Performability Modeling of Exceptions-Aware Systems in Multiformalism Tools

Barbierato, Enrico; Gribaudo, Marco; Iacono, Mauro; Marrone, S.

doi:10.1007/978-3-642-21713-5_19

Cited by 12 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The authors deal with the problem of complexity in [13][14][15] and [16] by means of multiformalism techniques in [17,18] and [19] by resorting to mean field analysis-based techniques suitable for systems that are highly modular and regular and have a high number of elements: in both cases, a model hierarchy-based approach has been used with different aims and purposes. In the first group of papers, the hierarchy is used to experiment with component models to obtain the simplest suitable form that will potentially empower the scale up of composed models, by also exploiting the best modeling formalism for each of them, that can guide in the simplification process; in the second group, in which the problem of scaling up is basically not a limit, the hierarchy is used to identify and derive the parameters of basic building blocks.…”

Section: Motivationmentioning

confidence: 99%

Modeling and analysis of performances for concurrent multithread applications on multicore and graphics processing unit systems

Cerotti

Gribaudo

Iacono

et al. 2015

Concurrency and Computation

View full text Add to dashboard Cite

The capabilities of multicore processors lead them to be widely adopted in systems at any scale, since their are able to provide more computing power at a lower consumption and dissipation cost. System designers are challenged to a deeper understanding of multicore functioning in order to fully exploit them while keeping the optimal balance between cores utilization and optimal throughput, response time and energy usage.Besides the advancement of general purpose CPUs, the same technological evolution leads to the rise of GPUs, dramatic evolution of graphical coprocessors, that are now affordable, efficient, dedicated computing units, capable of parallel computing and equipped with facilities that make them suited for supporting the main CPU of a system in running ordinary applications. The availability of commercial off-the-shelf (COTS) multicore computers, eventually equipped with one or more GPUs, makes them the basic building block of data centers devoted to cloud applications or scientific computing.The way to optimal exploitation of such a wide amount of computing power passes through the ability of matching the best scheduling of hardware resources with the software characteristics of the applications. This requires appropriate models and evaluation methods.Simulation and analytical techniques are essential tools to support the design and the management process of such architectures, but a sound characterization of the workloads is required. Typical workloads consist in multithreaded applications, with different characteristics, that dynamically span over the cores of multiple machines, connected by fast networks.In this paper we propose several parametric performance models for different configurations of multicore machines, with or without GPU support, running multiple class multithreaded applications, aiming to supply a detailed modeling help for complex data centers. ANALYSIS OF PERFORMANCES ON MULTITHREAD APPLICATIONS 445 ‡ The transcoding of image elements may be performed by means of multiple threads, but the most of the processing is run by a single thread. ANALYSIS OF PERFORMANCES ON MULTITHREAD APPLICATIONS 449 Figure 11. Queuing network of a system running two different types of application with one characterized by general-purpose computing on graphics processing units workload. I/O, input/output; CPU, central processing unit.

show abstract

Section: Motivationmentioning

confidence: 99%

Modeling and analysis of performances for concurrent multithread applications on multicore and graphics processing unit systems

Cerotti

Gribaudo

Iacono

et al. 2015

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Performance and performability measures have been successfully performed taking advantage of multiformalism modeling techniques. Many experiences in supporting and developing multiformalism modeling techniques and frameworks are reported in literature (Sharpe [28], SMART [4], [5], Mobius [8], [6], [7], OsMoSys [30], [13], [25], [14], SIMTHESys [19], [2], [1], [3], [18]). In this paper SIMTHESys is used to support the development of our approach.…”

Section: Related Workmentioning

confidence: 99%

“…In the approach proposed in this paper, PerfBPEL is used to describe the SOA software layer and is implemented as a (graph-based) formalism for the SIMTHESys framework [19], [2], [1], [3], and the architecture layer is described by a dedicated formalism, that describes the deployment of services. Using SIMTHESys enables the use of other formalisms too, such as Generalized Stochastic Petri Nets (GSPN), Queuing Networks (QN) or Fault Trees (FT), to represent services or other architectural components for the convenience of designers.…”

Section: A Why Perfbpelmentioning

confidence: 99%

“…This is based on GSPN, QN, FT and CTMC solving engines, capable of analyzing SOA system models relying on PerfBPEL and the ancillary formalism. Note that the case study presented in the paper does not explicitly show the use of exception handling BPEL constructs due to space restrictions (the use of exception handling primitives [3] in the case study would have complicated the description of the structure). Variables are specified as for the previous subsections with reference to state reduction and stochastic characterization, with the exclusion of correlation management.…”

Section: Current Implementation and Limitationsmentioning

confidence: 99%

See 1 more Smart Citation

PerfBPEL: a graph-based approach for the performance analysis of BPEL SOA applications

Barbierato

Iacono²,

Marrone³

2012

Proceedings of the 6th International Conference on Performance Evaluation Methodologies and Tools

Self Cite

View full text Add to dashboard Cite

Non-functional properties evaluation in Service Oriented Architecture (SOA) is still mostly an open challenge. Although this is a problem that has been already partially explored with some success, there is lack of consolidated results for more complex SOA applications based on services composition. This paper presents a contribution to performance evaluation of SOA-based applications integrated by BPEL. The evaluation technique is based on a performance-oriented reinterpretation of the BPEL specification as a performance modeling language within a multiformalism framework. The approach is based on automatic translation of PerfBPEL into Markov chains and it is implemented by means of SIMTHESys modeling and analysis framework to enable the interaction with other performance oriented formalisms.

show abstract

“…The prediction methods are successfully used in software engineering where time series are obtained through monitoring processes and also for performance evaluation (Barbierato et al 2011;Iacono et al 2010;Kryftis et al 2014Kryftis et al , 2015Skourletopoulos et al 2014;Sotiriadis et al 2010Sotiriadis et al , 2013.…”

Section: Introductionmentioning

confidence: 99%

Data analysis and statistical estimation for time series: improving presentation and interpretation

źErbăźNescu

Pop

2016

Soft Comput

View full text Add to dashboard Cite

In our days in the social sciences, time series (or longitudinal data) are ubiquitous, used in any analytic process, with the main scope to estimate or predict the future. The main issues are represented by the large variety of time series (sometime with an unknown size), the identification of outliers, and by the impossibility to estimate the error or numerical stability of statistical analysis. This paper proposed a matrix-based model for predictive analytics and, using a statistical estimation for different finite samples extracted from time series, estimated the residual and factorial variance for a group of samples. The proposed methods are applied on different samples of social data: number of births in a community, number of inhabitants, natural mobility of population, life expectancy (by sex and area), life expectancy at birth, fertility rate, infant mortality rate.

show abstract

Performability Modeling of Exceptions-Aware Systems in Multiformalism Tools

Cited by 12 publications

References 18 publications

Modeling and analysis of performances for concurrent multithread applications on multicore and graphics processing unit systems

Modeling and analysis of performances for concurrent multithread applications on multicore and graphics processing unit systems

PerfBPEL: a graph-based approach for the performance analysis of BPEL SOA applications

Data analysis and statistical estimation for time series: improving presentation and interpretation

Contact Info

Product

Resources

About