10 - Traffic Engineering Flashcards

1
Q

The process of re-configuring the network in response to changing traffic loads, to achieve some operational goal

A

Traffic engineering

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

network operator might want to reconfigure the network in response to changing traffic loads

A

1) to maintain traffic ratios in a peering relationship
2) to relieve congestion on certain links in the network
3) to balance load more evenly across the available links in the network

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

How should routing adapt to traffic

A

1) Avoid congested links

2) Satisfy application requirements

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Traffic engineering has 3 steps

A

1) Measure
2) Model
4) Control

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Intradomain Traffic Engineering:Tuning link weights

A

1) Routers flood information to learn topology

2) Operator configures link weight

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Ways that operators can set the links

A

1) Inversely proportional to capacity
2) Proportional to propagation delay
3) Network wide optimization

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Intradomain Traffic Engineering: Optimization

A

Input: Graph(R,L)
Output: Set of link weights (WL)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Utilization

A

The amount of traffic on the link divided by the capacity

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

In practice, we have these operational realities to worry about (Intra-domain routing)

A

1) Minimizing the number of changes to the network
2) Resistant to failure
3) Robust to measurement noise

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Intradomain

Interdomain

A

Intradomain: within a domain
Interdomain: between domains

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Examples of Interdomain routing

A

1) Peering between 2 ISPs
2) Peering between a university network & its ISP
3) Peering at an Internet exchange point
4) Routing across multiple datacenters

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Interdomain routing or Interdomain traffic engineering

A

Routing between ASes

 - Alleviating congestion on edge links
 - Using new/upgraded edge links
  - Changing end to end path
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Goals for Interdomain Traffic Engineering

A

1) Predictability
2) Limit influence of neighbors
3) Reduce overhead of routing changes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Equal cost multipath (ECMP)

A

1) Intradomain routing - Equal cost multipath (ECMP)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

How can a source router adjust paths?

A

1) Alternating between forwarding table entries.
By having multiple forwarding table entries for the same destination and splitting traffic flows across the the multiple next hops depending on the hash of th IP packet header.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Data Center Networking

A

Characteristics

1) Multi-tenacy
2) Elastic resources
3) Flexible service management

17
Q

Data center networking challenges

A

1) Traffic load balance
2) Support for virtual machine mugration
3) Power savings
4) Provisioning
5) Security

18
Q

Data center topology

A
Internet
Core
Aggregation
Access
->Servers
19
Q

Scale problem

A

10s of thousands of servers on a flat layer-2 topology where all of the servers have a topology independant MAC hardware address and in default every switch in topology has to store a forwarding table entry for every MAC address (Huge forwarding tables!)

20
Q

Scale problem solution

A

Pseudo Mac -> real MAC

Fabric manager is involved here.

21
Q

VL2 (Valiant Load Balance)

A

Goals

1) Spread traffic
2) Location independence

22
Q

Jellyfish

A

Networking data centers randomly

23
Q

Jellyfish goals

A

Goals:
High throughput -> Big data
Incremental expandability -> easy replacement of servers

Problems:
Structure constraints expansions

24
Q

Where does DC structure constrain expansion

A

Top-level switches

25
Q

Jellyfish’s topology

A

Random regular graph

26
Q

Jellyfish construction (RRG -> random Regular Graph)

A

N(ki-ri) servers

ri ->switches
ki->ports
N -> racks

Random Regular Graph
RRG(N,k,r)

27
Q

Constructing Jellyfish

A

1) Pick a random switch pair

28
Q

Open questions on Jellyfish topology

A
Topology Design
     How close are random graphs to optimal   
     What about heterogeneous switches
System Design
     Cabling
     Routing