Before talking about Services, Interfaces and Protocols concepts, due to OSI Model, we could define these words :
Service : it tells only what the layer does, and not how entities above it access it or how the layer work.
Interfaces : it tells the processes above it how to access it, and specifies parameters and results. Protocol : With this definition, the layer can characterize its own protocols, as many as he wants, and change them without affecting software in higher layers.
So we can introduce main differences between from these concepts :
| | | ||
| Service, interface and protocol | Service, interface and protocol are not clearly defined. For example, the only real services offered by the Internet layer are - Send IP Packet - Receive IP Packet |  | Protocols in the OSI model are better hidden and can be replaced relatively easily as the technology changes, which is one of the main objective of layered protocols. |
| Functionalities | Because models were invented before protocols, functionalities put in each layer are not very optimized. | | In this case, the protocols have been invented before models, so the functionalities are perfectly described. |
| Numbers of layers | Seven layers, Network (Internet), Transport and Application layers being similar to TCP/IP | | Only four layers. |
| Connectionless/ Connection-oriented communication | Both connectionless and connection-oriented communication are supported in the network layer, but only connection-oriented communication in the transport layer. | | Only one mode in the network layer (connectionless) but both modes in the transport layer are suppo |
A.b Difference
| TCP/IP |
The Department of Defense created TCP/IP to ensure and preserve date integrity. This model is a condensed version of the OSI model and only has four layers.
| |
Application Layer
Defines protocols for node-to-node application communication and also controls user interface specifications. Consists of a set of services that provide ubiquitous access to all types of networks. Applications utilise the services to communicate with other devices and remote applications
Protocols and Applications
| Port | Protocol | Discription |
| 23 | Telnet | Terminal Emulation (Telephone network) |
| 21 | FTO | ws file transfers between computers (File Transfer Protocol) |
| 69 | TFTP | Have to know what you want and where it is on the server, no directory browsing, no user authentication (Trivial File Transfer Protocol) |
| 2049 | NFS | Allows remote file systems to be mounted as local (Network File System) |
| 25 | SMTP | Used to send mail between mail servers (Simple Mail Transfer Protocol) |
| 515 | LPD | Used for print sharing of network printers with TCP/IP (Line Printer Daemon) |
| 161 | SNMP | Collect and manipulates network information (Simple Network Management Protocol) |
| 53 | DNS | Resolves FQDN to IP addresses (Domain Name Service) |
| 67 | BootP | Used by diskless workstations to receive boot file and other information via TFTP |
| DHCP | Assigns IP addresses to hosts from a pool. Can send IP address, Subnet mask, Domain Name, Default Gateway, DNS IP, WINS info. (Dynamic Host Configuration Protocol) |
Transport Layer
This layer shields the upper layers from the process of sending data. Also provides an end-to-end connection between two devices during communication by performing sequencing, acknowledgments, checksums, and flow control. Applications using services at this layer can use two different protocols: TCP and UDP.
Protocols at the Treansport Layer are:
TCP (Transmission Control Protocol)
TCP provides a connection-oriented, reliable services to the applications that use its services.
Main Functions of TCP
Segments application layer data stream--
TCP accepts data from applications and segments it into a desirable size for transmission between itself and the remote devices. The segment size is determined while TCP is negotiating the connection between the two devices. Either device can dictate the segment size.
Provides acknowledgment times--
TCP maintains timers to identify when packets have taken too long to get to their destination. When an acknowledgment is not received for a packet and the timer expires, TCP will resend the packet to the destination.
Enables sequence number checking--
TCP/IP uses sequence numbers to ensure that all packets sent by an application on one device are read in the correct order by an application on another device. The packets might not be received at the transport layer in the correct order, but TCP sequences them in their original order before passing them to the application layer.
Provides buffer management--
Any time two devices are communicating, the possibility exists that one device can send data faster than the other can accept it. If this happens, the receiving device puts the extra packets into a buffer to be read at the first chance it gets. When this data overflow persists, however, the buffer is eventually filled and packets begin to drop. TCP performs some preventive maintenance called flow control to avoid the problem.
Initiates connections with 3-way handshake--
TCP uses the concept of the three-way handshake to initiate a connection between two devices. A TCP connection begins with a device sending a request to synchronize sequence numbers (a SYN packet) and initiate a connection. The other device receives the message and responds with a SYN message and the sequence number increased by one. The first device responds by sending an acknowledgment message (an ACK) to the second device, indicating that the device received the sequence number it expected.
Performs error and duplication checking--
TCP uses a checksum to identify packets that have changed during transport. If a device receives a packet with a bad checksum, it drops the packet and does not send an acknowledgment for the packet. So the sending device will resend the packet. Any time TCP receives a duplicate packet it will drop the duplicate.
Performs acknowledgment windowing--
Any time a TCP device sends data to another device, it must wait for the acknowledgment that this data was received, To increase the bandwidth utilization, TCP can change the window size. Whatever the window size is negotiated to be, acknowledgments will only be sent after that many packets have been received at the receiving device. TCP sets the window size dynamically during a connection, allowing either device involved in the communication to slow down the sending data rate based on the other devices capacity. This process is known as sliding window because of TCP's ability to change the window size dynamically.
TCP Overview
Before data is sent, the transmitting host contacts the receiving host to set up a connection known as a virtual circuit. This makes TCP connection-oriented. During the handshake the two hosts agree upon the amount of information to be sent before an acknowledgment is needed (Windowing). TCP takes the large blocks of data from the upper layers and breaks them up into segments that it numbers and sequences. TCP will the pass the segments to the network layer, which will route them through the Internetwork. The receiving TCP can put the segments back into order. After packets are sent, TCP waits for an acknowledgment from the receiving end of the virtual circuit. If no acknowledgment is received then the sending host will retransmit the segment.
| TCP Header Information | |||||
| |||||
| Sequence Number 32 bits (Number to ensure proper sequence of data.) | |||||
| Acknowledgment Number 32-bits (Identifies next segment expected) | |||||
| Header Length 4 bits (Number of 32 bit words in header) | Reserved 6 bits (Always 0) | Code bits 6 bits (Identifies type of segment, setup/termination of session) | Window size 16 bits (Number of octets the device is willing to accept) | ||
| |||||
| Options 0 or 32 bits (Identifies maximum segment size) | |||||
| Data | |||||
UDP (User Datagram Protocol)
UDP transports information that doesn't require reliable delivery; therefore it can have less overhead than TCP as no sequencing or acknowledgments are used. NFS and SNMP use UDP for their sessions, the applications have their own methods to ensure reliability. UDP receives blocks of information from the upper layers, which it breaks into segments. It gives each segment a number, sends it, and then forgets about it. No acknowledgments, no virtual circuits, connectionless protocol.
| UDP Header Format | |
|
Source Port Number |
Destination Port Number |
|
UDP Length |
UDP Checksum |
|
Data | |
Differences between TCP and UDP
| TCP | UDP |
| Sequenced | Unsequenced |
| Reliable -sequence numbers, acknowledgments, and 3-way handshake | Unreliable -best effort only |
| Connection Oriented | Connectionless |
| Virtual Circuits | Low Overhead |
| Checksum for Error Checking | Checksum for Error Checking |
| Uses buffer management to avoid overflow, uses sliding window to maximize bandwidth efficiency | No flow control |
| Assigns datagram size dynamically for efficiency | Every datagram segment is the same size |
TCP and UDP Port Numbers
TCP and UDP use port numbers to communicate with the upper layers. Port numbers keep track of different sessions across the network. The source port will be above 1024 (unprivileged). 1023 and below (privileged) are known as well known ports and are assigned to common protocols. TCP and upper layer don't use hardware (MAC) and logical (IP) addresses to see the host's address; instead they use port numbers.
Internet Layer
The Internet Layer exists for routing and providing a single network interface to the upper layers. IP provides the single network interface for the upper layers.
Protocols at the Internet Layer are:
IP (Internet Protocol)
The Internet Protocol (IP) is a network-layer (Layer 3) protocol that contains addressing information and some control information that enables packets to be routed. IP has two primary responsibilities: providing connectionless, best-effort delivery of datagrams through an internetwork; and providing fragmentation and reassembly of datagrams to support data links with different maximum-transmission unit (MTU) sizes.
All machines on a TCP/IP network have a unique logical address, an IP address. The Internet Layer (IP) has a complete picture of the entire network and is responsible for path determination and packet switching. IP is the transport for TCP, UDP, and ICMP and provides an unreliable service. It lets the upper layer protocols that use it worry about reliability. IP will perform as a connectionless service because it handles each datagram as an independent entity. IP performs packet switching and path determination by maintaining tables that indicate where to send a packet based on its IP address. IP gets the destination address from the packet. IP receives segments from the Host-to-Host layer and fragments them into packets. IP will then reassemble the packets into segments on the receiving end to send to the Host-to-Host layer. Each packet has the source and destination IP address. Each router will make path determinations based on the destination IP address.
ICMP (Internet Control Message Protocol)
The Internet Control Message Protocol (ICMP) is a network-layer Internet protocol that provides message packets to report errors and other information regarding IP packet processing back to the source. ICMP utilizes IP to carry the ICMP data within it through a network.
ICMP Messages
ICMPs generate several kinds of useful messages, including Destination Unreachable, Echo Request and Reply, Redirect, Time Exceeded, and Router Advertisement and Router Solicitation. If an ICMP message cannot be delivered, no second one is generated. This is to avoid an endless flood of ICMP messages.
When an ICMP destination-unreachable message is sent by a router, it means that the router is unable to send the package to its final destination. The router then discards the original packet. Destination-unreachable messages include four basic types: network unreachable, host unreachable, protocol unreachable, and port unreachable.
Network-unreachable messages usually mean that a failure has occurred in the routing or addressing of a packet.
Host-unreachable messages usually indicates delivery failure, such as a wrong subnet mask.
Protocol-unreachable messages generally mean that the destination does not support the upper-layer protocol specified in the packet.
Port-unreachable messages imply that the TCP socket or port is not available.
An ICMP echo-request message, which is generated by the ping command, is sent by any host to test node reachability across an internetwork. The ICMP echo-reply message indicates that the node can be successfully reached. PING - Packet Internet Gropher, uses echo message to test physical connectivity.
An ICMP Redirect message is sent by the router to the source host to stimulate more efficient routing. The router still forwards the original packet to the destination. ICMP redirects allow host routing tables to remain small because it is necessary to know the address of only one router, even if that router does not provide the best path. Even after receiving an ICMP Redirect message, some devices might continue using the less-efficient route.
An ICMP Time-exceeded message is sent by the router if an IP packet's Time-to-Live field (expressed in hops or seconds) reaches zero. The Time-to-Live field prevents packets from continuously circulating the internetwork if the internetwork contains a routing loop. Routers discard packets that have reached their maximum hop count and tell the source machine that the packet is expired. Traceroute - uses ICMP timeouts to find the path a packet takes through the internetwork.
ARP (Address Resolution Protocol)
Used to find the MAC address from the known IP address. ARP sends a broadcast asking for the machine with the specified IP address to respond with its MAC address. If two devices want to communicate, the first device can send a broadcast ARP message requesting the physical address for a specified IP address. The receiving device responds with its IP address and the first device maintains the entry in its ARP cache. If a device doesn't exist on the same subnet, the sending device addresses the the default gateway's physical address and sends the packet to the default gateway.
RARP Reverse Address Resolution Protocol)
This protocol is used to find an IP address when the MAC address is known. A machine sends a broadcast with its MAC address and requests its IP address. An example of a device that uses RARP is a diskless workstation. Since it can't store its logical network address, it sends its MAC address to a RARP server to requests its IP address. A RARP server responds to the RARP request with the device's IP address.
Network Access Layer
The Network Access Layer monitors the data exchange between the host and the network. Oversees MAC addressing and defines protocols for the physical transmission of data.
A.C Diffirencen its infancy, a war to define the protocols of the Internet was fought between geeks and governments. Two standards, TCP/IP and OSI sought to provide a usable, structured design for the computer networking efforts which would eventually make our global Internet possible.
Despite their correspondences, significant technical differences exist between TCP/IP and OSI which stem from their design philosophies and which are far from simplistic. I admit at the outset that I've never worked with OSI, so the technical information presented is drawn mostly from the sources cited and my TCP/IP knowledge.
A technical audience should note that this discussion is based upon the reference models, rather than the protocol stacks as such. However, the issue of "TCP/IP versus OSI" is far more than just a technical discussion. It is a historical demonstration of what occurs when two opposite design philosophies are used to develop a protocol, and is far more a political and philosophical than a technical issue, and perhaps serves as a lesson in why politics makes for bad technical standards.
- "So the Internet perspective is implement it, make it work well, then write it down,
whereas the OSI perspective is to agree on it, write it down, circulate it a lot and now we'll
see if anyone can implement it after it's an international standard and every vendor in the
world is committed to it. One of those processes is backwards, and I don't think it takes
a Lucasian professor of physics at Oxford to figure out which."
-- Marshall Rose, "The Pied Piper of OSI"
On OSI:
OSI's major contribution to networking theory is in its distinct separation between three fundamental concepts:
- 1. Services: A service defines what a layer does, but abstracts the details of implementation from higher levels in the protocol stack.
- 2. Interfaces: The interface makes the layer available to higher layers. It defines the conventions of communication - what to send and what to expect, but also does not deal with implementation details.
- 3. Protocols: These are private methods of implementation which the higher layers have no access to or knowledge of. Thus, they can be changed (i.e. to allow adding support for a new hardware technology) without altering the basic functioning of higher layers.
It should be rather obvious to someone with a computer science background that these concepts are extremely similar to the design philosophy of object-oriented programming. The layer possesses a set of service methods which can be invoked through the layer interface to initiate internal protocols.
This similarity also means that OSI would gain some of the main advantages of object-based design: data encapsulation, modularity and reusability, and access protection. Because the protocols themselves are hidden, they can be changed as new and improved technologies become available, without compromising the integrity of the system, in a way transparent to higher layers (to say nothing of the end-users!).
We might in hindsight consider that this was a good idea. Unfortunately, the designers of the OSI model built the reference model before the protocols existed and did not understand from an engineering perspective where various pieces would optimally fit. This is because the majority of the protocols they were designing this for hadn't even been invented yet!
On TCP/IP:
TCP/IP's differences from the OSI model stem from its design requirements:
- A common application set
- Implicit support for dynamic routing
- Connectionless networking level
- Implicit support for packet-switching
- Universal connectivity
It should be clearly noted that TCP/IP has no provisions for the object-like separation between service, interface, and protocol of the OSI model. Whereas OSI was an abstraction created before technologies existed, TCP/IP simply described the existing hardware from an engineer's perspective and gave little thought to ensuring the model made sense at a higher level. The layers fit the technology perfectly, of course, but dialogue was impossible with other non-TCP/IP-based networks!
The main differences between OSI and TCP/IP exist in the higher layers (5-7) and in the Network layer (3). OSI supports connectionless as well as connection-oriented protocols at the network level (again, due to its inherently abstract design), but only connection-oriented communication at the transport layer. TCP/IP is connectionless at the network level and supports both modes at the transport level, where the user, who often has little knowledge of low-level networking hardware and protocols, has to select how his connection functions.
Problems with OSI
OSI was a poor performer in implementation, and there are definite flaws in the protocols. Flow control is a problem at every layer and error control must be implemented all layers as well. Network management is problematic and was actually omitted from the original OSI model. Semantic confusion about the Presentation and Application layers created so many major headaches that data security and encryption were eventually taken out altogether!
OSI was killed off because:
- Early slow and bug-filled, unusable implementations ruined its public image.
- OSI was thought to originate with the European Community and the U.S. federal government.
- Its probable market for use was proprietary. TCP/IP was bundled as part of Berkeley UNIX and was free.
- OSI is labyrinthine and full of almost bureaucratic levels of unnecessary complexity.
- The seven-layer model was somewhat arbitrary, and was basically done in an attempt to wrest control away from IBM's 7-layer SNATM protocol to a world standard controlled by a neutral organization (the ISO) rather than by a single corporation -- not to simplify actually using the model!
Problems with TCP/IP:
- It can't be used to intelligently describe another type of protocol stack (like SNA).
- Its network layer is more of an interface than a true layer of its own.
- There is no distinction between the Physical and Data Link layers. This is a poor choice from an engineering standpoint.
- Many of the original protocol implementations were hacks (in the "oldskool" sense, of course) with very limited usefulness and arbitrary constraints based on hardware limitations or on simplifying the coding task.
A Point of Networking Zen:
- The terminology of the OSI reference model is widely-used to describe behavior and design of networks.
- The implementations of that model are basically crap.
- The protocols of TCP/IP are generally well-thought-out and widely-used.
- Those protocols are usually best described by the design terminology of the OSI reference model.
- Tannenbaum, Andrew Computer Networks: Third Edition Prentice-Hall 1996.
- http://www.cellsoft.de/telecom/tcpiposi.htm
- "Forcing the TCP/IP suite into a non-standard 5-layer Model"
(http://mike.passwall.com/networking/netmodels/tcpip5layermodel.html)
No comments:
Post a Comment