S
ONS2014
Content Take Away
John
Donovan,
AT&T
How SDN enabled innovations
will impact AT&T’s plans to
transform it’s infrastructure
Openness
Transformation & Innovation
Flexibility + Simplicity + Speed
New Resources Model in vM
Redefining end point
Time required to provision services
SDN + NFV
Telecom Network
Running on vM Simpler
Scalable
AT&T
www.at&tmorationsp
ace.com
Build capabilities in
AT&T Cloud for real
time services
Mobile data traffic
has increased
50000% over the
last 7 years
Enhance business
efficiency
Less CAPEX &
OPEX
Transformation @ AT&T
Similar to Cloud Computing
(Network Design)
Common Infrastructure,
Consumption model
Open Network principle >100 vendors collaboration
Domain 2.0 DODO
Budget for new architecture
Ericson, Affirm Network -
ecp, Metal Switch Core, Tel-
ef – control software: open &
flexible & work with existing
devices
Open to new Ideas and
maintain competitiveness
Doing with the WAN as the world has done to Data Centre
cloud centric workloads
Elasticity to Scale
Provision services
in WAN network
Open Process
Agile development
Domain 2.0 DODO
2014 Domain 1.5
S Control on existing platforms – extend utilizing
New Platform in 2015 - NO overlay network – distributed cloud
2015 4600 Data Centre
S microseconds of compromise that exist between middle and applications will be
overcome by the milliseconds they avoid by highly distributed, very fast fiber network
OSS
S APIs – Do Policy & Provisioning
Software
S Orchestrate in highly distributed network
Why running their services on
your network?
S Platform Extra
S move security into cloud
S system getter smarter as components get smarter through
communication
S Privacy & Liability Guidelines
S Clarify boundaries – syntactic data build capability without
touching the raw
S 3 new technologies – control plane role, library, routing system
for data object – abstract syntactic data– no compromising
customers
Radical reshaping entire white
area
S Early Road Map Items
S Excess part of network and core backbone = 2 highest requirement
and list amount of generated incremental benefit
S Data Centre – Latency tolerance (policy & authothenticy) = mobile core
S converging wireless and wire line in universal services IMS platform
S Network Access
S Wireless – more like R&C pooling – really virtualization
S Complex (3G, 4G, wifi, AWS, WCS…) load very fast
S Optimized backbone switches and routing
Amin Vahdat,
Google
Google’s experience with
Software Defined Network
Function Virtualization at
Scale
• DE and Tech Lead for
Networking, Google
• SAIC Professor of Computer
Science and Engineering,
UCSD
• ACM Fellow, Sloan Fellow,
Duke University...
Networking @ Google
Network Corporation Team
CLOUD
S Infinite computer storage on demand
S Fundamentally easier operational model (not there yet – true
promise of the cloud – scale business)
S Much higher of time – 3or4 nines of availability for services
(multiplexing – leverage)
S State-of-the-art infrastructure services – DoS, Load Balancing,
Storage (can’t buy string out) – Offer these services = key to
success
S Programming models unavailable elsewhere – low lactency
programming, massive IOPS
Andromeda Network
Virtualization
Google Cloud Platform
CLOUD
S SDN control of entire hardware/software – QoS, Latency, fault
tolerance (Holistic Approach)
S Virtualize SDN with NFV (Non-standard Network functionality
per packet, can’t predict all the services, API for NFV = big
opportunity)
S Orchestrate & Manage: Network provisioning, High availability,
Balanced virtual infrastructure (Network, Storage & Compute in
right proportion for highest efficiency)
Andromeda Network
Virtualization
Google Cloud Platform
Google Infrastructure
S Google Global CDN (focus on
driving down to milliseconds
between Google services and
end users
S $2.9B Data Centre investments
worldwide
S Managing the energy in the
cooling - efficiency 2-3x better
Google Infrastructure
S Cluster Networking
S - storage building level
S - computation anytime
anywhere
S B4 SDN: Google Software
Defined WAN
Google Infrastructure
S First Google
File System
S Inspired HDFS
S Inspired Big
Data
revolution
S World largest NoSQL
Implementation
S Wide area
consistence
storage
infrastructure
for
transaction
across planet
New
Challenges
S Isolation
S DDoS
S Virtual IP networks
S Network Function
Virtualization
S Mapping external
services into internal
namespace
S Authentication,
authorization, billing
S Maintaining efficiency
while doing all of the
above
From Network Virtualization
Efficiency
Balance Compute, Storage
Capacity Bandwidth, Memory,
Network Bandwidth
How to provisioning your system:
Problem: Different
application has
different balanced
point
SDN solution: SDN
provisioning (Ser up the
balance point of network
compute and storage)v
Efficiency
Need Fundamental
transformation in virtual
networking
How to Spin up 1000 port
virtual network with
isolation, load balancing,
external access, bandwdith
provisionin?
How to deliver highest
availability transparently
to the end customer?
SDN + NFV
S Provision an isolated, high-performance network across
S NIC, soft switch, storage, packet processor, fabric switches, Tor
S Audit correctness – transactional distributed operation (the weakest
link is whatever part you get wrong)
S Provision resources end to end QoS and availability
S Logically centralized network management
S Programmable packet processors for extensible network functionality
S APIs for network application interoperability
SDN ROLE
SDN Opportunities and Challenges
S “Middle boxes like functionality” SDN API for others to build their own
network functionality with customized requirement
NFV
Andromeda Control Stack
Case study 1
S Datapath logical view
Andromeda Network Datapath with intergrated programmable
NFV
S Datapath pipelined, replicated multiple times as VM resources
scale out
S Critical optimization applied e2e (leveraging locality, forwarding
with less rules)
S Goal: near native performance, CPU efficiency
Case study 1
Network Performance
Case study 1
CPU Effiency
Case Study 2
S Rapid provisioning of virtual
networks
S Fast ramp to peak load
S Massive scale
S Low Cost!
Network Virtualization
@ Google
S Delivering high performance shared computing infrastructure
S Enable new programming model
S Open up to customer to use same infrastructure
S Logical centralized SDN control Orchestra across many
different components
S API for extensible NFV
S Goals: native performance of hardware, full isolation,
extensible NFV, high availability, scale out to many VMs.

presentationGAATT

  • 1.
  • 2.
    John Donovan, AT&T How SDN enabledinnovations will impact AT&T’s plans to transform it’s infrastructure
  • 3.
    Openness Transformation & Innovation Flexibility+ Simplicity + Speed New Resources Model in vM Redefining end point Time required to provision services
  • 4.
    SDN + NFV TelecomNetwork Running on vM Simpler Scalable
  • 5.
    AT&T www.at&tmorationsp ace.com Build capabilities in AT&TCloud for real time services Mobile data traffic has increased 50000% over the last 7 years Enhance business efficiency Less CAPEX & OPEX
  • 6.
    Transformation @ AT&T Similarto Cloud Computing (Network Design) Common Infrastructure, Consumption model Open Network principle >100 vendors collaboration Domain 2.0 DODO Budget for new architecture Ericson, Affirm Network - ecp, Metal Switch Core, Tel- ef – control software: open & flexible & work with existing devices Open to new Ideas and maintain competitiveness Doing with the WAN as the world has done to Data Centre cloud centric workloads Elasticity to Scale Provision services in WAN network Open Process Agile development
  • 7.
    Domain 2.0 DODO 2014Domain 1.5 S Control on existing platforms – extend utilizing New Platform in 2015 - NO overlay network – distributed cloud 2015 4600 Data Centre S microseconds of compromise that exist between middle and applications will be overcome by the milliseconds they avoid by highly distributed, very fast fiber network OSS S APIs – Do Policy & Provisioning Software S Orchestrate in highly distributed network
  • 8.
    Why running theirservices on your network? S Platform Extra S move security into cloud S system getter smarter as components get smarter through communication S Privacy & Liability Guidelines S Clarify boundaries – syntactic data build capability without touching the raw S 3 new technologies – control plane role, library, routing system for data object – abstract syntactic data– no compromising customers
  • 9.
    Radical reshaping entirewhite area S Early Road Map Items S Excess part of network and core backbone = 2 highest requirement and list amount of generated incremental benefit S Data Centre – Latency tolerance (policy & authothenticy) = mobile core S converging wireless and wire line in universal services IMS platform S Network Access S Wireless – more like R&C pooling – really virtualization S Complex (3G, 4G, wifi, AWS, WCS…) load very fast S Optimized backbone switches and routing
  • 10.
    Amin Vahdat, Google Google’s experiencewith Software Defined Network Function Virtualization at Scale • DE and Tech Lead for Networking, Google • SAIC Professor of Computer Science and Engineering, UCSD • ACM Fellow, Sloan Fellow, Duke University...
  • 11.
    Networking @ Google NetworkCorporation Team CLOUD S Infinite computer storage on demand S Fundamentally easier operational model (not there yet – true promise of the cloud – scale business) S Much higher of time – 3or4 nines of availability for services (multiplexing – leverage) S State-of-the-art infrastructure services – DoS, Load Balancing, Storage (can’t buy string out) – Offer these services = key to success S Programming models unavailable elsewhere – low lactency programming, massive IOPS
  • 12.
    Andromeda Network Virtualization Google CloudPlatform CLOUD S SDN control of entire hardware/software – QoS, Latency, fault tolerance (Holistic Approach) S Virtualize SDN with NFV (Non-standard Network functionality per packet, can’t predict all the services, API for NFV = big opportunity) S Orchestrate & Manage: Network provisioning, High availability, Balanced virtual infrastructure (Network, Storage & Compute in right proportion for highest efficiency)
  • 13.
  • 14.
    Google Infrastructure S GoogleGlobal CDN (focus on driving down to milliseconds between Google services and end users S $2.9B Data Centre investments worldwide S Managing the energy in the cooling - efficiency 2-3x better
  • 15.
    Google Infrastructure S ClusterNetworking S - storage building level S - computation anytime anywhere S B4 SDN: Google Software Defined WAN
  • 16.
    Google Infrastructure S FirstGoogle File System S Inspired HDFS S Inspired Big Data revolution S World largest NoSQL Implementation S Wide area consistence storage infrastructure for transaction across planet
  • 17.
    New Challenges S Isolation S DDoS SVirtual IP networks S Network Function Virtualization S Mapping external services into internal namespace S Authentication, authorization, billing S Maintaining efficiency while doing all of the above From Network Virtualization
  • 18.
    Efficiency Balance Compute, Storage CapacityBandwidth, Memory, Network Bandwidth How to provisioning your system: Problem: Different application has different balanced point SDN solution: SDN provisioning (Ser up the balance point of network compute and storage)v
  • 19.
    Efficiency Need Fundamental transformation invirtual networking How to Spin up 1000 port virtual network with isolation, load balancing, external access, bandwdith provisionin? How to deliver highest availability transparently to the end customer?
  • 20.
    SDN + NFV SProvision an isolated, high-performance network across S NIC, soft switch, storage, packet processor, fabric switches, Tor S Audit correctness – transactional distributed operation (the weakest link is whatever part you get wrong) S Provision resources end to end QoS and availability S Logically centralized network management S Programmable packet processors for extensible network functionality S APIs for network application interoperability SDN ROLE SDN Opportunities and Challenges S “Middle boxes like functionality” SDN API for others to build their own network functionality with customized requirement NFV
  • 21.
  • 22.
    Case study 1 SDatapath logical view Andromeda Network Datapath with intergrated programmable NFV S Datapath pipelined, replicated multiple times as VM resources scale out S Critical optimization applied e2e (leveraging locality, forwarding with less rules) S Goal: near native performance, CPU efficiency
  • 23.
  • 24.
  • 25.
    Case Study 2 SRapid provisioning of virtual networks S Fast ramp to peak load S Massive scale S Low Cost!
  • 26.
    Network Virtualization @ Google SDelivering high performance shared computing infrastructure S Enable new programming model S Open up to customer to use same infrastructure S Logical centralized SDN control Orchestra across many different components S API for extensible NFV S Goals: native performance of hardware, full isolation, extensible NFV, high availability, scale out to many VMs.