CATT Seminar on Networks Research Polytechnic University March

CATT Seminar on Networks Research Polytechnic University March 27, 1999 Next Generation Networks Richard D. Gitlin Chief Technical Officer and Data Networking Technology Vice President Data Networking Systems Lucent Technologies rich@lucent. com

Next Generation Networks • Introduction – The Network Revolution – Technology Trends – Applications and Requirements • Issues and Solutions – – – • • Quality of Service Security Network Management High Reliability Intelligent Networking Example: Voice on the Next Generation Network Summary

This R/Evolution Is Fueled By Unparalleled Customer Demand (and by telecom deregulation and the Internet) Worldwide Access Lines Global Internet Users 3 B 250 M 2 B 1 B Changing Traffic Patterns 134 M Internet Session 20 - 30 minutes Voice Call 3 minutes 30 M Average Hold Times 1994 1998 2001 · It took about a century to install the world’s first 700 million phone lines; an additional 700 million lines will be deployed over the next 15 -20 years · There are more than 200 million wireless subscribers in the world today; an additional 700 million more will be added over the next 15 -20 years · There are more than 200 million Cable TV subscribers in the world today; an additional 300 million more will be added over the next 15 -20 years · More than 100 million additional Internet users will come on-line by 2001 ---the Net is experiencing a 1000% per year growth! If this trend continues, by 2004 99% of the world’s bandwidth will be Net traffic ---including computer-to-computer communications.

Next Generation Networks (The New Public Network): Current situation • • • No longer any debate that wide-area networks based on packet technology will emerge as a compelling alternative to the PSTN The new public network will be optimized for IP-based applications and will become the platform for future voice and data service innovations---it will not be based on merging existing legacy voice and data [frame relay, SMDS, IP, …] networks Carriers expect that the simpler new network will also reduce costs of operations, equipment and staff and will capitalize on the faster pace of networking element development Migration strategies, quality of service (Qo. S), network management, security, rapid service creation, and reliability are the major concerns of the carrier --as well as the almost $1 Trillion invested in the PSTN Almost 80% of the service providers intend to build their multiservice network with an ATM core and about 20% based on IP Some principles for the new network – Give customers access choices (DSL, cable, wireless, ISDN, …) – Work hard to optimize IP switching (Diff. Serv, MPLS, RSVP, …. ) – Separate service intelligence from the network transport ---open interface between intelligent call control features and packet gear – Build IP-based billing and management

A Networking Paradigm Shift Occurring Separate (IP Becomes Dominant WAN and LAN Protocol) Circuit Switched Network Separate Data Networks (Frame Relay, X. 25, ATM, Router) Single Network Supporting Voice & IP Endpoints • Next-generation data networking –Excellent performance with IP –Qo. S breakthroughs: wire speed and per flow control –“Route once, switch often” Route at wire speed –Distance transparency and distributed “computing” –Policy driven network management –Directory Enabled –Broadband access –Wireless and optical networking –Silicon and software • Data on voice (circuits) Voice on data (circuits) • “ 80/20” Enterprise/WAN data traffic split “ 20/80” • Networks Network of networks

“Convergence” Driving Change & Qo. S PSTN DB More than moving voice over the Internet • Converged, multi-service networks – reduce costs – provide integrated services • Voice over cell/packet solutions -- Vo. ATM and Vo. IP • Virtual Private Networks -- VPNs • Quality of Service -- Qo. S • • • IP DBs PSTN DB SS 7 LEC IP IP LEC “IP” Network Media Gateways, Controllers Accommodate multiple protocols (e. g. , IP, ATM, frame relay) Provide at least today’s voice services (e. g. , 3 -way connections, hold, add, forward, toll free, 911) Interoperate with one another, the Internet and the Public Switched Telephone Network The real challenge is to build converged networks that are as reliable, robust and scalable as voice networks

Convergence of Communications Paradigms Leads to New Services and Requires New Technologies • Voice over IP • Virtual Private Networks • E-Commerce Data Communications p Connectionless p Loosely Coupled • Video & audio streaming, conferencing…multi-media • Mobile Access Telecommunications s Connections Applications s Tightly Coupled p Loose Controls, Distributed s Centralized Controls p SW Fault Tolerance s HW Fault Tolerance p Features During ‘Session’ Common Infrastructure s Features At Call Set-Up p Little Attention To Qo. S s Obsession With Qo. S p High Latency s Low Latency • Qo. S: Diff. Serv, MPLS, Qo. Saware Switches • Multi-Protocol Support: ATM (CBR, VBR, UBR, ABR), IP Over FR/ATM • Multicasting • Manageability & Intelligent Networking: Policy Driven Nets • Security • High Reliability • Scalability

The Pace of Technology Trend l Silicon Chips X 2 in density/speed every 18 -24 months l Optics X 2 in transmission capacity every year l Data/Web X 2 Internet subscribers every 2 -3 years X 2 Internet hosts/servers every year l Wireless X 1000 in capacity in 5 years l Power X 2 MIPs/MW every 2 years (DSPs) l Compression X 2 in information density every 5 years

Disruptive Technologies and their Impact on Networking • Access: – Mbps (home) and Gbps (office) will substantially increase data x. DSL traffic via x. DSL, cable modems, wireless, and optics Fiber • Semiconductors: Atomic-scale transistors will mean - 64 Gb DRAM, 10 GHz processor clocks and giga-instructions/sec (GIPs) Fiber Fixed Wireless - Heterogeneous and multi-protocol functions on a chip reduce power/cost - wire speed processing in data networks • Optical networking: WDM-fueled bandwidth explosion will - trade bandwidth for network complexity - lower risk with new networking solutions (e. g. , IP WDM) • Enterprise 1 IP IP Access WAN Integrated Services Node Communications Software: Will spawn - High performance databases/directories supporting advanced network RF Access ATM Access Enterprise 2 ATM features (e. g. , policy servers) - Speech recognition, media conversion (e. g. , text-to-speech), and network agents to realize value-added intelligent networks Cellular

Impact of Transmission Speeds on Networking Available WAN bandwidth has been less than LAN bandwidth --- this situation is expected to change at the millennium (WANs no longer a bottleneck for leading edge customers) – Fiber optic transmission speeds have increased by 50% per year since 1980 (x 100 in 10 years) – LAN bandwidth has increased at 25% per year and WAN bandwidth has remained expensive (shared) – “Available” curve purchased by leading-edge users (e. g. , OC-3 c); about 1% of WAN BW LAN Single Channel Fiber 105 Multi-Channel (WDM) Available Mbps • 104 Gigabit Ethernet 103 Fast Ethernet 102 OC-3 c T 3 10 Ethernet T 1 1975 1980 1985 1990 1995 2000

Impact of Speeds of Fiber Transmission and Microprocessors on Networking Speed gains for microprocessors have kept pace with fiber transmission speeds The number of instructions available to process an optically transported packet, using the “hottest” micro has remained constant Microprocessor speed (Mhz) Single Channel Fiber 105 Mbps or Mhz • • Multi-Channel (WDM) 104 Merced 103 Pentium III Pentium II 102 Power. PC 486 10 386 286 1975 1980 1985 1990 1995 2000

Impact of DRAM Memory Size and Transmission Speeds on Networking Mbps or k. B • With increasing transmission speeds, more packets are “in flight” for a given round trip propagation time; common error recovery protocols require that one round trip worth of data be stored • e. g. , NY-LA-NY round trip propagation time of 50 ms results in 1 MB for a 155 Mbps link • Size of DRAM increasing 58% per year – Effective BW of memory is increasing at about 40% • Storage capacity and transmission speeds are increasing at the same rate, thus number of chips to hold one “window” of data has remained constant DRAM Size Single Channel Fiber Multi-Channel (WDM) 106 256 MB 64 MB 105 16 MB 104 4 MB 103 102 10 1975 1980 1985 1990 1995 2000

Much More Traffic (leads to much more traffic --- Metcalfe’s Law) US Businesses WAN Peak Capacity Will Need to Increase at Least 10 X in Three Years 5. 0 4. 0 Tb/sec 3. 0 2. 0 1. 0 0. 0 1997 1998 Source: Estimated from projections of data port shipments (Dataquest, 12//97) 3. 5 Billion 1997 wth of ar Gro ges 3 Ye Messa Email 1999 2000 56 Billion Year 2000 Source: email projections: [Yankee Group] Metcalfe’s Law: the value of a network grows exponentially with the number of users and connected sources and a “network of networks” becomes the organizing principle for most communications

Major Requirements for Next Generation Network Applications will require: • Qo. S and security for successful convergence • Virtual Private Networks for converged networks and Qo. S • Network management directories, policies and intelligent agents for decision support, configuration and Qo. S

The Leading Protocols for Transporting Information on Next Generation Networks Are ATM and IP Economies of scale favor IP *Related Items

Issues to Be Solved for Next Generation Networks: Qo. S Issues Guarantees beyond Availability · Dial Access Blocking · Maximum Delay & Jitter · Minimum Effective Bandwidth Qo. S Guarantees Application & Source Performance Issues (e. g. , Latency, Jitter) Approaches · · Individualized SLAs by · Class of Service (Application) · Customer or groups of customers (VPN) · Flow or connection · · Reduction of large frequently encountered latency and response time Efficiency of network traffic · · · Allocation of dial ports per VPN or service Static (SLAs) & Dynamic (RSVP) Qo. S Requests Resource reservation (provisioning, MPLS explicit paths, RSVP) Use of Qo. S aware network elements Differentiated Services Integrated Services Classification, large multi-priority buffer pools and buffer management Edge vs Core congestion control Policing , shaping, marking Caching Network and Server Load Balancing Efficient Multicasting Mirroring Firewall/Proxy Server Farms Private Peering Agreements

How Will IP Networks Approach the Performance of ATM Networks? SLA The Past Reliability Blocking • • The Future Dynamic SLA Reliability Blocking Latency Jitter Loss Implementing wire speed switches Decreasing effect of IP packet variability and header size with transmission of higher speeds Selecting good designs and paths with VPN Designer expert system Making IP connection oriented via MPLS, per flow queueing Implementing Qo. S infrastructure akin to PNNI Using policies and directories to enable Qo. S Exploiting ASICS for congestion control directly on flows Executing congestion control within core instead of at edge

Next Generation Switches VPN Designer (Central) SLAs VPN Manager System (Distributed) VPN Designer (Distributed) ATM Switch Site 1 SR L • • SR Label Switching L Router SR L Site 2 Site 3 Wire speed traffic classification and filtering – No performance degradation when filtering or Qo. S is switched on Complete traffic isolation: – Can meet Service Level Agreements without the need for over-provisioning Guaranteed minimum bandwidth based on source address, destination address, protocol and/or TCP/UDP port numbers Hierarchical Weighted Fair Queuing

Decreasing Effect of IP Packet Variability and Header Size (Example Application: Voice over ATM vs. Voice over IP) Situation • Large IP packets cause longer delays than short ATM packets • Variable IP packets create more jitter than fixed ATM packets • 20 Byte IP header causes less economic efficiency than 5 Byte ATM header (Voice over ATM) Natural Solution • IP Performance and Economics Comparable at Speeds beyond OC-12

Make IP Connection Oriented via MPLS. . . VPN Designer (Central) SLAs • Translate SLAs for Configuration • Determine Qo. S Paths VPN Manager System (Distributed) VPN Designer (Distributed) ATM Switch Site 1 SR L • Determine and Propagate Enterprise & Network Topology • Translate SLAs for configuration • Determine Qo. S Paths • Set up ATM VC or MPLS Label Switched paths • Classifies incoming traffic (IP header, port, DS byte) • Forward/route traffic based on forwarding/routing table SR Label Switching L Router SR L Site 2 Site 3 Ÿ Allows Qo. S path optimization Ÿ SLAs are easy to implement. Ÿ Facilitates identifying individual flows Ÿ Can be used with IP, ATM, SONET, WDM, . . . Ÿ Supports multi-vendor environments Ÿ Complements Enterprise need for tunnels Ÿ Will require building Qo. S capabilities into OSPF, LDP, RSVP protocols

IP With MPLS and IP Over ATM For IP Qo. S Guarantees *ERLSP=Explicitly Routed Label Switched Path

Congestion Control of Bad Behavers: Value of Isolating Flows in Qo. S Management VPN 1 And VPN 2 Have The Same Contract ( 0. 4 of the DS 1 capacity) VPN 2 uses 0. 52 of the capacity (i. e. , 30% more than contract) Benefit of Isolating Flows Maximum delay in ms 70 70 60 50 VPN 2 40 VPN 1 30 20 Price of Not Isolating Flows 60 VPN 1 (without flow isolation) 50 40 30 VPN 1 (with flow isolation) 20 10 10 0 0. 1 0. 2 0. 3 Utilization of VPN 1 0. 4 Both at same priority with routers using flow isolation ( by VPN) and equal weights for the two VPNs 0. 5 0 0. 1 0. 2 0. 3 0. 4 0. 5 Utilization of VPN 1 Both at same priority with no discrimination • Without flow isolation, all VPNs get unacceptable delay when one creates congestion • With flow isolation, all well behaving VPNs get acceptable delay • With flow isolation, misbehaving VPNs can get acceptable delay only when other VPNs well below contracted load

Reducing Latency: Web Access With Next Generation Caching www. cnnfn. com www. lucent. com www. yahoo. com PULL Router L 4 Request Trap Request Central Cache Control http Cache Sites Multicast Load Balance Requests Reply Client Deploy cache sites in: --- NAP --- Backbone network --- Data center --- ISP --- POP --- Enterprise Request Current Situation • High End-to-end latency • High Network load • High Server load • High Cost for ISP and Enterprise Solution Principle: Move content closer to users – much lower web access latency – reduced network congestion – higher content availability Next Steps – pre-fetch “hot” objects – multicast to cache sites – load balance cache sites – high level trap of cache request – support “streaming” multimedia – cache dynamic content – support value-added services

Reducing Latency With Multicasting Current Situation • Redundant traffic causing needless loading of network and servers • Results in unacceptable latency Solution: Reduced Latency via • Reduced traffic on core network • Reduced load at data source server • Data closer to receivers • Combination with caching and replication Obstacles to Overcome • Lack of unique set of protocols • Data synchronization • Reliability, Recovery from lost data • Current implementations too static Multicast Group Data Receiver Multicast Cooperative Server Data Receiver Core Network Data Receiver Data Source Data Receiver Multicast Group Core Network Data Source Multicast Group Data Receiver Multicast Cooperative Server Data Receiver

Issues to be solved for Next Generation Networks: Security

Requirements for Access to VPNs PPP ISP RAS Certificate PPP/L 2 TP Server Authentication Server IPsec ISP Internet LNS Authentication Server • • R Internet IPSec. R RADIUS Dial: Telecommuters and remote office access to a corporate site VPN Requirements • Certificate Authority Dedicated: Branch office access to a corporate site Private Addressing: to allow access to corporate network resources (Tunneling and Network Address Translation) Security: authentication of users and privacy of user data as it goes over the network (RADIUS/DIAMETER, Tunneling) Legacy Protocols: allow user to use non-IP protocols (e. g. IPX, Apple. Talk) over an IP network (Tunneling) Performance: provide a level of performance comparable to that of private networks (Qo. S) Network Management: provide customer management of the VPN (monitoring, reconfiguration, . . ) Issue: Tunneling addresses many VPN requirements but makes Qo. S more difficult since flow information becomes hidden in the core

Evolving Tunneling Options SERVICE PROVIDER USER CORPORATE NETWORK IP-IP PC L E C RAS = Remote Access Server (modem pool) LAC = L 2 TP Access Client LNS = L 2 TP Network Server RAS/ LAC ISP Backbone RAS/ Router/ LNS Host Firewall RADIUS Server L 2 TP IPsec

Issues to Be Solved for Next Generation Networks: Network Management

Historical Network Management/Policy Paradigm Device Manager (NMS) Device Manager (2) SNMP Data store Agent Network Device NVRAM Current paradigm has following problems: • Individual Device management • Device Manager per vendor • Device Manager product • No unified configuration store • Network Manager and Device have Client-Server model and are not peers

Evolving to Next Generation Network Management Near Term Current Situation • Directories drive data • Independent device and unification independent services • Central policy management • Table-driven device functions on service basis • Dynamic device functions • Client(NM)-Server(Device) • Policy agents added architecture • SNMP Network Management Technology Specific Configurations DNS/DHCP Radius SNMP Network Device Network Management Technology Policy Servers Policy Administration Policy Distribution COPS Radius The Future • Distributed policy management • Integrated services through policies • Reactive agents added • Complex & reactive policy capabilities LDAP DNS/DHCP Network Device Business Policy Servers Policy Support Services (VPN Designer) Network Device

Complex Networks and New Dynamic Services Drive Changes to Policy Management and Infrastructure Issues • Management is device configuration; needs to be offer & service related • Associated data is per device per vendor and largely in tables; needs to be integrated and for the offer or service • Data inconsistency and synchronization problems since data repeated for devices • Management rules need to respond to changes in network conditions Software Solutions • Technology Policy Service Policy • Protocol Based Management Tables Common Information Model • Configuration Policy Management • Provisioned Dynamic Reactive Policy Unified Distributed Centralized Configuration Policy Management Monitoring Unmanaged Networks Self-healing Networks Static Filter Tables Device Management Dynamically Updated Filter Tables Procedural Policy Agents Network Management Reactive Policy Agents Policy Management

Directory Evolution: Near Future Directory LDAP Data store Directory Management Interface Data store Meta-Directory Address Policy Server Qo. S Policy Server DHCP COPS Network Device Meta Directory Solution • All directory changes are arbitrated through the Meta-Directory • Meta-Directory maintains consistency between information in each physical directory/database – Appearance of a single directory to Network Manager –Single entry link to other directories Meta-Directory Is A Band-aid • Does not resolve any overlapping schema issues

Network Management (The Future): Supporting Complex and Reactive Policies Are Represented as Scripts Solution • Policy scripts Directory Configuration Activities LDAP Directory Access Client Policy Interpreter and Processor Decision Support Info Policy Directory Distribution Manager Access (PIP) Client Filter Tables RTOS Network Device • Network Device uses Directory for configuration • Policy Server uses Directory for decision support and policy storage Management & Decision Support Config Data – Distributed by Policy Server – Interpreted by Network Devices – Alternative to COPS/DIAMETER Policy Server • Policy Server and Directory Access Client both manipulate device data structures

Example Voice over IP Application: What is Required to Support Vo. IP With Qo. S?

Voice over IP (Vo. IP) Architecture Requirements • Today’s products do not scale well. Need to separate signaling from media transport and control for large scalable networks – Media Gateways ~ 1000’s – Media Gateway Controllers/Gate Keepers ~ 10’s, – Signaling Gateways < 10 • Today’s solutions do not interface with value added feature data bases or Signaling Control Points (SCPs). Voice feature support requires interaction with existing and future SCPs such as – Local Number Portability (LNP), 800, SDN, . . . • Vo. IP is growing much faster than multimedia over IP. Thus, focus on voice protocol simplification first. • Commercial Success of Vo. IP (including VPNs) will require Qo. S – Call Admission – Media Transport

Near Term Evolution of Vo. IP Architecture: New Vo. IP Functional View Existing Voice New Vo. IP Data Bases, Feature Servers • User Authentication. Servers • Local Number • Accounting • Routing Portability • SDN • 800 SS 7 Net L E C LDAP/IP* , RADIUS TCAP/ IP TCAP/ SS 7 D-Channel Signaling Translation Gate Keeper H. 323+, SIP H. 323++/SIP+ Gateways Between Domains Call SS 7 Gate Keeper Call Control Media GW Functions Controller TBDMedia GW Control Signaling Gateway Existing Voice Feature Servers Controller Functions D-Channel Signaling Translation Signaling Gateway MGCP/MDCP/H. gcp* Media Gateway RTP/ T 1, PRI Voice Circuit UDP/ to IP, IP Connection Ethernet ER ER Media Gateway SS 7 Net L E C Voice Circuit to IP Connection “IP” Network Challenges (Mainly Due to Number of Devices) • Call Set Up Time • Reliability • Voice Quality • Qo. S Guarantees • Network Management • Cost/Minute * Proposed Protocol H. 323+ = H. 225+ & H. 245 H. 323++ = H. 225+, H. 245 & Annex G

Requirements for Future Qo. S Vo. IP Architecture • Qo. S Aware Network Elements • Qo. S Protocols • MPLS, RSVP, LDP in IP Network • 802. 1 p on Ethernet LANs • Diff. Serv on IP • Call/Connection Admission Control • Qo. S Policy • Qo. S Network Management CAC DS Gate Keeper DS 802. 1 p SS 7 Signaling Gateway Media GW Controller CAC Qo. S Policy, Manager Media GW Controller SS 7 Signaling Gateway 802. 1 p LEC 802. 1 p Media Gateway DS CAC “IP” Network CAC=Call/Connection Admission Control DS=Diff. Serv Byte in IP Header LEC

Summary: What to Expect in Transition to the Next Generation Network • Data applications dominate network traffic – – • • Multimedia, collaborative systems have increased acceptance Network driven to data networking solution Data network must also support voice applications and Must interwork with Public Switched Telephone Network (PSTN) Rapid new technology decreases cost; increases capabilities Network is packet based – Packet voice technology widely utilized • • Need to provide Qo. S, Security, Network Management … Intelligent, wire speed, Qo. S enabled switching elements for better efficiency and control Data networks achieve reliability comparable to voice networks Vendors provide solutions that – work in heterogeneous, multi-vendor environments – allow rapid introduction of new services – allow customers to provide service differentiation