An Efficient Fault-Tolerant Approach for Mobile IP in Wireless Systems

1
An Efficient Fault-Tolerant
Approach for Mobile IP in Wireless
Systems
CS 6204 Paper Presentation

• Jenn-Wei Lin and Joseph Arul
• Paper Presented by Vidhya Dass

10/31/2006
1
2
Agenda

• Introduction
• The Proposed Approach
• Fault tolerance of FA
• Fault tolerance of HA
• Evaluation
• Analytical comparison Simulation
• Conclusion

2
3
Introduction

• Mobile IP Support wireless users with
  continuous network connections while changing
  locations
• Functionality of Mobile IP in wireless system
  provided by
• Mobility agents in architecture of wireless
  systems(HA and FA)
• Drawbacks No fault tolerance for MA failure
• Approach Resource sharing to redirect workloads
  of faulty FA(HA) to other failure free FA(HA)

3
4
Basic mobile IP in wireless system
HA
CH
IP Network
Home network
FA
Wireless Data serving area
RAN
RAN
RAN
MN
4
5

• Related work
• 1.MA statically equipped with one or more
  redundant MAs to work in standby or load sharing
  mode
• MA fails, one backup member selected as primary
  mobility agent
• ARP Used to map IP address of faulty MA onto
  network link layer address of selected backup
  member
• Disadvantages Long registration delay since MN
  registers with all MA
• 2.Checkpointing and logging technique Store
  mobility bindings in stable storage

5
6

• GOAL Provide Fault tolerance capability in
  wireless system with mobile IP functionality
• Fault tolerance in telecom system is five
  nines(99.999) reliability requirement for
  network design but Hardware failures follow
  bathtub curve. Provide fault tolerance for MA
  failures.

6
7
Wireless data network model
System model
7
8

• OA M functions
• Configuration management Configures equipment
  with suitable resource parameters
• Fault management Detecting and reporting
  failures in equipment
• Performance management Measures resource
  utilization, loading status, concerned values in
  equipment
• Security management Monitors access rights to
  equipment
• Assumptions
• Failures only occur in MA Detect failure by not
  receiving agent advertisement messages within a
  time period
• Fail - Stop approach Faulty MA not send agent
  advertisement messages

8
9
The Proposed Approach

• Dynamically select multiple failure free MA as
  backup set for faulty MA when a failure is
  detected
• Workloads of faulty MA redirected to failure free
  MA in backup set
• Faulty FA One or more failure free FA
  dynamically selected (backup set), system
  initiated handoff issued to virtually move all MN
  to service area of backup FA (Continuous data
  executable property)
• Faulty HA One or more failure free HA
  dynamically selected (backup), intercept packets
  moving toward faulty HA and send them to
  corresponding MN

9
10
Fault tolerance of Foreign Agent

• (FA_failure-affected MNs) MN in serving area
  and arriving MN, cannot execute wireless data
  sessions
• System initiated handoff to dynamically select
  multiple failure free FAs, which are backup set
  of faulty FA
• FA_failure-affected MNs virtually moved to
  serving areas of failure free FA
• Failure free FAs adds visitor entries for
  FA_failure-affected MN, that have moved into it
• Informs MNs corresponding HAs of new serving FAs
  and CoA for mobility bindings
• Workloads of faulty FA redirected to other
  failure free FA

10
11

• Achieve virtual movement of FA_failure-affected
  MNs Modify RAN-FA interconnection network
  which is determined by RANs internal FA-serving
  record
• Initially
• FA-serving record of RAN Identifier of fixed FA
• FA Failure detected FA-serving record of
  failure-affected RAN(initially served by faulty
  FA) reset with identifiers of backup members
• FA_failure-affected MNs served by backup members
  but their location is same(still located in
  respective radio coverage area)

11
12
RAN-FA Remapping for fault tolerance
12
13
Implementation of Foreign Agent

• FA Failure detected Failure event sent to OAM
  fault management
• fault management initiates proposed fault
  tolerant approach for FA
• 1.Interacts with performance management to
  acquire loading status of failure free FAs, finds
  number of FA_failure-affected MNs
• 2.Select multiple failure free FA as backup
  members of faulty FA
• 3.Configuration management informed to configure
  backup members of faulty FA by resetting
  appropriate parameters to some equipment in core
  network update mobility bindings of MNs

13
14
Fault tolerance of Home Agent

• HA functions Mobility binding maintenance,
  packet interception packet tunneling
• HA_failure-affected MN MNs managed by faulty
  HA not able to receive packets from CHs
• Select one or more failure free HA dynamically as
  backup members
• Mobility bindings of faulty HA restored by
  searching all FAs visitor lists
• Distribute bindings to backup members Up-to-date
  location of all MNs known from FAs visitor list
  entry(MNs data link layer address, IP address
  and home agent address)

14
15

• Packet interception of faulty HA restored on
  backup using tunneling. Routers collocated with
  HA on same network segment, dont forward packet
  to faulty HA but tunnel packets to backup HA
  which again tunnels it to located FAs(packet from
  CH to HA_failure-affected MN sent by twice
  tunneling)
• Packet tunneling function already present in
  failure free HA

15
16
16
Packet route to HA_failure-affected MN
17
Implementation of Home Agent

• Select multiple failure free HAs with low
  traffic(from OAM) as backup set and one among
  them, as HAs backup manager
• Mobility binding restoration mobility-
  reconstruction message sent to each FA and
  responses divided by HA backup manager based on
  MNs IP address. Assigns groups to HA backup
  members
• Collocated routers remove routing entries of
  faulty HA and add routing entries of backup
  members, with its interface set to virtual
  interface pointing to software program to perform
  packet tunneling

17
18
Changing the packet interceptor
Mobility binding reconstruction
18
Redirecting the packet interception
19
Failure Recovery

• FA recovery procedure
• Recovered FA determines failure affected RANs
  (configuration management of OAM)
• Failure affected RANs reset FA-serving records to
  identifier of recovered FA
• Recovered FA creates visitor entries for
  FA-Failure-affected MNs HAs of these MNs
  updated with mobility bindings

19
20

• HA recovery procedure
• Modify routing tables of collocated routers of
  recovered HA
  - packet interceptors of
  HA_failure-affected MN changed
• Mobility bindings of recovered HA reconstructed
  - search all FAs visitor lists

20
21
Evaluation

• Workload redirection causes
• Performance degradation of failure free MA
• Control message overhead
• Traffic behavior of FA, HA modeled as M/G/c/c
  queuing model
• Assumption Data request sent to FA response
  packet intercepted by HA follow Poisson process
• Service time of data request and processing time
  of packet tunneling not follow any specific
  distribution

21
22
22
23
Performance degradation of MA

• Performance degradation of failure free FA due to
  resources being contended by MNs virtually moved
  and original MNs served
• Represented as increasing blocking probability
  PFA_blocking - new data request possibly blocked
  at failure free FA in comparison to prefailure
• Erlangs loss formula from the M/G/c/c queuing
  model

Blocking probability of data request to a failure
free FA
Pre-failure
23
24

• New blocking probability of a data request to a
  failure free FA when FFA FAs fail
• Increasing blocking probability

Blocking probability due to original and
redirected workload
Post-failure
Pre-failure
Post-failure
24
25

• Performance degradation of a failure free HA(HAk)
  - Increasing blocking probability that causes an
  intercepted packet to be blocked at failure free
  HA in comparison with prefailure

Pre-failure
Post-failure
25
26
Control message overhead

• Control messages issued from OAM for assisting
  fault tolerance of MA
• FA_Loading
• RAN_Mapping
• Binding_Update
• HA_Loading
• Interceptor_Change
• Binding_Restoration

26
27

• Cost of FA_Loading TFA_Loading TFA_Response
• Cost of RAN_Mapping TRAN_Mapping

Transmission time from performance management to
fault management
Transmission time from fault management to
performance management
Transmission time from configuration management
to RAN(single memory access)
27
28

• Cost of Binding_Update

Simultaneous transmission of mobility binding
update command from fault management
Total time required for failure free FAs to send
mobility binding updates about all
FA_failure-affected MNs
Fraction of time due to serial, simultaneous
transmissions from FAs to HAs
Total number of FA_failure-affected MN
Average transmission time of registration from FA
to HA
28
29

• Cost of HA_Loading THA_Loading THA_Response
• Cost of Interceptor_Change TInterceptor_Change

Transmission time from fault management to
performance management
Transmission time from performance management to
fault management
Transmission time from configuration management
to collocated router of faulty HA(only memory
access of routing table)
29
30

• Cost of Binding_Restoration

Transmission time of mobility binding restoration
to each HA from fault management
Total time required for restoring lost mobility
binding table of faulty HA
Fraction since all FAs perform serial
simultaneous transmission to HA manager
Average time of sending qualified visitor entry
from FA to HA manager
Total number of HA_failure-affected MN
30
31
are negligible due to high speed physical
interface of
OAM network. Size of messages is negligible and
so is cost.
where
Probability of n in processing data
requests/response packets in faulty MA
Assumption Each MN is not allowed to
simultaneously issue more than one data session .
So PFA_n (P HA_n) represented as probability of
n FA_failure-affected MNs(HA_failure-affected
MNs) in faulty FA(HA)
31
32
32
33
Analytical comparison Simulation

• Previous Approaches
• Primary MA has redundant MA in network segment
• Co-working mode of primary and redundant MA
  Standby and Load sharing (Different performance
  degradation)
• Standby modePre-failureBlocking probability of
  selected redundancy 0

(No workload)
Post-failure
performance degradation of
selected redundancy
Pre-failure
0
33
34

• Load Sharing mode
• Pre-failureLoad distributed among primary and
  all redundancies of that primary
• Post-failureIf selected as primary then
  redundant MA has to handle twice the original load

Post-failure
Pre-failure
Where (1RAgent) is primary MA Redundant MAs in
network segment
Arrival rate of data to MA based on load sharing
mode
34
35
35
Tolerate N-1 failures in N MA system
Improvement
least
36

• Traffic intensity of MA Expected number of
  arrivals per mean service time at a MA
• Performance degradation of MA Increasing
  blocking probability of failure free MA

36
37

• Proposed Approach Workload of faulty MA evenly
  redirected to all failure free MA
• Ratio of redirecting workloads to failure free MA
• Previous Approach Load sharing mode Number of
  redundancies in network segment affects
  increasing blocking probability

37
38
38
Previous Approach One redundant MA take over one
faulty MA. Independent of Fagent Increasing
blocking probability ? as ?Agent/?Agent ?
CAgent50
Current Approach Increasing blocking prob ? as
FAgent ? Increasing blocking probability not
always ? as ?Agent/?Agent ?
39
CAgent50
Total number of resource units in FA and HA is 50
Average number of in processing data requests in
an FA(HA) cannot be greater than 50. Maximum
number of NFA_MN(N HA_MN) is 50 Conclusion
Overhead of mobility binding update is restricted
by the total resources in FA(HA)
Equal
Failure recovery overhead depends on this graph
NFA_MN NHA_MN under one faulty FA (HA)
39
40

• Simulated using NS-2
• Same workloads to an FA and HA then PFA_Blocking
  and PHA_Blocking should be equal but simulation
  doesnt agree???

Below 10
Stopped with 8 faulty MAs???
Why didnt they consider all the cases for Number
of faulty MA???
Difference rate is varying randomly???
Only considering MN in data session for
FA_failure-affected MN not justified
40
41
Conclusion

• Utilizes available resources in other failure
  free MA to dynamically generate backup set for
  each faulty MA
• Advantages
• No Hardware support required
• No failure free overhead
• Distribute fault tolerant overhead to avoid
  significant performance degradation on single
  failure free MA
• Good when is 200 and FAgent is small

41
42
Thank you
Questions