Triva: Interactive 3D visualization for performance analysis of parallel applications

Schnorr, Lucas Mello; Huard, Guillaume; Navaux, Philippe O. A.

doi:10.1016/j.future.2009.10.006

Cited by 19 publications

(11 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In conclusion, the "multilevel cache" helps the 3D display speed reach ~0.005 s per device for one user, and ~0.04 s per device for 20 users simultaneously. Such experimental results verify the simulated results well, and this cache optimization also could be widely used to promote the efficiency of visualizations for many other applications [16]- [17]. …”

Section: Experimental Analysissupporting

confidence: 71%

3D visualized data management for network devices based on multilevel caches

Jiang¹,

Chen²,

Cai³

et al. 2014

ijsgce

View full text Add to dashboard Cite

Three dimensional visualization technologies are dramatically improving convenience and efficiency in modern industrial managements. In this paper, a novel multilevel caches-based 3D visualization system for managing device data, which could be applied for displaying 3D scenes and processing device data simultaneously, is investigated. The system is fast-response, compared to conventional strategies. This work may provide helpful references and inspirations for applications about computer based visualization, complex data management and network communications.

show abstract

Section: Experimental Analysissupporting

confidence: 71%

3D visualized data management for network devices based on multilevel caches

Jiang¹,

Chen²,

Cai³

et al. 2014

ijsgce

View full text Add to dashboard Cite

show abstract

“…Our new experimental scenarios have larger errors, making "bad cases" easier to identify. To understand the root causes of the errors we use a multi-scale, interactive and exploratory visualization tool called Triva [Schnorr et al 2011;Schnorr et al 2010;Triva 2011]. We select some of the scenarios that have large maximum error and run them until completion of all flows so that we can compare flow completion times and not only instantaneous bandwidth shares.…”

Section: Reverse Traffic Effectsmentioning

confidence: 99%

On the validity of flow-level tcp network models for grid and cloud simulations

Velho

Schnorr

Casanova

et al. 2013

ACM Trans. Model. Comput. Simul.

Self Cite

View full text Add to dashboard Cite

Researchers in the area of grid/cloud computing perform many of their experiments using simulations that must capture network behavior. In this context, packet-level simulations, which are widely used to study network protocols, are too costly given the typical large scales of simulated systems and applications. An alternative is to implement network simulations with less costly flow-level models. Several flow-level models have been proposed and implemented in grid/cloud simulators. Surprisingly, published validations of these models, if any, consist of verifications for only a few simple cases. Consequently, even when they have been used to obtain published results, the ability of these simulators to produce scientifically meaningful results is in doubt. This work evaluates these state-of-the-art flow-level network models of TCP communication via comparison to packet-level simulation. While it is straightforward to show cases in which previously proposed models lead to good results, instead we follow the critical method, which places model refutation at the center of the scientific activity, and we systematically seek cases that lead to invalid results. Careful analysis of these cases reveal fundamental flaws and also suggest improvements. One contribution of this work is that these improvements lead to a new model that, while far from being perfect, improves upon all previously proposed models in the context of simulation of grids or clouds. A more important contribution, perhaps, is provided by the pitfalls and unexpected behaviors encountered in this work, leading to a number of enlightening lessons. In particular, this work shows that model validation cannot be achieved solely by exhibiting (possibly many) "good cases." Confidence in the quality of a model can only be strengthened through an invalidation approach that attempts to prove the model wrong.

show abstract

“…Triva (Schnorr et al, 2010) analyses paje trace files from the execution of parallel applications. Together with the GraphViz library (Ellson et al, 2001), Triva presents graphical information about the behavior of monitored applications.…”

Section: Simulation Toolsmentioning

confidence: 99%

Energy Consumption in Peer-to-Peer Protocols for Ubiquitous Devices

Camargo

Pilla

Fonseca

et al. 2012

jacr

View full text Add to dashboard Cite

This paper studies energy consumption in peer-to-peer protocols in the context of file distribution among border devices and wireless sensors, which are limited in processing power and battery duration. Gnutella, Chord, CAN, Pastry, and Tapestry protocols were compared to the client-server approach in simulations using SimGrid and Triva, analyzing the energy cost of message exchanging. The peer-topeer protocols presented less consumption to distribute a file among 10 devices, and the best results were achieved by the Chord protocol. The client-server architecture presented the worst results, as expected, due to the high concentration of load in a single server. Processing costs of the Gnutella protocol were compared to the client-server's, with very similar results.

show abstract

Triva: Interactive 3D visualization for performance analysis of parallel applications

Cited by 19 publications

References 9 publications

3D visualized data management for network devices based on multilevel caches

3D visualized data management for network devices based on multilevel caches

On the validity of flow-level tcp network models for grid and cloud simulations

Energy Consumption in Peer-to-Peer Protocols for Ubiquitous Devices

Contact Info

Product

Resources

About