Presentation is loading. Please wait.

Presentation is loading. Please wait.

Fault Detection, Isolation, and Diagnosis In Multihop Wireless Networks Lili Qiu, Paramvir Bahl, Ananth Rao, and Lidong Zhou Microsoft Research Presented.

Similar presentations


Presentation on theme: "Fault Detection, Isolation, and Diagnosis In Multihop Wireless Networks Lili Qiu, Paramvir Bahl, Ananth Rao, and Lidong Zhou Microsoft Research Presented."— Presentation transcript:

1 Fault Detection, Isolation, and Diagnosis In Multihop Wireless Networks Lili Qiu, Paramvir Bahl, Ananth Rao, and Lidong Zhou Microsoft Research Presented by -Maitreya Natu

2 Network Management Faulty network … Root cause Faults directory Corrective measure Healthy network

3 Tasks involved in Network Management Continuously monitoring the functioning Collecting information about the nodes and the links Removing inconsistencies and noise from the reported information Analyzing the information Taking appropriate actions to improve network reliability and performance

4 Challenges in wireless networks Dynamic and unpredictable topology  link errors due to fluctuating environment conditions  Node mobility Limited capacity  Scarcity of resources Link attacks

5 Proposed framework Reproduce inside a simulator, the real- world events that took place Use online trace driven simulation to detect faults and analyze the root causes

6 Network Management … Healthy network Types of faults Network model Faults directory Creating a network model

7 Network Management Faulty network … Types of faults Network model Detected faults Fault diagnosis Faults directory

8 Network Management … Types of faults Network model what-if analysis Detected faults Faults directory Corrective measures

9 Key issues How to Accurately reproduce what happened in the network inside a simulator How to build fault diagnosis on top of a simulator to perform root cause analysis

10 Accurate modeling Use real traces from the diagnosed network  Removes dependency on generic theoretical models  Captures nuances of the hardware, software and environment of the particular network Collect good quality data  By developing a technique to effectively rule out erroneous data

11 Fault diagnosis Performance data emitted by trace driven simulation is used as baseline Any significant deviation indicates a potential fault Simulator selectively injects a set of suspected faults and searches a set that most produces the expected performance An efficient algorithm is designed to determine root causes

12 System Overview simulator Topology changes Traffic Simulator Interference Injection Link RSS Link Load Routing update Faults Directory +/- Expected loss rate Throughput noise Loss rate Throughput noise Error Link/Node failure 1. Receive Cleaned Data2. Drive Simulation 3. Compute Expected Performance 4. Compare Expected & Average Performance 5. Discrepancy Found 6. Search for set of faults that result in best explanation 7. Report the cause of failure

13 Why Simulation Based Diagnosis? Much better insights into the network behavior than any heuristic or theoretical technique Highly customizable and applies to a large class of networks Ability to perform what-if analysis  Helps to foresee the consequences of a corrective action Recent advances in simulators have made possible their use for real-time analysis

14 Accurate modeling … Healthy network Types of faults Network model Faults directory

15 Current network models Bayesian networks to map symptom-fault dependencies Context Free Grammars Correlation Matrix

16 Can on-line simulations be used as core tool?

17 Building confidence in simulator accuracy Problem  Hard to accurately model the physical layer and the RF propagation  Traffic demands on the router are hard to predict

18 Building confidence in simulator accuracy Problem  Hard to accurately model the physical layer and the RF propagation  Traffic demands on the router are hard to predict Solution  “after the fact” simulation  Agents periodically report information about the link conditions and traffic patterns to the link simulators

19 Simulations when the RF condition of the link is good Modeling the overheads of the protocol stack such as parity bits, MAC-layer back-off, IEEE 802.11 inter-frame spacing and ACK, and headers. Modeling the contention from flows within the interference and communication ranges.

20 Simulations with varying received signal strength Throughput matches closely with the simulator’s estimate, when signal quality is good Simulator estimate deviates from real, when signal strength is poor

21 Why simulation results deviate in case of poor signal strength? Lack of accurate packet loss as a function of packet size, RSS and ambient noise.  Depends on signal processing hardware and the RF antenna within the wireless cards Lack of accurate auto-rate control  Adjustment of sending rate done by WLAN cards based on the transmission conditions

22 How to model auto-rate control done by WLAN cards? Use Trace driven simulation When auto-rate is in use  Collect the rate at which the wireless card is operating and provide the reported rate to the simulator Otherwise  Data rate is known to the simulator

23 How to model accurate packet loss as a function of packet-size, RSS and ambient noise? Use offline analysis Calibrate the wireless cards and create a database associating environmental factors with expected performance  E.g., mapping from signal strength and noise to loss rate

24 Experiment to model the loss rates due to poor signal strength Collect another set of traces  Slowly send out packets  Place packet sniffers near both the sender and the receiver, and derive loss rate from the packet level trace Seed the wireless link in the simulator with a Bernoulli loss rate that matches loss rate with the real traces

25 Estimated and measured throughput when compensating for the loss rate due to poor signal strength Even though the match is not perfect, its not expected to be a problem, because many routing protocols try to avoid the use of poor quality links Poor quality links are used only when certain parts of mesh network have poor connectivity to the rest of the network In a well-engineered network, not many nodes depend on such bad link for routing Loss rate and the measured throughput do not monotonically decrease with the signal strength due to the effect of auto-rate

26 Stability of channel conditions How rapidly do channel conditions change and how often a trace should be collected?

27 Temporal fluctuation in RSS Fluctuation magnitude is not significant Relative quality of signals across different number of walls remain stable

28 Stability of channel conditions How rapidly do channel conditions change and how often a trace should be collected?  When the environment is generally static, nodes may report only the average and standard deviation of the RSS to the manager every few minutes

29 Dealing with imperfect data By neighborhood monitoring  Each node reports performance and traffic statistics for its incoming and outgoing links  And for other links in its communication range Possible when node is in promiscuous mode Thus multiple reports are sent for each link Redundant reports can be used to detect inconsistency Find the minimum set of nodes that can explain the inconsistency in the reports

30 Summary How to accurately model the real behavior?  Solution: Use trace-based simulation Problem: Simulation results are good for strong signals but deviate for bad RF conditions  Need to model the autorate control Use trace-driven data  Need to model the loss rate due to poor signal strength Use offline analysis How often a trace should be collected?  Very little data (average and standard deviation of RSS), at fairly low time granularity, as channels are relatively stable How to deal with imperfect data  By neighborhood monitoring

31 Fault diagnosis Faulty network … Types of faults Network model Detected faults Faults directory

32 Current fault diagnosis approaches AI techniques  Rule based systems  Neural networks Model traversing techniques  Dependency graphs  Causality graphs  Bayesian networks

33 Fault Isolation and Diagnosis Establish the expected performance in the simulation Find difference between expected and observed performance Search over the fault space to detect which set of faults can re-produce performance similar to what has been observed

34 Collecting data from traces Trace data collection  Network topology Each node reports its neighbor and routing tables  Traffic statistics Each node maintains counters of traffic sent and received from immediate neighbors  Physical medium Each node reports signal strength of wireless links to neighbors  Network performance Includes both the link and end-to-end performance, which can be measured through loss rate, delay, throughputs Focus is on link level performance

35 Simulating the network performance Traffic load simulation  Link based traffic simulation  Adjust application sending rate to match the observed link-level traffic counts Route simulation  Use actual routes taken by packets as input to the simulator Wireless signal  Use real measurement of signal strength Fault injection  Random packet dropping  External noise sources  MAC misbehavior

36 Fault diagnosis algorithm General approach Simulator Expected performance Network settings Simulator Observed performance Network settings Faults set How to find ?

37 How to search the faults efficiently? Different types of faults often change one or few metrics  E.g., random dropping only affects link loss rate Thus use metrics in which observed and expected performance is significantly different, to guide the search

38 Scenario where faults do not have strong interactions Consider large deviation from expected performance as anomaly Use decision tree to determine the type of fault Fault type determines the metric to quantify performance difference Locate faults by finding the set of nodes and links with large difference between expected and observed performance

39 Scenario where faults have strong interactions Get the initial diagnosis set from the decision tree algorithm Iteratively refine the fault set  Adjust the magnitudes of faults in the fault set Translate difference in performance into change in faults’ magnitude It maps the impact of a fault into its magnitude Remove fault whose magnitude is too small  Add new faults that can explain large differences between the expected and observed performances Iterate till the change in fault set is negligible

40 Example scenario 1 2 3 45

41 1 2 3 45 Observed performance Increased loss rate at 1-4 and 1-2 No increase in the sending rate of 1-4, 1-2 No increase in noise experienced by neighbors Inference Increased Sending Rate Increased Noise Increased Loss Too low CW Noise Packet DropNormal Y N Y Y N N

42 Example scenario 1 2 3 45 Observed performance Increased loss rate at 1-4 and 1-2 No increase in the sending rate of 1-4, 1-2 No increase in noise experienced by neighbors Inference Increased Sending Rate Increased Noise Increased Loss Too low CW Noise Packet DropNormal Y N Y Y N N Packet dropping at node 1

43 Accuracy of fault diagnosis Correctness of the model  Complete information  Consistent information  Timely information Correctness of the reported symptoms  Right size of the threshold to report a symptom  Difference in the behavior of faults  Timely reporting of symptoms

44 System implementation Windows XP Agents run on every wireless node and reports information collected on demand Managers collect and analyze information Collected information is cast into performance counters supported by Windows Manager is connected to a backend simulator. Collected information is converted to script to drive the simulation Testbed:  Multihop wireless testbed built using IEEE 802.11a cards  Commercially available network sniffer called Airopeek is used for data collection  Native 802.11 NICs provide rich set of networking information

45 Evaluation: Data collection overhead Management traffic overheadPerformance of FTP flow with and without data collection No data cleaning: Each link is reported only once With data cleaning: Each link is reported by all observers for consistency check Overhead < 800 bits/s/node Data collection traffic has little effect

46 Data cleaning effectiveness Higher accuracy with denser networks Higher accuracy with client-server traffic Coverage greater than 80% in all cases Higher accuracy with grid topology Higher coverage when using history

47 Evaluation: Fault diagnosis Detecting random dropping Detecting external noise Symptom: Significant difference in loss rates in links Less than 20% of fault links are left undetected No-effect faults are faulty links sending less that threshold (250) packets of data Symptom: Significant difference in noise level in nodes Noise sources are correctly identified with at most one or two false positives Inference error in magnitudes of noises is within 4%

48 Evaluation: Fault diagnosis Detecting MAC misbehavior Detecting combinations of all Symptom: Significant discrepancy in throughput on links Coverage is mostly around 80% or higher False positives within 2

49 what-if analysis … Types of faults Network model Detected faults Faults directory Corrective measures

50 What-if analysis DiagnosisTopology Corrective measures

51 Limitations Limited by accuracy of the simulator Time to detect the faults is acceptable for detecting long term faults but not transient faults Choices of traces to drive the simulation has important implications Focus has only been on faults resulting in different behavior

52 Conclusion Used trace data for modeling the network Data collection techniques are presented to collect network information and detect a deviation from the expected performance Fault diagnosis algorithm is proposed to detect the root causes of failure A scheme for what-if analysis is proposed to evaluate alternative network configuration for efficient network operation

53 Future work Validation on a large test-bed Performance analysis in presence of mobility Detecting malicious attacks Diagnosis in presence of incomplete network information More deeply investigating the potential of what-if analysis

54 References L. Qiu, P. Bahl, A. Rao, L. Zhou, Fault Detection, Isolation, and Diagnosis in Multihop Wireless Networks, Microsoft Technical Report, Microsoft Researh-TR-2004-11, Dec. 2003 M. Steinder, A. Sethi, A survey of fault localization techniques in computer networks, Technical Report 2001, CIS Dept., Univ of Delaware, Feb 2001 M. Steinder, Probabilistic inference for diagnosing service failures in communication systems, PhD thesis, Univ. of Delaware, 2003

55 Questions What is proposed solution to model the throughput when the signal strength is poor? In Table 2, the simulated throughput monotonically decreases with the loss rate while the measured throughput does not. Why? What could be the causes of generation of false positives in the fault diagnosis results? When can the false positive ratio increase? http://www.cis.udel.edu/~natu/861/861.html


Download ppt "Fault Detection, Isolation, and Diagnosis In Multihop Wireless Networks Lili Qiu, Paramvir Bahl, Ananth Rao, and Lidong Zhou Microsoft Research Presented."

Similar presentations


Ads by Google