Precision, Recall, and Sensitivity of Monitoring Partially Synchronous Distributed Systems

@inproceedings{Yingchareonthawornchai2016PrecisionRA,
  title={Precision, Recall, and Sensitivity of Monitoring Partially Synchronous Distributed Systems},
  author={Sorrachai Yingchareonthawornchai and Duong N. Nguyen and Vidhya Tekken Valapil and Sandeep S. Kulkarni and Murat Demirbas},
  booktitle={RV},
  year={2016}
}
Runtime verification focuses on analyzing the execution of a given program by a monitor to determine if it is likely to violate its specifications. There is often an impedance mismatch between the assumptions/model of the monitor and that of the underlying program. This constitutes problems especially for distributed systems, where the concept of current time and state are inherently uncertain. A monitor designed with asynchronous system model assumptions may cause false-positives for a program… 
Efficient Two-Layered Monitor for Partially Synchronous Distributed Systems
TLDR
This work presents a new, efficient two-layered monitoring approach that overcomes both the time and space limitations of earlier monitors and shows that the combination of these two layers reduces the cost of monitoring by 85-95%.
Crash-Resilient Decentralized Synchronous Runtime Verification
TLDR
This paper proposes an automata-based synchronous monitoring algorithm that copes with crash monitor failures and emits a symbolic verdict that efficiently encodes their partial views.
Efficient Algorithms for Predicate Detection using Hybrid Logical Clocks
TLDR
This work focuses on using hybrid logical clocks (HLCs) to perform wait-free and efficient predicate detection and developing efficient algorithms for detecting weak conjunctive predicates (WCPs) with the help of HLC and then extending them to detect arithmetic predicates such as those necessary for expressing resource usage, network density, and so on.
Monitoring Partially Synchronous Distributed Systems Using SMT Solvers
TLDR
The results show that the framework can be used for real-life applications, and because the framework uses SMT solvers, the range of appropriate applications will increase as these solvers become more efficient over time.
Using Weaker Consistency Models with Monitoring and Recovery for Improving Performance of Key-Value Stores
TLDR
This work proposes a detect-rollback based approach that makes it possible to provide efficient recovery from such faults with a minimal amount of work wasted due to rollbacks, and shows that using eventual consistency with monitoring can provide a 50 – 80% increase in throughput when compared with sequential consistency.
Using weaker consistency models with monitoring and recovery for improving performance of key-value stores
TLDR
Overall, for applications considered in this paper, it is found that even with rollback, eventual consistency provides better performance than using sequential consistency.
Optimistic Execution in Key-Value Store
TLDR
This work proposes a detect-rollback based approach to the problem of what a designer should do if she has an algorithm that works correctly with sequential consistency but is faced with an underlying key-value store that provides a weaker consistency.
Technical Report: Optimistic Execution in Key-Value Store
TLDR
This work focuses on the problem of what a designer should do if he/she has an algorithm that works correctly with sequential consistency but is faced with an underlying key-value store that provides a weaker consistency, and proposes a detect-rollback based approach.
Biased Clocks: A Novel Approach to Improve the Ability To Perform Predicate Detection with O(1) Clocks
TLDR
It is shown that biased clocks improve the ability to detect predicates by 100–200 times when compared to standard hybrid logical clocks.
Retroscope: Retrospective Monitoring of Distributed Systems
TLDR
The Retroscope search algorithm is embarrassingly-parallel and can employ many worker processes (each processing up to 150,000 consistent snapshots per second) to handle a single query.
...
1
2
...

References

SHOWING 1-10 OF 26 REFERENCES
Consistent detection of global predicates
TLDR
This paper presents three algorithms for detecting global predicates in a well-defined way by interpreting predicates with respect to the communication that has occured in the system.
Decentralized Runtime Verification of LTL Specifications in Distributed Systems
TLDR
This paper proposes the first sound and complete method for runtime verification of asynchronous distributed programs for the 3-valued semantics of LTL specifications defined over the global state of the program.
Runtime Verification of Traces under Recording Uncertainty
TLDR
An on-line algorithm for the runtime checking of temporal properties, expressed as past-time Linear Temporal Logic (LTL) over the traces of observations recorded by a "black box"-like device based on a three-valued semantics defined in this paper.
Recognizing patterns in streams with imprecise timestamps
TLDR
This work proposes a temporal model that assigns a time interval to each event to represent all of its possible occurrence times and proposes the formal semantics of such pattern evaluation, two evaluation frameworks, and algorithms and optimizations in these frameworks.
Distributed algorithms for detecting conjunctive predicates
  • V. Garg, C. Chase
  • Computer Science
    Proceedings of 15th International Conference on Distributed Computing Systems
  • 1995
TLDR
This paper introduces algorithms which distribute the computation and space requirements of the detection procedure, and shows that the worst-case time complexity can not be improved beyond O(mn) with any on-line detection algorithm.
Time, clocks, and the ordering of events in a distributed system
TLDR
A distributed algorithm is given for synchronizing a system of logical clocks which can be used to totally order the events, and a bound is derived on how far out of synchrony the clocks can become.
Bounded Version Vectors
TLDR
Bounded version vectors is presented, the first bounded solution to data causality tracking between replicas subject to local updates and pairwise symmetrical synchronization, and its role with respect to vector clocks is clarified.
On Real-Time Monitoring with Imprecise Timestamps
TLDR
This work identifies classes of formulas for which existing monitors can be leveraged to correctly reason about observed system traces in a metric temporal logic.
Timestamps in Message-Passing Systems That Preserve the Partial Ordering
TLDR
This paper presents algorithms for timestamping events in both synchronous and asynchronous n1essage-passing programs that allow for access to the partial ordering in a parallel system.
Logical Physical Clocks
TLDR
This work proposes a hybrid logical clock, HLC, that combines the best of logical clocks and physical clocks, and shows that HLC has many benefits for wait-free transaction ordering and performing snapshot reads in multiversion globally distributed databases.
...
1
2
3
...