Systematic Debugging of Parallel Programs in DIWIDE Based on Collective Breakpoints and Macrosteps1

Kacsuk, Péter; Lovas, Róbert; Kovács, József

doi:10.1007/3-540-48311-x_8

Cited by 6 publications

(11 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One feasible way to prove the cloud platform agnostic feature of complex deployment and maintenance strategies is to leverage on advanced systematic debugging/troubleshooting methods in order to find the timing or architecture dependent failures in the designed deployment description and orchestrator. For this purpose, our research team applied and extended the macrostep-based debugging [4] methodology that has been introduced originally for message-passing parallel programs and developed in the P-GRADE graphical programming environment [5].…”

Section: Solutionmentioning

confidence: 99%

See 1 more Smart Citation

Fundamentals of a Novel Debugging Mechanism for Orchestrated Cloud Infrastructures with Macrosteps and Active Control

et al. 2021

View full text Add to dashboard Cite

In Infrastructure-as-a-Service (IaaS) clouds, the development process of a ready-to-use and reliable infrastructure might be a complex task due to the interconnected and dependent services that are deployed (and operated later on) in a concurrent way on virtual machines. Different timing conditions may change the overall initialisation method, which can lead to abnormal behaviour or failure in the non-deterministic environment. The overall motivation of our research is to improve the reliability of cloud-based infrastructures with minimal user interactions and significantly accelerate the time-consuming debugging process. This paper focuses on the behaviour of cloud-based infrastructures during their deployment phase and introduces the adaption of a replay, and active control enriched debugging technique, called macrostep, in the field of cloud orchestration in order to provide support for developers troubleshooting deployment-related errors. The fundamental macrostep mechanisms, including the generation of collective breakpoint sets as well as the traversal method for such consistent global states, have been combined with the Occopus cloud orchestrator and the Neo4J graph database. The paper describes the novel approach, the design choices as well as the implementation of the experimental debugger tool with a use case for validation purposes by providing some preliminary numerical results.

show abstract

Section: Solutionmentioning

confidence: 99%

“…Early debugging methods in parallel systems relied on the so-called "monitor & replay" approach [4]. In the first monitoring (or recording) phase, the monitoring tool collects as much data about the parallel program as required to reproduce the run in a deterministic way in the second replay phase.…”

Section: Original Concept For Message-passing Programsmentioning

confidence: 99%

Fundamentals of a Novel Debugging Mechanism for Orchestrated Cloud Infrastructures with Macrosteps and Active Control

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In the debugging mode, the execution of the program is under the control of the DIWIDE [11] distributed debugger which provides options to create breakpoints, perform step-by-step execution, animation of the flow of control, etc. Once the program has been successfully debugged, it can be executed in trace mode.…”

Section: Tools In P-gradementioning

confidence: 99%

Using P-GRADE for Monte Carlo Computations in a Distributed Environment

Alexandrov

Thandavan

Kacsuk

2004

Computational Science - ICCS 2004

View full text Add to dashboard Cite

Abstract. Computations involving Monte Carlo methods are, very often, easily and efficiently parallelized. P-GRADE is a parallel application development environment which provides an integrated set of programming tools for development of general message-passing applications to run in heterogeneous computing environments or supercomputers. In this paper, we show how Monte Carlo algorithms for solving Systems of Linear Equations and Matrix Inversion can easily be parallelized using P-GRADE.

show abstract

“…Accordingly, systematic testing and debugging of a metacomputing program require (i) generation of its execution tree (ii) exhaustive traverse of its execution tree. With the help of the extended macrostep-by-macrostep concept both of these issues can be solved and implemented in a very similar way as they have been implemented in DIWIDE [8] [12]. Some minor changes are required by the wildcard send operations as well as the event tracing and replaying.…”

Section: Systematic Debugging In Harnessmentioning

confidence: 99%

“…As we studied the related work (such as DejaVu [I], DJVM [ll], DIWIDE [8] [12] integrated in P-GRADE [lo], Macrostep-by-macrostep concept [7], Totalview, P2D2 [4]) we realized that there is a lack of an integrated graphical systematic debugger for metacomputing applications equipped by visualization techniques. Even the most relevant gridlmetacomputing projects such as Globus [3], Condor [13] and Legion [6] do not give efficient solution to the emerging debugging issues.…”

Section: Conclusion and Related Workmentioning

confidence: 99%

Extension of Macrostep Debugging Methodology Towards Metacomputing Applications

Lovas

Sunderam

2001

Computational Science — ICCS 2001

View full text Add to dashboard Cite

This paper focuses on the non-deterministic behaviour and architecture dependencies of metacomputing applications from point of view of debugging. As a possible solution we applied and also extended the macrostep systematic debugging methodology fir metacomputing applications. Our extended methodology is based on modified collective breakpoints and macrosteps furthermore, we introduce host-translation tables generated automatically for exhaustive testing. The prototype is developed under the Harness metacomputing framework for message box communication based applications. The main implementation issues as well as the architecture of our systematic debugger are also described as the further development of X-IDVS Harness-based metadebugger.

show abstract

Systematic Debugging of Parallel Programs in DIWIDE Based on Collective Breakpoints and Macrosteps1

Cited by 6 publications

References 3 publications

Fundamentals of a Novel Debugging Mechanism for Orchestrated Cloud Infrastructures with Macrosteps and Active Control

Fundamentals of a Novel Debugging Mechanism for Orchestrated Cloud Infrastructures with Macrosteps and Active Control

Using P-GRADE for Monte Carlo Computations in a Distributed Environment

Extension of Macrostep Debugging Methodology Towards Metacomputing Applications

Contact Info

Product

Resources

About