Enabling Ultra Low Latency Applications Over Ethernet

1
Enabling Ultra Low Latency Applications Over
Ethernet

• Grant Kirkwood
• Chief Technology Officer
• Mzima Networks

2
Services are fast becoming packet-based
Residential
Digital broadcast IPTV, VoIP Internet video gaming
Circuit-switched voice and broadcast video
Enterprise
Ethernet Services Virtual circuits IP PBX VoIP
Dedicated connections T1/T3, FR, ATM Digital PBX
Wireless
Data optimized Video enabled 3G/4G/WiMax
Voice optimized 2G network
3
Services are stream-based, not packet-based
Digital broadcast IPTV, VoIP Internet video gaming
Ethernet Services Virtual circuits IP PBX VoIP
Data optimized Video enabled 3G/4G/WiMax
4
Services are stream-based, not packet-based
Voice, video and other services streams of
data Internet Protocol (IP) relies on small
packets, not streams of data
Digital broadcast IPTV, VoIP Internet video gaming
Ethernet Services Virtual circuits IP PBX VoIP
Data optimized Video enabled 3G/4G/WiMax
5

IP Networks designed to carrybest effort
traffic Email, web browsing, instant
messaging.Applications are not
time-sensitive. Networks designed to
carry packet-based data are now being askedto
carry streaming data
6

Perceived value has shifted from service provider
to application provider
Enterprise
Consumer

• Penetration of Ethernet, fiber, collaboration,
  and interactive video
• Increased workforce mobility requiring seamless
  access
• Maturing virtualization, cloud-based applications
  telepresence

• Maturity of multiservice offerings
• Increased mobile data penetration and improved
  ease of use
• Growing adoption of online video and Web 2.0
  applications

7

New applications created daily Applications
become more and more sensitive to network
conditions QoS policies are being created to
support these technologies An increasing number
of QoS policies are creating challenges for
service providers
8

Service provider realities
9


How do we deliver real-time services with
carrier-grade QoS?
10
QoS Model
11
Ethernet Performance Metrics

• Packet Loss
• Latency
• Jitter

12
Ethernet Performance Metrics
Jitter
Packet Loss
Latency
13
Ethernet Performance Metrics
Jitter
Packet Loss
Latency
14
Ethernet Performance Metrics
Different applications are sensitive to
different performance metrics
15
Class/Quality of Service
QoS/CoS
16
QoS Model
17
QoS Model
Methods to separatetraffic into different
bucketsor QoS policies
18
QoS Model
FIFO is not intelligent orderingOrder by
priority
19
QoS Model
Non-stream services
Drop packets for applications not sensitive to
packet loss
20
QoS Model
Size throughput to availability Limit/buffer
traffic that wont be impacted by policing
21
QoS Model
22
QoS Model
23
Policing vs Classifying

• Bandwidth policing is dumb
• Metering drops packets without discretion
• Congestion causes buffering (on routers)
• Buffering causes latency
• Each application sensitive to specific metrics
• Key is accurate classification

24
End to end QoS

Classification/Policing
Schedule
Transmit
VoIP
Gold
VoIP
Gold
Silver
VoIP
Gold
Silver
Silver
Best Effort
25
End to end QoS

• End-to-end QoS requires a technology that exists
  end-to-end
• Difficult to achieve in multi-vendor,
  multi-carrier or multi-technology networks
• New technologies are being developed to address
  this limitation

26
Mzima Case Study

• Problem provide true Carrier Ethernet
• Differentiate product from other carriers
• Provide End-to-End QoS
• Carry carrier-grade Services
• Application-level granularity of QoS profiles
• Meet requirements of true Carrier Ethernet

27
Mzima Case Study

28
Mzima Case Study

• VLAN bridging or tunneling
• Layer 2 MPLS
• Layer 3 MPLS
• PBB-TE

29
PBB-TE

• Provider Backbone Bridge Traffic Engineering
• End-to-End QoS
• Stringent performance metrics
• First Carrier Ethernet protocol not integrated
  into a layer 3 protocol (MPLS)

30

PBB-TE Tunnel Performance Management
Y.1731 Performance Management
PRIMARY

PBB-TE BCBs
PBB-TE BEB
PBB-TE BEB
BACKUP

• PBB-TE with Y.1731 Performance Management
• Performance Management between Tunnel Endpoints
• Provides Service Independent Tunnel Monitoring
• Enhanced Scalability as 1,000s of services may
  traverse the tunnel without the need to monitor
  every service
• Leverages 802.1ag frames for reduced overhead
• Multiple packets sent at 100ms interval to
  perform the test
• Frame Delay / Frame Delay Variation / Loss
  Measurement
• 2-way Delay Roundtrip Measurement
• 1-way Delay Measurement (requires common time
  base)
• Single Ended Frame-Loss (MEP to MEP)

31

Management Plane Y.1731

• Round trip delay/jitter and single ended frame
  loss (MEP to MEP)
• Non-Service Affecting
• Utilizes IEEE 802.1ag format frames for test
  packets
• Unicast messages to a specific MEP
• Multiple packets sent at 100ms interval to
  perform the test
• Delay, Jitter, and Frame Loss measurements
• Test results remain until the next test is run
  or until reboot of switch
• MIPs do not participate in delay/jitter/frame
  loss measurements

MEP 12
MIP
802.1ag CCMs
MEP 10
MEP 11
MIP
MIP
32
Thank You