fields, the network address field and thehost address field IP addresses. IPv4 addresses fall into one of five classes,although only three classes are available for unique public use. For themost part, these addresses are ones that can be uniquely assigned to amachine on the Internet. These three classes are shown below in Table 2.Table 2.Class Applicable NetworksNumber ofNetworksNumber of Addressper NetworkA1.0.0.0 – 126.0.0.012616,777,214 B128.0.0.0 – 191.255.0.016,384 65,534C192.0.1.0 – 223.255.255.02,097,151254Each class has a range of usable address which are unique and registeredwith a registration authority. This eliminates the possibility of duplicateaddress between interconnected private entities.The reasoning behind the segregation of addresses into classes is so thataddress blocks can be given out in a manner that is most economical forthe requirement. Say for instance that Acmeco requested a class A IPaddress block for its 3 offices and 65 staff members. Seeing that Acmecois such a small company, most all of the 16 million addresses allocatedto Acmeco would go wasted. It would make more sense to give Acmecoa single class C address space allowing them to have up to 254 devicesdirectly on the Internet. So Acmeco might receive the sole usage of theregistered network 199.12.21.0, which allows for connected devices on191.12.2.1 to 191.12.21.254. You may ask, “What about the 199.12.21 0 and199.12.21.255 addresses?” There are two rules governing the top mostand lower most addresses in any network.1. The top most address is reserved for broadcast on that network,an address machines can use to send a single message to all othermachines on that network.2. The lowest most address is reserved as the network address and usedfor routing purposes.So on the Acme network, the network address is 191.12.21.0 and thebroadcast address is 191.12.21.255.
Networking Fundamentals » Volume 5, TCP/IP NetworkingPrivate IP Address SpaceIPv4 addressing has special address ranges which are either allowedfor special circumstances, or not allowed for use by the public. Here is alisting of those rangesTable 3.Page 6 The lowest possible address for the subnet, also called the zero subnetaddress or subnet address. This address is used to identify the subnetfor routing. The highest possible address in any subnet. This is the subnet broadcastaddress.So the subnets we have created above have to following addresses (Table 4):Address SpaceReserved ForAllowedUse10.0.0.0 – 10.255.255.255Private Class A networksPrivateSubnet AddressBroadcast AddressHost Range172.13.0.0 – 172.31.255.255Private Class B networksPrivate10.10.21.010.10.21.6310.10.21.1 to 10.10.21.62192.168.0.0 – 192.168.255.255Private Class C networksPrivate10.10.21.6410.10.21.12710.10.21.65 to 10.10.21.126224.0.0.0 – .19110.10.21.129 to 92 to 10.10.21.254240.0.0.0 – 247.255.255.255ReservedThe top three are called the private IP address spaces. These addressesmay be used by anyone, as long as they are not used to connect directlyto registered address spaces.When IP addressing was first released, the 3.7 billion available uniqueaddresses were assumed to be plenty for the future on the Internet. Theproblem today is that there are over 200 million registered addressesassigned each year, bringing the total used registered addresses to almost 2.7billion. Without some help, we would run out of address space in the very nearfuture. This is where the use on the above private addresses and somethingcalled Network Address Translation (NAT) comes in. We will discuss NATin the Access Control and Address Translations section later. Anothermethod of expanding the usable IP address space was the creation of IPv6.Classless Internet Domain Routing (CIDR) Addresses and SubnettingAs just a quick note, addresses may be classful, as seen above or beclassless. Classless addresses do not strictly follow the rules of beinga class A, B or C address but rather a variable subnet mask is assignedto maximize IP address efficiency by creating subnetworks. This is donein a bit-wise fashion borrowing bits from the host address space andassigning them to the network address space. As an example let’s take alook at the network 10.10.21.0. This network could be implemented witha subnet mask of 255.255.255.0 (/24) and will allow one network with254 usable addresses. Let’s say we actually have four smaller networksthat we want to use this address space for and we need to be able to routebetween the networks. Here is what this network address looks like in bitformat with the subnet mask.Network .0255.255.255.0By borrowing the 2 most significant (left most) bits from the host portion ofthe network address, we create more networks with fewer potential hosts.New Subnet 192New Network etwork Address Translation (NAT)While subnetting is helpful in avoiding the waste of IP address space, NAToffers a unique solution to preserving IP address, the controlled reuse ofIP addresses. The idea behind NAT is this: If we allow users to reuse alimited number of controlled IP address in completely distinct IP domains,we can have routers translate those address to unique, registered addresswhen access to registered IP space in needed. Below is an example ofhow NAT can be applied. Acmeco uses a 10.10.1.0 private network address. Because this address ispart of the private IP range, it cannot be used on the public Internet. Therouter to the Internet has been set up with a pool of usable, registeredpublic address it can use to communicate over the Internet when aprivately addressed machine requests Internet services. The router sees the request to http:cisco.com from 10.10.1.55 Using an available registered address of 211.12.13.14 the router marksthat public address as being mapped to 10.10.1.5 on the inside andsends the request to cisco.com from the public address 211.12.13.14. When cisco.com responds to 211.12.13.14 the router remaps the packet tothe internal address of 10.10.1.5 and sends the packet to the workstation.In this way, if there are 500 employees at Acmeco but only 6 needs Internetaccess at a time, Acmeco only needs to use 6 registered IP addresses.This is only an example of one type of NAT and there are many. With theuse of NAT, the pressing need for more registered IP address has beensomewhat relieved. This subject will be covered in further detail in theupcoming IP Address paper.IPv6Subnet Mask11111111.1111111.11111111.00000000Table rowing the 2 host bits gave us three additional usable networks(subnets) from the single /24 original network. But how many hosts caneach subnet have? Seeing that we have 6 bits left over in each subnet, thisleaves 26 or 64 hosts in each subnet. But there are 2 reserved addressesin all subnets:Along with some features not addresses in IPv4, IPv6 conquers the issueof diminishing IP address space by expanding the usable IP address rangefrom 32 bits to 128 bits. The result is 3.4 X 1028 usable IP addresses. This isan extremely large number. Not surprising, IPv6 numbers look very large.Hex is the common format for showing numbers in the least amount ofspace, but even in hex numbers, IPv6 addresses are large. For examplehere is a hypothetical IPv6 number in hex.2001:24c8:85d3:08de:3145:c82e:0371:1237There is a feature to short hand of IPv6 numbers. Often the numberswill contain a series of zeros. These can be replaced by a double colon(::). For example2001:24c8:0000:0000:3145:c82e:0000:1237Can be expressed as2001:24c8::3145:c82e::1237As there is a complete New to Network volume dedicated to IP Addressing,we will not further examine IP addressing here.
Networking Fundamentals » Volume 5, TCP/IP NetworkingAddress Resolution Protocol (ARP) — Connecting layers 2 and 3One of the protocols that tie layers together is ARP. ARP functions totie logical IP addresses on local segments to the MAC address on thatsegment. When an IP packet reaches the final network segment and isready for delivery from the gateway router to the end system, the gatewayneeds some way to know what MAC address to deliver it to. Since IPaddresses don’t contain any MAC address information, the gatewaymust map IP addresses to MAC addresses. This is accomplished bythe gateway sending local broadcasts asking who has an IP address thegateway needs to deliver. The local system with that IP address tells thegateway it has the IP address of interest and also gives the gateway thelocal system’s MAC address. The gateway can now deliver the data tothe layer 2 MAC address. The gateway also stores this mapping of IP andMAC addresses in its ARP cache so that the gateway does not have tomake new ARP requests for every packet.Page 7S ecti o n 4IP Packets and IP RoutingAs with IP Addressing, there is a complete New to Networking volume inthe works covering IP Routing. For this reason I will keep this section to ahigh level explanation of routing. Again, the primary function of IP is theorderly routing of data to the proper network or subnetwork. Routers arerequired to move IP datagrams from one network to another if possibleor drop the packet if routing is not possible. In order to route, the routerrequires a physical connection to two or more networks and a method ofunderstanding where the datagrams needs to be sent to move towardsto its final destination. This understanding comes in the form of routingtables. Routing tables are formed according to the routing protocol enabledon the routers. There are very simple routing protocols, used primarily forsmall networks and complex routing protocols for very large networks.An IP workstation is configured with IP address and an IP gateway. Thegateway is where the workstation is instructed to send all IP communications.Once the gateway receives the IP datagram, the gateway is responsiblefor sending it on its way to the destination IP address if possible, ordropping the packet. If the packet destination address is on the samesegment as the sending device, the gateway sends the packet to theMAC address of the destination. If the IP destination is on an IP networkin the router’s routing table, the packet is sent out according to the rulesof the routing protocol. If the packet is for an unknown destination, it isdropped. Once an IP datagram reaches its destination network, the routeris responsible for delivering it to the proper local MAC address. This isaccomplished by referring to the routers Address Resolution Protocol(ARP) table which maps all local machines IP address to MAC addresses.The IP DatagramThe IP packet structure is fairly simple as it is designed for one main task— To move data from one network to another. IP is not responsible forestablishing a connection with the remote system, reliability, acknowledgingmessages, reporting dropped packets, and other connection features.This is why IP is called connectionless. Below (see Figure 8) is the IPpacket header format.Figure 8.Each field on the IP datagram has specific functions as described below: Version — 4 bits. Indicates the version of the IP header. Normally thiswill be set to v4. IHL — 4 bits. The Inter Header Length identifies where the IP headerends and the payload data begins. Type of Service — 8 bits, 6 used 2 unused. This is used to be the 3 bitToS which was expanded to allow for the 6 bit DSCP values requestingspecial or prioritized delivery. Length — 16 bits. The length of the entire IP datagram, including payload. Identification — 16 bits. Used to identify a datagram. Set by the senderto assist the receiver in reordering fragmented datagrams.
Networking Fundamentals » Volume 5, TCP/IP Networking Flags — 3 bits Used to control fragmentation. Defines 4 states Fragmentation allowed Fragmentation not allowed More fragments to come Last fragment Fragment Offset — 13 bits. Maps where this fragment fits in a seriesfragments. Time to Live — 8 bits. Sets a maximum time the datagram can exist.Each system routing an IP datagram decrements this field. Once thevalue is zero the packet is dropped. This keeps orphaned packets fromendlessly circling a network. Protocol — 8 bits. Specifies the transport layer protocol used.Page 8S ecti o n 5The Transport Layer,UDP and TCPUDP is a transport layer protocol offering connectionless and unreliabletransport of IP packets from one machine to another. Because UDP innot burdened with connection establishment or acknowledging deliveryit supplies faster communications compared to TCP. UDP operates byaccepting data from the session layer requesting UDP services, assigning aUDP port to the data, and handing it down to IP for network layer service.UDP can detect errors but does nothing to correct them. UDP is verysimilar to IP as they are both connectionless and unreliable. Header Checksum — 16 bits. Contains the results of a checksum on theheader only. If the receiving device does not obtain the same calculatedvalue, the packet is dropped.What the simple UDP header adds to IP is a UDP port number. Thisallows UDP/IP to multiplex conversations between hosts by separatingthe individual conversations with UDP ports. The 64 bit UDP header hasonly four fields as shown below (see Figure 9). Source and Destination IP Addresses —
Networking Fundamentals » Volume 5, TCP/IP Networking Page 3 SECTIoN 2 Networking Models The OSI model and the TCP/IP model are the prevalent methods to describe the interdependency of networking protocols. Both of these are conceptual models only and simply describe, not prescribe how networking
The following FCC Notice applies to the DLT TM4500 and DLT 4700 mini-libraries: FCC NOTICE : This equipment has been tested and found to comply with the limits for a Class A device, pursuant to Part 15 of the FCC Rules. These limits are designed to provide reasonable protection against harmful interference when the equipment is operated in a
Networking Fundamentals Vo l uM E 2 The Basics of Cisco IP S lA This paper examines Cisco's IP SLA technology with an emphasis on network and device impacts as well as deployment strategies. As part of the Networking Fundamental series, it is intended for the user who is new to IP SLA.
he innovation dilemma of distributed ledger technology 3 Interest in distributed ledger technology (DLT) has never been higher. Most people know it from its Blockchain cryptocurrency manifestation, but DLT is much broader—and so are the opportunities for industrial companies. In particular, DLT's transparency and security
Dell EMC Networking S4148F-ON 2.2 Dell EMC Networking S4248FB-ON The Dell EMC Networking S4248FB-ON is a 1-RU, multilayer switch with forty 10GbE ports, two 40GbE ports, and six 10/25/40/50/100GbE ports. Two S4248FB-ON switches are used as leaf switches in the examples in this guide. Dell EMC Networking S4248FB-ON 2.3 Dell EMC Networking Z9100-ON
Networking 101 . Agenda Introduction Networking Defined Purpose of Networking Types of Networking Meet & Greets Recap Disney Agenda . Did You Know? Approximately 70 percent of all jobs are found through networking Most people you meet have at least 250 contacts
Docker Networking with Linux Guillaume Urvoy-Keller Reference Scenario Basic tools: bridges, VETH Basic tools 2: Networking in namespaces Minilab : Anatomy of a docker container networking environment (45 min) Docker (host-level) Networking Docker Networking Model Docker Swarm Docker Network Overlay Sources documents Laurent Bernaille blog .
Network Skills Layer 2 & 3 Fundamentals Quality of Service Security and Segmentation Linux Networking Container Networking Cloud Networking IOT Networking Model Driven Programmability Network Function Virtualization Platform Skills Linux Administration Container Fundamentals Micro Service .
in advanced mathematics used in US universities are also popular in Australian universities for students studying engineering and some areas of applied sciences. However, the advanced mathematics .