Teknologi Pendukung Next Generation Network (NGN)
1
Tujuan • Memahami cara kerja jaringan TCP/IP dengan cara memahami fungsi dari lapisan pembentuk jaringan IP • Memahami evolusi jaringan data publik • Memahami proses kompresi suara (Codec)
2
Agenda
• Konsep Protokol dan Jaringan Komunikasi Data – – – –
Konsep Protokol Physical Layer Data Link Layer Network Layer
• Teknologi Jaringan Data • Jaringan TCP/IP • Voice Primer (Codec) 3
Model Referensi OSI dari ISO (1/3) • International Standard Organization (ISO) telah mendefinisikan Model Referensi OSI yang terdiri atas 7 layers (lapisan) sebagai acuan untuk mendesain protokol jaringan • OSI = Open System Interconnection Application
Layer 7
Presentation
Layer 6
Session
Layer 5
Transport
Layer 4
Network
Layer 3
Data Link
Layer 2
Physical
Layer 1 4
Model Referensi OSI dari ISO (2/3) EXAMPLES Application
Presentation
Session
User Interface • Bagaimana data ditampilkan • Pengolahan khusus seperti encryption
Membangun, mengatur, dan mengakhiri suatu sesi antar aplikasi
Telnet HTTP ASCII EBCDIC JPEG Operating System/ Application Access Scheduling
Transport Layer Network Layer Data Link Physical
5
Model Referensi OSI dari ISO (3/3) Application Presentation
Contoh
Session
Transport
Network
Data Link
Physical
• Pengiriman Reliable atau unreliable • Error correction sebelum pengiriman ulang •Penyediaan alamat logik yang digunakan router untuk pemilihan path • Menggabungkan bit menjadi byte dan byte menjadi frame • Access to media using MAC address • Error detection not correction • Pengiriman bit melalui suatu kanal komunikasi
TCP UDP SPX IP IPX 802.3 / 802.2 HDLC EIA/TIA-232 V.35 6
Header Protokol • Setiap lapisan pada protokol berkomunikasi dengan lapisan yang setara (pada sisi remote) melalui informasi yang disimpan dalam header • Setiap lapisan akan menambahkan header-nya di depan pesan yang diterima dari lapisan di atasnya. Original User Data Layer 7 header Layer 6 header Layer 5 header Layer 4 header Layer 3 header Layer 2 header
7
Enkapsulasi Data Application Presentation Session
Upper Layer Data TCP Header
Transport
Upper Layer Data
IP Header
Data
LLC Header
Data
FCS
MAC Header
Data
FCS
0101110101001000010
PDU Segment
Network
Packet
Data Link
Frame
Physical
Bits PDU = Protocol Data Unit 8
Dekapsulasi Data Application Presentation Session
Upper Layer Data
Transport
Network
TCP
der a e H
Upper Layer Data
TCP+ Upper Layer Data
der a e IP H
Data Link LLC
Physical
IP + TCP + Upper Layer Data
der a e H
der a e H MAC
LLC Hdr + IP + TCP + Upper Layer Data
0101110101001000010 9
Agenda • Konsep Protokol dan Jaringan Komunikasi Data – – – –
Konsep Protokol Physical Layer Data Link Layer Network Layer
• Teknologi Jaringan Data • Jaringan TCP/IP • Voice Primer (Codec)
10
Fungsi Physical Layer
• Bentuk Sinyal
V.35
EIA/TIA-232
802.3
• Jenis Konektor
Physical
• Jenis Media
Ethernet
Mendefinisikan
11
Physical Layer: Ethernet/802.3
10Base2—Thick Ethernet 10Base5—Thick Ethernet
Host Hub 10BaseT—Twisted Pair
Hosts 12
Hub beroperasi di Physical Layer
Physical
A
B
C
D
• Semua stasion berada di domain kolisi yang sama • Semua stasion berada di domain broadcast yang sama • Bandwidth digunakan bersama 13
Hub: Satu Domain Kolisi • Lebih banyak stasion berarti potensi kolisi makin tinggi • CSMA/CD digunakan
14
Agenda • Konsep Protokol dan Jaringan Komunikasi Data – – – –
Konsep Protokol Physical Layer Data Link Layer Network Layer
• Teknologi Jaringan Data • Jaringan TCP/IP • Voice Primer (Codec)
15
Fungsi Data Link layer (1/3)
802.2 Ethernet
Physical
• Alamat fisik pengirim dan penerima • Menyediakan akses ke Higher layer protocol (Service Access Point) berkaitan dengan frame • Topologi Jaringan • Frame sequencing • Connectionless oriented
Data Link
Mendefinisikan
802.3
16
Fungsi Data Link layer (2/3) MAC Layer - 802.3 # Bytes
8
6
6
2
Preamble Dest add Source add Length
0000.0C IEEE assigned
xx.xxxx Vendor assigned
Variable Data
4 FCS
Ethernet II uses “Type” here and does not use 802.2.
MAC Address
17
Fungsi Data Link layer (3/3) 802.2 (SNAP) 1
1
1 or 2
3
2
Dest SAP Source SAP Ctrl OUI Type ID AA AA 03
OR
Variable
Data
802.2 (SAP) 1
1
1 or 2
Dest SAP
Source SAP
Ctrl
Preamble Dest add Source add Length
Variable
Data
Data
FCS
MAC Layer - 802.3 18
Switch dan Bridge Beroperasi di Data Link Layer Data Link
1
2
3
4
OR
1
2
• Masing-masing segmen memiliki kolisi domain sendiri-sendiri. • Semua segmen berada di domain broadcast yang sama. 19
Agenda • Konsep Protokol dan Jaringan Komunikasi Data – – – –
Konsep Protokol Physical Layer Data Link Layer Network Layer
• Teknologi Jaringan Data • Jaringan TCP/IP • Voice Primer (Codec)
20
802.3
Frame Relay
Data Link
• Interkoneksi berbagai data link
802.2
HDLC
Network
• Menentukan jalur yang akan dipergunakan untuk pengiriman data.
IP, IPX
Ethernet
• Mendefinisikan alamat lojik pengirim dan penerima.
Physical
Fungsi Network Layer (1/4)
EIA/TIA-232 v.35
21
Fungsi Network Layer (2/4) Network Layer End Station Packet IP Header
• Logical Address
Source address
Destination address
Data
172.15.1.1 Network
Node
22
Fungsi Network Layer (3/4) Address
Mask
172.16.122.204 255.255.0.0 172
16
122
204
Binary Address 10101100 00010000 01111010 11001100 255
Binary Mask
255
0
0
11111111 11111111 00000000 00000000 Network
Host 23
Fungsi Network Layer (4/4) 1.1
1.2
1.0
4.0
1.3 E0
2.1 S0
Routing Table NET INT Metric 1 E0 0 2 S0 0 4 S0 1
2.2 S0
4.3
4.1
4.2
E0
Routing Table NET INT Metric 1 S0 1 2 S0 0 4 E0 0
• Logical addressing allows for hierarchical network • Configuration required • Uses configured information to identify paths to networks 24
Router : Beroperasi di Network Layer • Broadcast control • Multicast control • Optimal path determination • Traffic management • Logical addressing • Connects to WAN services
25
Agenda • Konsep Protokol dan Jaringan Komunikasi Data • Teknologi Jaringan Data – Local Area Network (LAN) – Wide Area Network (WAN) • X.25 • FR • ATM
• Jaringan TCP/IP • Codec 26
Standard LAN IEEE 802.2. IEEE 802.3. IEEE 802.3u. IEEE 802.3z. IEEE 802.4. IEEE 802.5. IEEE 802.6. IEEE 802.11. IEEE 802.12.
Logical Link Control CSMA/CD (Ethernet) Fast Ethernet Giga Ethernet Token Bus Token Ring DQDB WireLess LAN 100 VG Anylan
27
Teknologi Ethernet F Ethernet (IEEE 802.3)
n Cara akses ke media ≈ fisik acak n Topologi bus l 10Base5 (thick coaxial, 10 Mbps, segment 500 m) l 10Base2 (thin coaxial, 10 Mbps, segment 200 m) API, e.g. NDIS, ODI LLC MAC Physical
FIle Server
Laser printer
}
Data link
PC
Ethernet Coaxial Cable
PC
PC
28
Diagram Algoritma CSMA/CD
Packet?
Sense Carrier
No
Send
Detect Collision Yes
Discard Packet attempts < 16
Jam channel b=CalcBackoff(); wait(b); attempts++;
attempts == 16
29
Teknologi Ethernet …
n Topologi star/hub l 10BaseT (shielded/unshielded twisted pair, 10 Mbps, segment 100 m) l 10BaseF (fiber optic, 10 Mbps, segment 2000 m) Hub
Twisted pair
FIle Server
PC
F Fast Ethernet
Laser printer
PC
PC
n Topologi star/hub l 100BaseF (fiber optic, 100 Mbps, segment 2000 m) l 100BaseT (shielded/unshielded twisted pair, 100 Mbps, segment 100 m)
30
Jenis-jenis Ethernet Name
Maximum Data Rate
Cables
10Base5
10 Mbps
Coaxial
10Base2
10 Mbps
Coaxial
10BaseT
10 Mbps
UTP cat 3, UTP cat 5
100BaseT
100 Mbps
UTP cat 5, fiber
1000BaseX
1 Gbps
UTP cat 5, UTP cat 5e, UTP cat 6, fiber
10 GbE
10 Gbps
UTP cat 5e, UTP cat 6, UTP cat 7, fiber
40 GbE
40 Gbps
fiber
31
Fast Ethernet IEEE Ethernet Standards/Working Groups 802.3 CSMA/CD
802.3 Ethernet
802.3u Fast Ethernet
802.3x Full Duplex Ops
802.3z Gigabit Ethernet
10base-T 10base-5 10base-2 10broad-36
100base-TX 100base-FX 100base-T4 Autonegotiaton
Flow Control Control Frames Pause Frames Type Field
1000base-SX 1000base-LX 1000base-CX
Fast ethernet adalah suatu istilah generik untuk keluarga jenisjenis High speed Local Area Network yang mempunyai kecepatan 100Mbps melalui kabel UTP atau Fiber optic
32
Token Ring Token Passing
33
IBM Token Ring Network Vs. IEEE 802.5
34
Gigabit Ethernet (1000 BASE X) • Memberikan kecepatan sebesar 1000Mbps(kapasitas one billion bits per second) untuk operasi half-duplex dan fullduplex. • Menggunakan format frame Ethernet frame format dan teknologi MAC ethernet – metoda akses CSMA/CD dengan dukungan untuk satu repeater per collision domain – Backward compatible dengan 10 BASE-T dan 100 BASE-T
• Menggunakan 802.3 full-duplex Ethernet technology • Menggunakan 802.3x flow control.
35
Gigabit Ethernet Architecture Standard
36
Gigabit Ethernet Technology Fiber: 1000 BASE SX 1000 BASE LX
short wavelength long wavelength
Copper: 1000 BASE CX 1000 BASE T
shielded twisted pair unshielded twisted pair
* Based on Fiber Channel physical signaling technology.
37
IEEE 802.3Z Gigabit Ethernet Medium Alternatives
38
Agenda • Konsep Protokol dan Jaringan Komunikasi Data • Teknologi Jaringan Data – Local Area Network (LAN) – Wide Area Network (WAN) • X.25 • FR • ATM
• Jaringan TCP/IP • Codec 39
Overview WAN
Service Provider
• WAN menghubungkan LAN dan untuk akses remote. • Koneksi dapat dilakukan sesuai kebutuhan dan biaya.
40
Koneksi WAN : Layer 1 Synchronous serial
Leased Line Asynchronous serial, ISDN Layer 1
Circuit-switched
Telephone Company
Synchronous serial
Packet-switched
Service Provider
41
Interfacing WAN Service Provider WAN service provider toll network CO Switch
S
S
S
S Local Loop
S
S
S
Trunks and switches
Demarcation Customer Premises Equipment
Point-to-point or circuit-switched connection
42
Koneksi Serial Point-to-Point Router connections End user device
DTE
CSU/ DSU
DCE Service Provider EIA/TIA-232
EIA/TIA-449
V.35
X.21
EIA-530
Network connections at the CSU/DSU 43
Protokol Enkapsulasi WAN : Layer 2 Leased Line
HDLC, PPP, SLIP
X.25, Frame Relay, ATM Packet-switched
Service Provider
PPP, SLIP, HDLC Circuit-switched
Telephone Company
44
Format Frame HDLC HDLC Flag
Address Control
Data
FCS
Flag
• Hanya dapat membawa satu jenis protokol higher layer
HDLC = High Level Data Link Control
45
Overview PPP Multiple protocol encapsulations using NCPs in PPP
TCP/IP Novell IPX AppleTalk
PPP Encapsulation
Link setup and control using LCP in PPP
•PPP dapat membawa beberapa suit protokol menggunakan Network Control Program (NCP) •PPP mengontrol setup beberapa opsi link menggunakan LCP 46
Elemen pada Lapisan PPP IP
IPX
IPCP
PPP
IPXCP
Layer 3 Protocols Many Others
Network Layer
Network Control Protocol Authentication, other options
Data Link Layer
Link Control Protocol Synchronous or Asynchronous Physical Media
Physical Layer
•PPP—A data link with network-layer services 47
Opsi Konfigurasi LCP PPP Fitur
Operasi
Protokol
Autentikasi
Require a password PAP Perform Challenge Handshake CHAP
Kompresi
Compress data at source; reproduce data at destination
Stacker or Predictor
Deteksi Error
Monitor data dropped on link Avoid frame looping
Magic Number
Multilink
Load balancing across multiple links
Multilink Protocol (MP) 48
Agenda • Konsep Protokol dan Jaringan Komunikasi Data • Teknologi Jaringan Data – Local Area Network (LAN) – Wide Area Network (WAN) • X.25 • FR • ATM
• Jaringan TCP/IP • Voice Primer (Codec) 49
An Introduction to X.25 X.25 cloud LAN protocol
LAN protocol X.25
X.25
Virtual circuit
• • • • •
IP AppleTalk Novell IPX Banyan VINES XNS
• • • •
DECnet ISO-CLNS Compressed TCP Bridging
X.25 Protocol Stack OSI Reference Model
X.25 Protocol
7
Application
•
6
Presentation
•
5
Session
•
4
Transport
•
3
Network
X.25
3
2
Data Link
LAPB
2
1
Physical
Physical
1
X.25 User level Packet level Link level Physical level
packet packet packet
packet packet packet
packet
packet
01001110001
52
X.25 DTE and DCE Public data network (PDN) X.25 DTE
X.25 DCE
X.25 DCE
X.25 DTE
– X.25 DTE—Usually a subscriber's router or PAD – X.25 DCE—Usually a PDN's switch or concentrator
Identifying the PAD Public data network (PDN) PAD DTE
DCE
X.25
DCE
X.25 DTE host
Asynchronous terminals
– PAD collects data and outputs it into X.25 packets
X.25 (X.121) Addressing Format 4 decimal digits
Up to 10 or 11 decimal digits
Data network ID code
Network terminal number
– Addressing set by service provider
X.25 Address Resolution Token Ring
X.25
ARP ETH Destination Source MAC MAC
X.25 map
IP
X.25
Source X.121
Destination X.121
IP
X.25 Encapsulation IP network
IP network
X.25
Data-link frame (LAPB) (L2)
X.25 header (L3)
IP datagram (L3)
– Protocol datagrams are reliably carried inside LAPB frames and X.25 packets
X.25 Virtual Circuits
Switched virtual circuits (SVCs)
Permanent virtual circuits (PVCs)
– Numbering for up to 4095 VCs per X.25 interface
SVC Usage IPX
Host
– SVCs may be combined to improve throughput for a particular protocol
Single-Protocol Virtual Circuits
IPX TCP/IP Host
AppleTalk
– Each network-layer protocol is associated with its own virtual circuit
Multiprotocol Virtual Circuits
IPX TCP/IP AppleTalk Host
– Multiple protocols are carried within a virtual circuit to a single destination – A maximum of nine protocols may be mapped to a host
Agenda • Konsep Protokol dan Jaringan Komunikasi Data • Teknologi Jaringan Data – Local Area Network (LAN) – Wide Area Network (WAN) • X.25 • FR • ATM
• Jaringan TCP/IP • Voice Primer (Codec) 62
Frame Relay Overview (1/3) y Frame Relay adalah suatu cara untuk mengirimkan informasi melalui suatu WAN dengan membagi-bagi data dalam suatu paketpaket y Frame Relay beroperasi pada lapisan fisik dan data link dari protokol model referensi OSI y Frame Relay adalah protokol data link yang menangani multiple virtual circuits (VCs) y Sangat tergantung pada protokol lapisan atasa seperti TCP untuk proses error correction y Menggunakan Frame yang disederhanakan tanpa mekanisme error correction melalui suatu jaringan digital kualitas yang tinggi, Frame Relay mampu mengirimkan data pada kecepatan tinggi.
63
Frame Relay Overview (2/3) þ Frame relay lebih effisien dari pada protokol X.25 þ Frame relay interface dapat digunakan dalam suatu jaringnan publik atau pada jaringan private pada satu perusahaan þ Biaya internetworking dapat dikurangi § Dapat membagi suatu jaringan fisik tunggal kedalam beberapa subinterface § Biaya perangkat yang kecil § Biaya yang kecil dibanding dengan leased lines
þ Meningkatkan performansi jaringan dan mengurangi kompleksitas dari jaringan § Jumlah proses dapat dikurangi dibanding dengan X.25 § Penggunaan jaringan transmisi kecepatan tinggi secara efisien § Frame Relay dapat meningkatkan peformansi dan waktu respon dari aplikasi.
64
Frame Relay Overview (3/3) DCE or Frame Relay Switch CSU/DSU
Frame Relay.
– Koneksi menggunakan Virtual circuits (VCs) – Connection-oriented service
Frame Relay Stack OSI Reference Model
Frame Relay
Application Presentation Session Transport Network
IP/IPX/AppleTalk, etc.
Data Link
Frame Relay
Physical
EIA/TIA-232, EIA/TIA-449, V.35, X.21, EIA/TIA-530
66
Frame Relay Terminology PVC DLCI: 100 DLCI: 200
LMI LMI 100=Active 100=Active 400=Active 400=Active
DLCI: 400 Local Access Loop=64 kbps
Local Access Loop=T1 PVC
DLCI: 500
Local Access Loop=64 kbps
Konsep Jaringan Frame Relay - virtual connection - virtual path
Ethernet
DLCI = 575
DLCI = 441
FR Switch
DLCI = 222
DLCI = 144 Token-ring RouterFR UNI
FR Switch
FR UNI Router FR NETWORK FR Switch
Ethernet
FR UNI = Frame Relay User Network Interface DLCI = Data Link Connection Identifier FR Network IP
IP
{
Q.922 Core
DLCI = 144
FR Service
Physical Layer Router
Q.922 Core
Q.922 Core
Physical Layer
Physical Layer
FR Switch
FR Switch
FR UNI
DLCI = 575 DLCI = 441
FR UNI
Q.922 Core Physical Layer Router
68
Frame Relay Service n FR UNI
l Data link : FR frame l Physical : Interface V.35, X.21, atau G.703, Bitrate port akses : n x 64 kbps s/d 2 Mbps n Connection Oriented Mode (pada data link layer) l Permanent Virtual Connection PVC (saat pendaftaran) l Switched Virtual Connection SVC (via prosedur signalling) n Connection Admission Control l Deklarasi & negosiasi parameter trafik per PVC/SVC : CIR, Bc, Be, T n Aplikasi l Fast Internet Access l Interkoneksi LAN (file transfer antar LAN) l Image Transfer, CAD/CAM, dll.
69
FR Frame n n
Variable length frame (n > 1600 octets) DLCI digunakan untuk mengidentifikasi suatu FR connection Flag (01111110)
Address Field
User Information
FCS
Flag (01111110)
1
2
n
2
1
8
7
6
5
4
DLCI (High Order)
3
2
1
C/R 0/1
EA 0
8
7
6
5
DLCI (Low Order)
4 FECN
3
2
1
BECN
DE
EA 1
Octets
Bits
F Network layer protocol yang didukung n n n
IP (Internet Protocol)
n
Apple Talk, dll.
XNS (Xerox Network System) IPX (Internet Packet eXchange/Novell Netware)
70
Kontrak Trafik
Trafik per Virtual Connection n n n n
CIR = Committed Information Rate Bc = Committed Burst Size T = Measurement Interval Be = Excessive Burst Size
71
Manajemen Kongesti Skenario n Data flow dari A ke B n FR switch 1≈ kongesti n Kontrol kongesti l l
FECN = Forward Explicit Congestion Notification BECN = Backward Explicit Congestion Notification
Data Flow A to B
Router B
FR Switch 4
FECN FR Switch 1
FR Switch 2 BECN FR Switch 3
Router A 72
Frame Relay Traffic Shaping (1/2)
Port speed Rate
CIR
=Bc >Bc
Time
73
Frame Relay Traffic Shaping (2/2)
Slope of this line is the access rate Frames sent in excess of Be will be discarded by the network
Bc+Be
Frames sent in excess of Bc but below Be will have their DE bit set by the network
Bc
Frames sent below Bc will be processed correctly by the network
Bits sent To
Tc
Slope of this line is the CIR
Time 74
Agenda • Konsep Protokol dan Jaringan Komunikasi Data • Teknologi Jaringan Data – Local Area Network (LAN) – Wide Area Network (WAN) • X.25 • FR • ATM
• Jaringan TCP/IP • Voice Primer (Codec) 75
Asynchronous Transfer Mode (ATM) • Originally developed in ITU (CCITT) as part of B-ISDN • Initially designed for high-speed fiber optic lines • Worldwide support from: • computer vendors • data networking vendors • telecommunications vendors • network planners, designers, and administrators
76
What is ATM? ATM is packet switching! • Switched or permanent connections • Traffic type independent (voice, data, interactive video) • Fixed length packet - 53 bytes (cell)
header
payload
Fixed length packet = cell
77
What is ATM? ATM
Conventional Telecom
Conventional LAN
Traffic Type
Data, Voice, Video
Voice
Data
Transmission Unit
Fixed Cell
Fixed Frame
Variable Packet
Cell
Circuit
Packet
Connection-oriented
Connection-oriented
Connectionless
Delivery
Defined Classes
Guaranteed
Best Effort
Access
Dedicated
Dedicated
Shared
Application Dependent
Channel Dependent
Protocol Dependent
Switching Connection Type
Rate & Media
78
ATM Cell Relay: The Underlying Technology Cell Features
Benefit
Small
Low latency to support real-time services like audio and video
Fixed Length
Fast hardware switching and scalability
Standardized
Usable in all networks (LAN and WAN) Transmission Cable
48 Bytes
5 Bytes
Header Data Payload
79
ATM Switching Connections (routes) set up by software • Routing (path through multiple-switch network) and resource allocation is performed once per connection by switch control CPU
Cells are switched by hardware • Hardware (table lookup + switching fabric) switches each incoming cell to appropriate output port • Once a connection is established, cells are not touched by software
ATM LANs grow by adding more switches • More aggregate bandwidth • Negligible additional latency (10-50 microseconds per switch hop vs. 10000 microseconds per router hop)
80
ATM Switching: Contentionless Time Division Ports 1-4 622 Mbps (OC-12)
2.5 Gbps Switch Fabric
Ports 13-16
Ports 5-8
Benefits: • • • • •
4 x 155 Mbps OC-3c or 6 x 100 Mbps TAXI with Multiple Priority Output Buffers
Non-blocking Deterministic performance Flexible port speeds Congestionless hardware multicast Lowest transit delays
Ports 9-12 81
ATM Switching: Contentionless Time Division Advantages • Non-blocking • Deterministic performance—probability of cell loss = 0 • Flexible port speeds (DS-1, E-1, DS-3, E-3, 100M, 155M, 622M)
• Hardware multicast without increasing fabric cell traffic • Low transit delays Disadvantages • Limited scalability on single TDM fabric Use time-space-time to expand 82
Protocol Reference
OSI ATM Adaptation ATM Layer
NetworkLayer Data Link Layer Physical Layer
Physical Layer
Convergence Segmentation and reassembly Generic flow control Cell header generation/extraction VPI/VCI translation Cell mux and demux Cell route decoupling HEC header sequence generation/verification Cell deleniation Transmission frame adaptation Transmission frame generation/recovery Bit timing Physical medium
83
ATM Physical Layer CCITT SDH • STM-1 (155 Mbps) • STM-4 (622 Mbps) • STM-16 (2.5 Gbps) ANSI/Bellcore SONET • OC-3 (155 Mbps) • OC-12 (622 Mbps) • OC-48 (2.5 Gbps)
ATM Forum • SONET OC-3c (155 Mbps) • UTP-5 and STP • Multimode Fiber (155 Mbps) • Multimode Fiber (100 Mbps) • UTP-3 (51 Mbps) • DS-3 (45 Mbps) • E-3 (34 Mbps) • E-1 (2 Mbps) • DS-1 (1.544 Mbps) ATM 25 Consortium (led by IBM) • UTP-3 (25 Mbps)
84
SONET/SDH Hierarchy • SONET = Synchronous Optical Network (US) • SDH = Synchronous Digital Hierarchy (Europe) • SONET = SDH SONET OC-1 OC-3 OC-12 OC-48
SDH STM-1 STM-4 STM-12
Rate 51.8 Mbps 155.52 Mbps 622.08 Mbps 2.488 Gbps
• STS = physical framing • OC = optical implementation • OC-3c = OC-3 concatenated
85
The ATM Layer
header
payload
Fixed length packet = cell
86
Anatomy of an ATM Cell 8 Byte 1 Byte 2 Byte 3 Byte 4 Byte 5
7
6
5
4
3 2 VPI VCI
GFC (UNI) or VPI (NNI)
VPI
1 Header
VCI VCI
PTI
CLP
HEC
48 Bytes
VPI: Virtual Path Identifier VCI: Virtual Channel Identifier PTI: Payload Type Indicator
Payload
CLP: Cell Loss Priority HEC: Header Error Check GFC: Generic Flow Control 87
ATM Cell Header • PTI: Payload Type Indicator • Bit 1: User Indication (EPD in AAL5, marks last cell in packet) • Bit 2: indicates congestion up-stream (FECN) • Bit 3: discriminates between data & OAM cells
• CLP: Cell Loss Priority • CLP bit is set to prioritize cell to be dropped (Policing)
• HEC: Header Error Check • GFC: Generic Flow Control
88
Virtual Paths & Virtual Channels • VPI: Virtual Path Identifier • VCI: Virtual Channel Identifier
VCs
VP
VCs
VP
Physical Transmission Link
VP
VCs
VP
VCs
• Unique on a link-by-link basis • Virtual channels are contained within virtual paths • Interpreted at each switch to: • determine output link • determine outgoing VPI/VCI • Two-level structure: • allows “trunking” of virtual channels as one virtual path • virtual path can be switched • both used to route cells through network
89
Virtual Paths & Virtual Channels VC1
VC1
VC1
VC1
VC1
VC1
VC2
VC2
VC2
VC2
VC2
VC2
VC3
VC3
VC3
VC3
VC3
VC3
VC4
VC4
VC4
VC4
VC4
VC4
VC5
VC5
VC5
VC5
VC5
VC5
VC6
VC6
VC6
VC6
VC6
VC6
ATM Switch
VP1
ATM Switch
VP1
ATM Switch
90
ATM Network Interfaces
Public NNI
B-ICI
WAN
Service Provider
Public UNI
Private UNI
WAN
Private NNI
Customer Premises
Private UNI
91
ATM Signaling ATM is connection oriented. A virtual connection must be established before cells can be sent. Connections can be established: • administratively as Permanent Virtual Circuits (PVC) (telephone switchboard operator) • dynamically as Switched Virtual Circuits (SVC) (automatic, like today’s telephone switches)
Connection establishment includes: • route set-up (path) • resource allocation (QoS - bandwidth)
92
Traffic Management Types of Traffic –CBR (Constant Bit Rate) –VBR (Variable Bit Rate) –ABR (Available Bit Rate) Traffic Management Function –Traffic Shaping –Traffic Policing –Congestion Control
93
Classes Of Service Service class Parameter Class A Timing relationship Bit rate
requiered constant
Connection mode AAL type
Class B
Class D
Not-requiered variable
Connection oriented Type-1
Class C
Type-2
Connection less
Type-3/4 atau Type-5
94
Traffic Management Adapter
Switch Work Station
Work Station
Shaping
• Bursty traffic
• Adapters shape traffic
Policing
Peak Average
• Switch polices traffic (UPC) • Dual Leaky Buckets • CLP bit
95
ATM Internetworking Network Configuration (ILMI) LAN Emulation
IP-over-ATM (Classical IP) • IP address registration • ARP (RFC 1577)
• MAC address registration • Address resolution Q.2931 Signaling • Call setup • End of connection
96
IP-over-ATM Structure ASX-200 ATM
ATM End Station
LAX-20 ATM
ATM Switch
Server
Ethernet
LAN Access Switch
LAN Host
Higher Layer Applications
Higher Layer Applications
IP
IP
IP
RFC 1577 AAL 5
LAN MAC
RFC 1577 AAL 5 ATM Layer
ATM Layer
ATM Layer
Physical Layer
Physical Layer
Physical Layer
LAN PMD
IP LAN MAC
LAN PMD
97
Agenda • Konsep Protokol dan Jaringan Komunikasi Data • Teknologi Jaringan Data • Jaringan TCP/IP – TCP – IP – IPv6
• Voice Primer (Codec)
98
Internet • An internet or internetwork is the connection of two or more distinct networks so that computers on one network are able to communicate with computers on another network. • This can hide the details of the differences of the networks. • It also allows the different networks to function as a coordinated unit. 99
TCP/IP • TCP stands for Transmission Control Protocol, and IP is Internet Protocol. • It is not vendor-specific. • It has been implemented on everything from personal computers to minicomputers to mainframes to largest supercomputers. • It is used for both LANs and WANs. • It is used by many different government agencies, commercial sites and research institutions.
100
TCP/IP Protocol Stack
Network Interface Layer
101
TCP and UDP • TCP (Transmission Control Protocol): is a connectionoriented protocol that provides a reliable, full-duplex, byte stream for a user process. Most Internet application programs use TCP which uses IP, the entire Internet protocol suite is often called TCP/IP protocol family. • UDP (User Datagram Protocol): is a connectionless protocol for user processes. Unlike TCP, which is a reliable protocol, there is no guarantee that UDP datagrams ever reach their intended destination.
102
Format Segmen TCP Bit ke- 0
15
Source Port
31
Destination Port Sequence Number
Acknowledgement Number Header Length
Reserved
Code Bits
Checksum
Window Urgent Pointer Padding
Option Data
103
Format Segmen UDP (User Datagram Protocol)
Bit ke- 0
15
Source Port
31
Destination Port
Message Length
Checksum
Data
104
ICMP and IP • ICMP (Internet Control Message Protocol): handles error and control information between gateways and hosts. While ICMP messages are transmitted using IP datagrams, these messages are normally generated by and processed by the TCP/IP networking software itself, not user processes. • IP (Internet Protocol): provides a connectionless and unreliable service. It provides the packet delivery service for TCP, UDP and ICMP. 105
ARP and RARP • ARP (Address Resolution Protocol): maps an Internet address into a hardware address. (This protocol and RARP are not used on all networks. Only some networks used it.) • RARP (Reverse Address Resolution Protocol): maps a hardware address into an Internet address.
106
Address Resolution Protocol (ARP) Saya mendengar broadcast tsb. Alamat IP tsb adalah alamat saya dan ini adalah alamat Ethernet saya.
Saya membutuhkan alamat Ethernet dari 10.14.130.4
B
A
IP:10.14.130.4=????? IP : 10.14.130.4 = Ethernet : 0800.0020.1111
• Memetakan alamat IP ke alamat Ethernet
107
Reverse ARP (RARP)
Berapa alamat IP saya ?
Saya mendengar broadcast tsb. Alamat IP nya adalah 10.14.129.5
B
A
Ethernet = 0800.0020, IP = ??? Ethernet : 0800.0020.1111 = IP : 10.14.129.5
• Memetakan alamat Ethernet ke alamat IP • Baik ARP maupun RARP diimplementasikan langsung di atas datalink layer 108
Internet Address • An Internet address occupies 32 bits and encodes both a network ID and a host ID. The host ID is relative to the network ID. • Every host on a TCP/IP Internet must have a unique 32-bit address that is used in all communication with that host. • TCP/IP addresses on the Internet are assigned by a central authority, Internet Network Information Center (InterNIC) located at SRI International.
109
IP Addresses class A
0network
B 10 C 110 D 1110
1.0.0.0 to 127.255.255.255
host
network
128.0.0.0 to 191.255.255.255
host
network multicast address
host
192.0.0.0 to 223.255.255.255 224.0.0.0 to 239.255.255.255
32 bits
110
Addresses Example 15.1.1.1
131.108.200.1
E0
E1 15.250.8.11
131.108.3.10
131.108.12.12
15.180.30.118
IP : 131.108.2.1 131 . 108 . 12 . 12
Network
Host
IP:15.6.24.2 Routing Table Network Interface 131.108.0.0 15.0.0.0
E0 E1 111
Internet Protocol Connectionless: because it considers each IP datagram independent of all others. Every IP datagram contains the source address and the destination address so that each datagram can be delivered and routed independently. Unreliable: because it does not guarantee that IP datagrams ever get delivered or that they are delivered correctly (may be lost, duplicated or out of order). Best-effort: because the Internet makes an earnest attempt to deliver packets.
112
Datagram Format
113
The Datagram Fields (1/3) • Version (4-bit): Version number, included to allow evolution of the protocol. Either header format or semantics might change. • Internet Header Length, IHL (4-bit): Length of header in 32-bit words. The minimum value is 5. Thus, a header is at least 20 bytes long, which is typical. • Type of Service (8-bit): Specifies reliability, precedence, delay and throughput parameters. • Total Length (16 bits): Total datagram length, including header, in bytes.
114
The Datagram Fields (2/3) • Identification (16-bit): Together with source address, destination address, and user protocol, it can uniquely identify a datagram. • Flags (3-bit): One bit, the More flag, used for fragmentation and reassembly. Another bit, if set, prohibits fragmentation. This facility may be useful if it is known that the destination does not have the capability to reassemble fragments. The third bit is currently not used. • Fragment Offset (13-bit): Indicates where in the datagram this fragment belongs. It is measured in 64-bit units. • Time to Live (8-bit): Measured in hops! (originally in 1 sec increments).
115
The Datagram Fields (3/3) • Protocol (8-bit): Indicates the next level protocol that is to receive the data field at the destination. • Header Checksum (32-bit): uses 1’s complement. • Source Address (32-bit): Internet address • Destination Address (32-bit): Internet address • Options (variable): Encoded the options requested by sender. • Padding (variable): Used to ensure the internet header ends on a 32-bit boundary. • Data (variable): The data field must be a multiple of eight bits in length. Total length of data field plus header is a maximum of 65535 bytes. 116
Agenda • Konsep Protokol dan Jaringan Komunikasi Data • Teknologi Jaringan Data • Jaringan TCP/IP – – – –
TCP IP IPv6 QoS
• Voice Primer (Codec)
117
IPv6 The New Internet Protocol
118
IPv6 • Initial motivation: 32-bit address space completely allocated by 2008. • Additional motivation: – header format helps speed processing/forwarding – header changes to facilitate QoS – new “anycast” address: route to “best” of several replicated servers
• IPv6 datagram format: – fixed-length 40 byte header – no fragmentation allowed 119
IPv6 Header Priority: identify priority among datagrams in flow Flow Label: identify datagrams in same “flow.” (concept of“flow” not well defined). Next header: identify upper layer protocol for data
120
Notations of IPv6 Addresses • 128 bit is represented as: – 8 integers (16-bit) separated by colons • each integer is represented by 4 hex digits
Example: FEDC:BA98:7654:3210:FEDC:BA98:7664:3210
121
Simplifications • Skip leading zeros – Example:1080:0000:0000:0000:0008:0800:200C:41 7A – is reduced to: 1080:0:0:0:8:800:200C:417A
• A set of consecutive nulls is replaced by :: (at most one :: inside an address) – the above address is reduced to: • 1080::8:800:200C:417A
122
Simplifications • Fixed format headers – no options -> no need for header length – options expressed as Extension headers
• No header checksum – reduce cost of header processing, no checksum updates at each router – minimal risk as encapsulation of media access protocols (e.g..., Ethernet, PPP) have checksum
• No segmentation – hosts should use path MTU discovery – otherwise use the minimum MTU (536 bytes) 123
Other Changes from IPv4 • Checksum: removed entirely to reduce processing time at each hop • Options: allowed, but outside of header, indicated by “Next Header” field • ICMPv6: new version of ICMP – additional message types, e.g. “Packet Too Big” – multicast group management functions
124
Transition From IPv4 To IPv6 • Not all routers can be upgraded simultaneous – no “flag days” – How will the network operatewith mixed IPv4 and IPv6 routers?
• Two proposed approaches: – Dual Stack: some routers with dual stack (v6, v4) can “translate” between formats – Tunneling: IPv6 carried as payload n IPv4 datagram among IPv4 routers
125
Dual Stack Approach
126
Tunneling
IPv6 inside IPv4 where needed
127
Agenda • Konsep Protokol dan Jaringan Komunikasi Data • Teknologi Jaringan Data • Jaringan TCP/IP – – – – –
TCP IP IPv6 QoS MPLS
• Voice Primer (Codec) 128
QoS (Quality of Service) in the Internet • The Internet Protocol (IP) does not guarantee QoS to applications • Idea: Re-engineer IP to provide quality of service – Let routers distinguish classes of flows
• Q: What is the model for a class? • Solution must consider: – the needs of (some/most/all) applications – the add’l state that routers must maintain – the add’l communication overhead (add’l packets or bits) – # of flows or classes a router must handle 129
Integrated Service and Differentiated Service
• Integrated Services (IntServ) – Resource Reservation Protocol: RSVP
• Differentiated Services (DiffServ) – Assured Forwarding – Expedited Forwarding – Comparison: AF vs. EF
130
IntServ vs. DiffServ IntServ (1992) • per-flow reservations – i.e., needs RSVP • flows provide traffic characterization • “heavy” state: per-flow
• “strong” guarantees, e.g., – conformance to leakybucket characterization [RFC 2215] – bound on max e2e delay [RFC 2212]
DiffServ (1997) • packet classification
• edge & core routers – edge: “heavy” state – core: “light” state • “weak” guarantees, e.g., – Flow A gets better service than Flow B
131
Integrated Services (1992) As described in [RFC 1633] from 1994: • Philosophy: “guarantees cannot be achieved without reservations” • Four components to IntServ architecture: – – – –
packet scheduler classifier admission control routine reservation setup protocol
132
IntServ Components (1/2) All components implemented at all routers! • Packet Scheduler – Manages forwarding of different streams – Required resources: sets of queues, timers – Example: Implementation of Weighted Fair Queuing (WFQ)
• Classifier – Maps packets to a class – Packets in same class treated similarly – Examples: • per-flow class • video-packet class 133
IntServ Components (2/2) • Admission Controller – Determines whether or not to admit a new flow – Q: why would a flow be rejected? – Requirements: • Knowledge of available resources at router • (conservative) model of flow’s resource consumption – e.g., leaky bucket – The hard part: getting apps to characterize their flows • Reservation Setup Protocol – Sets up and maintains (distributed) flows’ network resource usage • “negotiates” between admission controllers at routers • establishes active classifiers at routers – e.g., RSVP protocol 134
RSVP Protocol • The commonly suggested reservation setup protocol • Designed for multicast sessions (unicast is a special case) • Receiver-oriented: receivers initiate requests – allows for receiver heterogeneity
• Reservation styles allow merging of reservations (i.e., use the style that’s appropriate for the app) • Uses soft-state: reservations need to be refreshed or they expire. Why soft-state? • Dynamic: able to reconfigure reservation rather than perform complete teardown / setup
135
RSVP Messaging R1 S R2
RSVP state info
• Rcvrs make requests for reservations • Sufficient resources: Router reserves per outgoing interface (i.e., link) and forwards request upstream • Insufficient: send ResvError message downstream • Path messages: from sender toward rcvr so that routers know where to forward receiver requests. – Why not just head toward sender using Internet routing tables?
136
RSVP Reservation Styles (1/3) • Fixed-Filter: Allocation per sender indicated – Sample application: multimedia (e.g., send audio (S1) and video (S2) at same time)
R1 S1 S2
S1: 10 S 2: 7
S1: 10 S 2: 2
S1: 7 S 2: 7
S1: 10 S 2: 2
R2
S1: 5 S 2: 7
R3
S1: 7
requests
137
RSVP Reservation Styles (2/3) • Shared-Explicit: Allocation shared by list of senders – Sample application: multimedia (e.g., debate w/ 2 speakers) S1: 10 R1 S2: 10
S1 S2
(S1,S2: 10) (S1,S2: 10)
(S1,S2: 10)
R2
S1: 5 S 2: 5
R3
S1: 7
requests
(S1,S2: 7)
138
RSVP Reservation Styles (3/3) • Wildcard-Filter: Allocation shared by all senders – Sample application: town meeting (one sender, but not clear who the speakers might be)
S1 S2
10 10
R1
10
R2
5
R3
7
10 requests
7
139
Style Summary • Fixed-Filter: reservation per sender – Senders don’t “share” bandwidth – Dynamic event: rcvr wants to change a sender allocation
• Shared-Explicit: reservation per list-of-senders – Fixed set of senders “share” bandwidth – Dynamic event: rcvr wants to add/remove sender or change group allocation
• Wildcard-Filter: no sender specified w/ reservation – Any sender can “share” bandwidth – Dynamic event: new sender begins transmitting, rcvr wants to increase its receiving allocation 140
IntServ: Problems • Reservation protocols and structure complicated – lots of message passing – coordination problems
• All routers maintain state – state maintenance requires additional processing / memory resources – Lots of flows traverse core (backbone) routers • Lots of state: need more memory • Lots of RSVP msgs to process: slows transfer speeds • Scheduler and Classifier have too much to deal with
141
Differentiated Service (1/2) • Q: What if IntServ is too complex/costly to deploy? • A: Build a simpler scheme that takes into account – many apps have simple requirements (e.g., need fixed bandwidth, low jitter) – App can’t/doesn’t always conform to/provide “strict” model of resource usage – different levels of functionality can be placed at different “types” of routers • network edge • network core 142
Differentiated Services (2/2) H
H
H
edge router
H H
core router H
host
• Idea: keep the architecture simple within the core. – higher complexity permitted at edge routers
• Just provide service differences, no explicit guarantees – i.e., high and low priority classes (extra $$$ for high) 143
DiffServ Architecture sha per
ier f i ss a l c
mark: /data
mark:n /data edge
core
• Edge router – classify packet and mark packet – shape flow (control entry rate into core, drop pkts, change mark, etc.) • Core router – handle packet based on its mark – possibly remark at peering points • Maintain Per-Hops Behavior (PHB): the desired service (e.g. rate) provided to a class at a given hop (router) 144
Traffic Classification and Conditioning Delay/Drop outof-profile packets
In-profile/ Out-of-profile Metering
Demote out-ofprofile packets
Shaper/ Dropper
Remarking
Using meters, markers and shapers/droppers, as necessary, the Traffic conditioner ensures the TCA is adhered to.
Traffic Profile In-profile/ Out-of-profile
A Traffic Policer / Conditioner 145
Assured Service Example
Drop if congested
Uncongested Congested Assured Service
146
Premium Service Example
Drop always
Fixed Bandwidth
147
Differentiated Services Codepoint (1/4) 20 - 65536 bytes 20 - 60 bytes
Header
VER
HLEN
4 bits
4 bits
Data
DS 8 bits
Total Length 16 bits
Identification
Flags
Fragmentation Offset
16 bits
3 bits
13 bits
Time to Live
Protocol
Header Checksum
8 bits
8 bits
16 bits Source IP Address
Destination IP Address
Options
IP Datagram 148
Differentiated Services Codepoint (2/4)
Service Type
Codepoint
Differentiated Services 149
Differentiated Services Codepoint (3/4) Category
Codepoint
Assigning Authority
1
XXXXX0
Internet
2
XXXX11
Local
3
XXXX01
Temporary or Experimental
• When 3 rightmost bits are 0s, the 3 left-most bits are interpreted as the precedence bits in the Service Type interpretation. • When 3 right most bits are not all 0s, the 6 bits define 64 services based on priority assignment. 150
Differentiated Services Codepoint (4/4) • First category – 0, 2, 4, …., 62 -- assigned by IETF – contains 32 service types
• Second category – 3, 7, 11, 15, …., 63 -- can be used by local organizations – contains 16 service types
• Third category – 1, 5, 9, …., 61 -- temporary. Can be used if First category is used up – contains 16 service types BecauseService then it would incompatible TOS interpretation • Note: typebecome numbers are not with contiguous. Why? of XXX000. Numbers 0, 8, 16, 24, 32, 40, 48, 56 would fall into all 3 categories.
151
2 Competing PHBs • Expedited Forwarding (EF) [RFC 2598] – Router must support classes’ configured rates – EF class allocated fixed portion of router processing per unit time, e.g., • Class-based queueing (CBQ) w/ priority to EF queue • Weighted Fair Queuing
• Assured Forwarding (AF) [RFC 2597] – N classes (current standard: N=4) – M possible drop preferences w/in class (current standard: M=3) – Each classes’ traffic handled separately – Packet drop “likelihood” increases w/ drop preference 152
PHB Specs Omit... • EF and AF PHBs do not specify mechanism, e.g., not specified: – edge classification, shaping or marking policy – core router queuing mechansim – ranges of rates, relative class/preference service ratios, etc.
• Why are these details omitted? – Allow flexibility - as long as specified requirements are met. – DiffServ is a new idea - still unclear on which mechanism is best - so standardize later
• Which is better, EF or AF? 153
Comparing PHB Models [Sahu] • How does isolating traffic (EF) compare with preferential treatment (AF w/ preferences)? • Measures: – expected loss rate – expected delays • Reqm’t: overall buffer / bandwidth fixed high priority traffic low priority traffic
• Queue models: – EF: separate queues per class. High priority queue always serviced first (when non-empty) – AF: one queue w/ threshold for accepting high drop-preference pkts high & low priority traffic high drop-preference threshold 154
Intuition high priority traffic low priority traffic
high & low priority traffic high drop-preference threshold
• Which is – better for reducing high-priority delay? – for high-priority loss? • How should buffer be allocated in the 2 models to make a fair comparison? – EF: low priority gets its own buffer – AF: low priority must share its buffer with high
155
EF vs. AF Comparison • Choose buffer partitions and threshold such that low-priority traffic sees similar loss rates in two systems • Examine impact on high priority traffic • Main Results for high priority traffic: – AF router needs to process 30-70% faster than an EF router to maintain same delays (function of partition point and threshold location) – EF router needs only 15% add’l buffer to yield same loss rates to low priority traffic as AF
156
DiffServ Open Issues • How to decide “how much” to reserve • How to do DiffServ for multicast – Much more complicated – Multicast reservation issues significantly complicated IntServ. What about DiffServ?
157
Agenda • Konsep Protokol dan Jaringan Komunikasi Data • Teknologi Jaringan Data • Jaringan TCP/IP – – – – –
TCP IP IPv6 QoS MPLS
• Voice Primer (Codec) 158
Multi-Protocol Label Switch (MPLS)
159
Latar Belakang Pengembangan MPLS
• Pelanggan Carriers/Telco • Layanan Frame Relay dan ATM menginginkan layanan IP: menyediakan : – Layanan IP connectionless. – Layanan Connection oriented – Dibutuhkan jaminan QoS IP yang lebih fleksibel. – Bandwidth point-to-point tidak fleksibel – Dibutuhkan layanan yang lebih ‘privacy’ daripada – Memiliki ‘privacy’ yang baik. yang disediakan Internet. 160
Solusi : Multi-Protocol Label Switching (MPLS) • Mekanisme yang memberikan aspek terbaik dari dua dunia: – Privacy dan QoS ATM, Frame Relay – Flexibilitas and skalabilitas IP • Dasar untuk IP business services – Flexible untuk membentuk kelompok user dan value-added services • Managed IP services yang efisien – Skalabel untuk jaringan privat besar dan kecil 161
Konsep MPLS • Mengintegrasikan switching label dengan routing network layer. – Traditional routing • Setiap router memiliki seluruh tabel routing dan meneruskannya ke hop berikutnya (destination based routing)
– MPLS mengkombinasikan routing L3 dengan swapping dan forwarding label – MPLS Forwarding • Label ditempelkan pada router ingress • Seluruh keputusan berdasarkan label (tidak diperlukan ‘table lookups’)
162
Tujuan Pengembangan MPLS • Tujuan –Meningkatkan skalabilitas network layer –Switching kecepatan tinggi –Kemampuan rekayasa trafik
163
Kemampuan Kunci MPLS RSVP
IP Multicast
IP CoS
Integrasi IP/ATM
Rekayasa Trafik
Ethernet Ethernet
ATM
Membentuk VPN-VPN
ATM
L2-VPN’s 164
Traditional IP over ATM
• Menempatkan router di sekitar perbatasan jaringan ATM. • Terhubung ke router menggunakan Permanent Virtual Circuits. • Dengan cara ini tidak diperoleh integrasi yang IP dan ATM yang optimal. 165
Integrasi IP+ATM • Skalabilitas routing internal –Limited adjacencies
• Skalabilitas routing external –Mendukung penuh BGP4, dengan seluruh fitur-fiturnya
• VC menyatu untuk jaringan yang besar 166
Label Switching Devices Label Switching Routers (LSRs) (ATM Switch or Router)
Label Edge Routers
Label Switching Example (1/3) Destination-Based Routing Module Address Prefix Interface
Address Prefix Interface
128.89.10
1
128.89.10
0
171.69
1
171.69
1
...
128.89.10
...
i/f 0 i/f 1 i/f 1
Advertises Reachability to 128.89.10 and 171.69
Advertises Reachability to 128.89.10
171.69
Advertises Reachability to 171.69 168
Label Switching Example (2/3) Address Prefix Interface
Address Prefix Interface
128.89.10
1
128.89.10
0
171.69
1
171.69
1
...
128.89.10
...
i/f 0 i/f 1
Advertises Binding <5,128.89.10> Using LDP
i/f 1
Advertises Bindings <3,128.89.10> <4,171.69> Using LDP
171.69
Advertises Binding <7,171.69> Using LDP LDP=Label Distribution Protocol 169
Label Switching Example (3/3)
Local Remote Address Label Label Prefix Interface
x
3
128.89.10
1
x
4
171.69
1
Local Remote Address Label Label Prefix Interface
3
5
128.89.10
0
4
7
171.69
1
...
128.89.10
...
0
1 171.69.12.1 data
1
4
7
171.69.12.1 data
171.69.12.1 data
‘Edge’ Router Does Longest Match, Adds Label
Subsequent Routers Forward on Label Only
171.69
Rekayasa Trafik (1/2) • Mengapa rekayasa trafik ? Optimalisasi utilisasi link Jalur khusus untuk kastamer atau kelas service
Route yang dipilih oleh protokol routing IP
Route yang ditentukan dengan rekayasa trafik
Menyeimbangkan beban trafik • Trafik mengikuti jalur yang telah ditentukan • Jalur dapat berbeda dari jalur route biasa • Mengontrol aliran paket melalui infrastruktur jaringan L2 atau L3.
171
Rekayasa Trafik (2/2)
• Rekayasa Trafik MPLS tidak menawarkan QoS tambahan yang dapat dirasakan end-user. • Rekayasa Trafik akan mengurangi biaya penyediaan layanan kepada end-user (mis. Diff-Serv) dengan cara mengoptimalkan pemanfaatan resource yang ada. • Dapat memperbaiki QoS. • RT MPLS memanfaatkan karateritik “connection-like” nature of MPLS to distribute traffic based on Bandwidth demand/use • like current Voice Traffic Engineering
IP Routing & “the Fish” R8 R3 R4 R2
R5
R1 R6
R7
IP (Mostly) Uses Destination-Based Least-Cost Routing Flows from R8 and R1 Merge at R2 and Become Indistinguishable From R2, Traffic to R3, R4, R5 Use Upper Route Alternate Path Under-Utilized 173
Tunnel Rekayasa Trafik R8 R3 R4 R2
R5
R1 R6
R7
Labels, like ATM VCs can be used to establish virtual circuits which are “Qos Routed” Normal Route: R1->R2->R3->R4->R5 TE Tunnel:
R1->R2->R6->R7->R4->R5 174
Mekanisme RT • Tunnels RT perlu diroutekan secara otomatis • Dilaksanakan Constraint Based Routing dimana yang termasuk constraints antara lain: – Bandwidth yang dibutuhkan suatu tunel versus bandwidth yang tersedia di semua link – Policy constraint yang dikonfigurasi oleh Operator (mis. Suatu jenis tunnel tidak boleh menggunakan tipe link tertentu)
175
Contoh Deployment RT Find route & set-up tunnel for 20 Mb/s from POP1 to POP4 Find route & set-up tunnel for 10 Mb/s from POP2 to POP4
WAN area
POP4
POP1 POP
POP2 POP POP
MPLS VPNs (1/2) Connection-Oriented VPN Topology
• Private, connectionless IP VPNs VPN B
• Outstanding scalability • Customer IP addressing freedom • Multiple QoS classes • Secure support for intranets and extranets • Simplified VPN Provisioning • Support over any access or backbone technology
VPN A VPN C
VPN C
VPN B
VPN A VPN A
VPN B VPN C
VPN C VPN B
VPN A
Connectionless VPN Topology
VPN B
VPN A VPN C
VPN C
VPN B
VPN A VPN A VPN B VPN C
VPN C VPN A
VPN B 177
MPLS VPNs (2/2) Separately engineered customer private IP networks
Build once, Sell once
vs.
Single carrier network supporting multiple customer IP VPNs
BGP+once, MPLS Build Network Sell many
Return 178
Agenda • Konsep Protokol dan Jaringan Komunikasi Data • Teknologi Jaringan Data • Jaringan TCP/IP • Voice Primer (Codec)
179
Komunikasi Suara PSTN
Sinyal Analog
Sinyal Digital
Sinyal Analog
010
time
time
time
- Pada LE terjadi digitasi sinyal suara: sinyal analog à sinyal digital - Noise pada sinyal digital lebih mudah dihilangkan - Data rate sinyal digital PSTN adalah 64 kbps, terlalu besar untuk digunakan dalam jaringan IP 180
Digitalisasi Sinyal Suara • Tujuan: – Mentransformasikan format sinyal suara analog menjadi sinyal digital agar sesuai dengan jaringan yang membawanya dan mudah untuk diolah lebih lanjut
• Proses generik : – Sampling – Kuantisasi – Coding
181
Sampling • Mengambil beberapa nilai yg dianggap mewakili • Teori Nyquist: frekuensi sampling minimal 2x frekuensi suara. Umumnya 8000 sample/dt, atau 125 mikro det setiap sample. • Makin tinggi frek sampling, kualitas makin baik
Sinyal suara analog
Tahap sampling 182
Kuantisasi • Membagi nilai amplitudo sample sinyal suara ke dalam beberapa tingkatan diskrit. • Semakin rapat tingkatan yang dibuat, kualitas semakin baik, tetapi semakin banyak bandwidth dibutuhkan.
Segmen 1
Segmen 0
183
Contoh: Kuantisasi A Law 128 112 96 80 64
Segment--7 Segment
48 32 16 0
1 Volt
-1 Volt
Ada 16 interval per segment
1/2 1/4
Gambar-6 : Kuantisasi ALaw.
1/8
184
Coding • Pemberian kode unik kepada setiap tingkatan diskrit pada proses kuantisasi. • Jumlah bit pengkodean tergantung jumlah tingkatan. Diupayakan jumlahnya seminimal mungkin • Contoh: Polaritas : 1 bit Segmen : 3 bit Step : 4 bit
P
Se
Se
Se
St
St
St
St
Total : 8 bit setiap sampling nya
• Pada PSTN: 8 bit untuk setiap 125 mikro-sekon; Dalam 1 detik terdapat 8000 sample atau dibutuhkan 64.000 bit tiap detiknya
185
VoIP membutuhkan Kompresi • Pada jaringan Internet, user sangat jarang mendapatkan bandwidth yg besar (<< 64 kbps) • Sinyal suara digital PSTN harus diperkecil ukurannya • Keuntungan kompresi: – Mengurangi konsumsi bandwidth – Mengurangi biaya transmisi sinyal suara
• Kerugian kompresi: – Terjadi distorsi suara – Terjadi delay pada proses kompresi/ dekompresi 186
Voice Coding pada VoIP • Mencakup proses digitalisasi suara dan kompresi • Pada konfigurasi H.323, voice coding dilakukan di terminal atau gateway (tergantung konfigurasi VoIP yg digunakan)
187
Teknik Kompresi Suara • Algoritma Wave form vPCM: Pulse code modulation vADPCM: Adaptive differential pulse code modulation
• Algoritma Source vLD-CELP: Low delay, code-excited linear processing vCS-ACELP: Conjugate structure algebraic code-excited linear processing 188
Mean Opinion Score (MOS) (1/2) • Mean Opinion Score atau MOS adalah suatu cara yang biasa digunakan untuk mengukur performansi kualitas suara. • Test MOS juga digunakan untuk membandingkan bagaimana sebuah CODEC bekerja dalam berbagai keadaaan pada saat itu, termasuk perbedaan noise level, sistem multiplex, dll. • Harga tertinggi memiliki score 5, sedangkan yang terendah 1,
189
Mean Opinion Score (MOS) (2/2) Channel Simulation
Impairment Codec X
11
22
33
44
55
11
22
33
44
55
Test statement: “Nowadays, a chicken leg is a rare dish”
Rating
Speech Quality Level of Distortion
5
Excellent
Imperceptible
4
Good
Just perceptible but not annoying
3
Fair
Perceptible and slightly annoying
2
Poor
Annoying but not objectionable
1
Unsatisfactory Very annoying and objectionable
MOS of 4.0 = Toll Quality 190
Rating MOS untuk Digital Voice Codec
Bit MIPS Comp. Framing MOS Rate Delay (ms) Size (ms)
G.711
PCM
64
0.34
0.75
0.125
4.1
G.726
ADPCM
32
13
1
0.125
3.85
G.728
LD CELP
16
33
3-5
0.625
3.61
G.729
CS-ACELP
8
20
10
10
3.92
G.729a
CS-ACELP
8
10.5
10
10
3.9
G.723.1
MPMLQ
6.3
16
30
30
3.90
G.723.1
ACELP
5.3
16
30
30
3.8 191
Thank You
192
