LC Hotline: 2-4531

From offsite: (925) 422-4531

 

Hours

Monday–Friday
8am–12pm, 1–4:45pm
B453 R1103 | Q-clearance area

 

Parallel Performance Evaluation Using TAU

August 28–29, 2017
 

Date / Time This event has already occurred.  Page is archival.
Description

To meet the needs of computational scientists to evaluate the performance of their parallel, scientific applications, we present the TAU Performance System and its interfaces to other tools such as PAPI, Score-P, Scalasca, OTF and Vampir. This one day workshop will cover performance evaluation of applications on Tri-lab OCF platforms. This workshop will focus on performance data collection, analysis, and performance optimization. After describing and demonstrating how performance data (both profile and trace data) can be collected in a straightforward manner using TAU's (Tuning and Analysis Utilities) automated instrumentation, the workshop will cover how to analyze the performance data collected and drill down to find performance bottlenecks and determine their causes. The workshop will include some sample codes that illustrate the different instrumentation and measurement choices available to the users. Topics will cover generating performance profiles and traces with OpenMP intrumentation using the OpenMP Tools API with Intel compilers, memory utilization, I/O, and hardware performance counters data using PAPI. Hardware counter data can show not only which routines are taking the most time, but why? For example, because of cache misses, TLB misses, excess address arithmetic, or poor branch prediction behavior. We will demonstrate scalable tracing using Score-P and OTF and visualization using the Vampir trace visualizers. Performance data analysis using ParaProf and PerfExplorer will be demonstrated using TAU's database technology (TAUdb). The workshop will also feature cross experiment analysis including comparing the effects of multi-core architectures on code performance. We will attempt to collect and analyze performance data for additional user codes during the hands-on portion of the workshop. Users and developers are welcome to contact the instructor ahead of time to begin collecting data so as to have it on hand for the workshop. Additional information about TAU can be found at tau.uoregon.edu 

Workshop (Aug 28): Attendees may use their own Livermore Computing (LC) computer accounts on clusters such as cab, quartz vulcan, etc. If you do not have an account on an LC cluster, you can use a temporary workshop account provided during the workshop, or you can request an account in advance through the LC Hotline ( lc-support@llnl.gov). 

Individual Sessions (Aug 29): Interested developers can schedule individual meeting times with the speaker, which can include hands-on work with their codes/projects (bring your own laptop if you wish to do this). These individual sessions must be scheduled in advance by contacting Blaise Barney at 422-2578 or blaiseb@llnl.gov

Agenda Aug 28 Workshop: (T1889 classroom 2) 
Morning
  • Introduction to TAU
  • Instrumentation: PDT, MPI, OpenMP OMPT, tau_exec
  • I/O, and memory evaluation
  • Hands-on
  • PAPI
  • Hands-on using loop level instrumentation, PAPI

Afternoon

  • Demonstration of analysis tools: Paraprof, TAUdb and PerfExplorer
  • Vampir and Jumpshot
  • Hands-on

Aug 29 Individual Sessions: (B453 R1020)

  • Applying performance evaluation tools to user codes
  • Time slots must be reserved in advance - contact Blaise Barney (925-422-2578 / blaiseb@llnl.gov) for details.
About the Instructor Sameer Shende (sameer@paratools.com) serves as the Director of the Performance Research Laboratory at the University of Oregon and the President of ParaTools, Inc. He received his Ph.D. from the University of Oregon and B.Tech from the Indian Institute of Technology, Bombay. He has helped develop the TAU Performance System, Program Database Toolkit (PDT), Parallel Tools Runtime Environment (PToolsRTE), and the HPCLinux distribution. His areas of interest include tools and techniques for performance instrumentation, measurement, performance analysis, runtime systems, and compiler optimizations.
Location Workshop: Aug 28, 2016 9:00am – 5:00pm; Laboratory Training Center 2, Trailer 1889 (near the West Gate Badge Office). This is a Common Access Area. Directions and contact information are available at: https://lab-training-center-int.llnl.gov/xlocation.html

Individual Sessions: Aug 29, 2016 9:00am – 5:00pm; Building 453, Room 1020 (Wildcat Canyon). Note: Time slots for these must be scheduled in advance.
This is a Property Protection Area. Foreign national temporary escorted building access procedures apply.

What to Bring Computers are provided for the workshop on Aug 28.  For the individual sessions on Aug 29, attendees should bring a laptop computer
Fee No cost
Level/Prerequisites Introductory level. A basic understanding of parallel programming with C or Fortran is essential.
Registration

Aug 28 Workshop: You must register in advance. Registration is limited to LLNL employees, students and collaborators. Note that enrollment is limited to 20 attendees, due to the number of available workstations.

  • Register for course number EC3528 through LTRAIN.
  • Foreign Nationals: T1889 is a Common Access Area, so no escort is required for employee foreign nationals.


Aug 29 Individual Sessions: Those wishing to schedule individual sessions with the instructor must do this in advance by contacting Blaise Barney at 422-2578 or blaiseb@llnl.gov

Questions? Please call or email Blaise Barney (925-422-2578 / blaiseb@llnl.gov)

 

 

UCRL-MI-138859