SvPablo: A Multi-language Performance Analysis System

Rose, Luiz De; Zhang, Ying; Reed, Daniel A.

doi:10.1007/3-540-68061-6_31

Cited by 55 publications

(28 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Again, though, this work only supports visualizations of how the observed data relates to the application source code, and not how it relates to application semantics. Finally, many existing parallel performance tracing frameworks [7], [9], [15], [16], [17] attempt to visualize the behavior of large-scale parallel programs, either by visualizing communication between processes, by visualizing hardware metrics on a torus, or by examining communication traces using three-dimensional views. None of these, however, support the projection of application data into performance domains or vice versa, limiting their ability to pinpoint performance bottlenecks through the kind of correlation analysis presented in this paper.…”

Section: Per-core Per-phase Datamentioning

confidence: 99%

Novel views of performance data to analyze large-scale adaptive applications

Bhatelé

Gamblin

Isaacs

et al. 2012

2012 International Conference for High Performance Computing, Networking, Storage and Analysis

View full text Add to dashboard Cite

Abstract-Performance analysis of parallel scientific codes is becoming increasingly difficult due to the rapidly growing complexity of applications and architectures. Existing tools fall short in providing intuitive views that facilitate the process of performance debugging and tuning. In this paper, we extend recent ideas of projecting and visualizing performance data for faster, more intuitive analysis of applications. We collect detailed per-level and per-phase measurements for a dynamically load-balanced, structured AMR library and project per-core data collected in the hardware domain on to the application's communication topology. We show how our projections and visualizations lead to a rapid diagnosis of and mitigation strategy for a previously elusive scaling bottleneck in the library that is hard to detect using conventional tools. Our new insights have resulted in a 22% performance improvement for a 65,536-core run of the AMR library on an IBM Blue Gene/P system.

show abstract

Section: Per-core Per-phase Datamentioning

confidence: 99%

Novel views of performance data to analyze large-scale adaptive applications

Bhatelé

Gamblin

Isaacs

et al. 2012

2012 International Conference for High Performance Computing, Networking, Storage and Analysis

View full text Add to dashboard Cite

show abstract

“…Finally, many existing parallel performance tracing frameworks [11][12][13][14]22] attempt to visualize the behavior of largescale parallel programs, either by visualizing communication between processes, by visualizing hardware metrics on a torus, or by examining communication traces using threedimensional views. None of these, however, support the projection of application data into performance domains or vice versa, limiting their ability to pinpoint performance bottlenecks through the kind of correlation analysis presented here.…”

Section: Related Workmentioning

confidence: 99%

Simplifying Performance Analysis of Large-scale Adaptive Scientific Applications

Bhatelé¹,

Gamblin²,

Gunney³

et al. 2012

View full text Add to dashboard Cite

Performance analysis of parallel scientific codes is becoming increasingly difficult due to the rapidly growing complexity of applications and architectures. Existing tools fall short in providing intuitive views that facilitate the process of performance debugging and tuning. In this paper, we exploit a recent idea of projecting and visualizing performance data on the communication and hardware domain for faster, more intuitive analysis of applications. We leverage several performance analysis and visualization tools to showcase the discovery of scalability bottlenecks in a structured AMR library. Using novel techniques to project per-phase timing data, application data, and communication data on a communication graph, we identify a previously elusive scaling bottleneck in the library. We present solutions that mitigate this problem, resulting in 22% improvement in the performance for a 65,536-core run on an IBM Blue Gene/P system.

show abstract

“…These tools include prof [14], gprof [15], DCPI [16], HPCToolkit [17], and Speedshop [18]. Similar instrumentation based profiling tools include TAU [13] and SvPablo [19]. Like the mappings discussed above, the primary focus of these tools is code regions and base data structures.…”

Section: Related Workmentioning

confidence: 99%

Assigning Blame: Mapping Performance to High Level Parallel Programming Abstractions

Rutar

Hollingsworth

2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Parallel programs are increasingly being written using programming frameworks and other environments that allow parallel constructs to be programmed with greater ease. The data structures used allow the modeling of complex mathematical structures like linear systems and partial differential equations using high-level programming abstractions. While this allows programmers to model complex systems in a more intuitive way, it also makes the debugging and profiling of these systems more difficult due to the complexity of mapping these high level abstractions down to the low level parallel programming constructs. This work discusses mapping mechanisms, called variable blame, for creating these mappings and using them to assist in the profiling and debugging of programs created using advanced parallel programming techniques. We also include an example of a prototype implementation of the system profiling three programs.

show abstract

SvPablo: A Multi-language Performance Analysis System

Cited by 55 publications

References 1 publication

Novel views of performance data to analyze large-scale adaptive applications

Novel views of performance data to analyze large-scale adaptive applications

Simplifying Performance Analysis of Large-scale Adaptive Scientific Applications

Assigning Blame: Mapping Performance to High Level Parallel Programming Abstractions

Contact Info

Product

Resources

About