skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Performance Prediction Toolkit

Abstract

The Performance Prediction Toolkit (PPT), is a scalable co-design tool that contains the hardware and middle-ware models, which accept proxy applications as input in runtime prediction. PPT relies on Simian, a parallel discrete event simulation engine in Python or Lua, that uses the process concept, where each computing unit (host, node, core) is a Simian entity. Processes perform their task through message exchanges to remain active, sleep, wake-up, begin and end. The PPT hardware model of a compute core (such as a Haswell core) consists of a set of parameters, such as clock speed, memory hierarchy levels, their respective sizes, cache-lines, access times for different cache levels, average cycle counts of ALU operations, etc. These parameters are ideally read off a spec sheet or are learned using regression models learned from hardware counters (PAPI) data. The compute core model offers an API to the software model, a function called time_compute(), which takes as input a tasklist. A tasklist is an unordered set of ALU, and other CPU-type operations (in particular virtual memory loads and stores). The PPT application model mimics the loop structure of the application and replaces the computational kernels with a call to the hardware model's time_compute() functionmore » giving tasklists as input that model the compute kernel. A PPT application model thus consists of tasklists representing kernels and the high-er level loop structure that we like to think of as pseudo code. The key challenge for the hardware model's time_compute-function is to translate virtual memory accesses into actual cache hierarchy level hits and misses.PPT also contains another CPU core level hardware model, Analytical Memory Model (AMM). The AMM solves this challenge soundly, where our previous alternatives explicitly include the L1,L2,L3 hit-rates as inputs to the tasklists. Explicit hit-rates inevitably only reflect the application modeler's best guess, perhaps informed by a few small test problems using hardware counters; also, hard-coded hit-rates make the hardware model insensitive to changes in cache sizes. Alternatively, we use reuse distance distributions in the tasklists. In general, reuse profiles require the application modeler to run a very expensive trace analysis on the real code that realistically can be done at best for small examples.« less

Authors:
 [1];  [1];  [1];  [1];  [1];  [1];  [1];  [1];  [2];  [2];  [2]
  1. LANL
  2. Florida International University
Publication Date:
Research Org.:
Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1401959
Report Number(s):
PPT; 005497MLTPL00
C17098
DOE Contract Number:
AC52-06NA25396
Resource Type:
Software
Software Revision:
00
Software Package Number:
005497
Software CPU:
MLTPL
Open Source:
Yes
Open source under the BSD license.
Source Code Available:
Yes
Related Software:
SIMIAN PDES
Country of Publication:
United States

Citation Formats

Chennupati, Gopinath, Santhi, Nanadakishore, Eidenbenz, Stephen, Zerr, Robert Joseph, Rosa, Massimiliano, Zamora, Richard James, Park, Eun Jung, Nadiga, Balasubramanya T., Liu, Jason, Ahmed, Kishwar, and Obaida, Mohammad Abu. Performance Prediction Toolkit. Computer software. https://www.osti.gov//servlets/purl/1401959. Vers. 00. USDOE. 25 Sep. 2017. Web.
Chennupati, Gopinath, Santhi, Nanadakishore, Eidenbenz, Stephen, Zerr, Robert Joseph, Rosa, Massimiliano, Zamora, Richard James, Park, Eun Jung, Nadiga, Balasubramanya T., Liu, Jason, Ahmed, Kishwar, & Obaida, Mohammad Abu. (2017, September 25). Performance Prediction Toolkit (Version 00) [Computer software]. https://www.osti.gov//servlets/purl/1401959.
Chennupati, Gopinath, Santhi, Nanadakishore, Eidenbenz, Stephen, Zerr, Robert Joseph, Rosa, Massimiliano, Zamora, Richard James, Park, Eun Jung, Nadiga, Balasubramanya T., Liu, Jason, Ahmed, Kishwar, and Obaida, Mohammad Abu. Performance Prediction Toolkit. Computer software. Version 00. September 25, 2017. https://www.osti.gov//servlets/purl/1401959.
@misc{osti_1401959,
title = {Performance Prediction Toolkit, Version 00},
author = {Chennupati, Gopinath and Santhi, Nanadakishore and Eidenbenz, Stephen and Zerr, Robert Joseph and Rosa, Massimiliano and Zamora, Richard James and Park, Eun Jung and Nadiga, Balasubramanya T. and Liu, Jason and Ahmed, Kishwar and Obaida, Mohammad Abu},
abstractNote = {The Performance Prediction Toolkit (PPT), is a scalable co-design tool that contains the hardware and middle-ware models, which accept proxy applications as input in runtime prediction. PPT relies on Simian, a parallel discrete event simulation engine in Python or Lua, that uses the process concept, where each computing unit (host, node, core) is a Simian entity. Processes perform their task through message exchanges to remain active, sleep, wake-up, begin and end. The PPT hardware model of a compute core (such as a Haswell core) consists of a set of parameters, such as clock speed, memory hierarchy levels, their respective sizes, cache-lines, access times for different cache levels, average cycle counts of ALU operations, etc. These parameters are ideally read off a spec sheet or are learned using regression models learned from hardware counters (PAPI) data. The compute core model offers an API to the software model, a function called time_compute(), which takes as input a tasklist. A tasklist is an unordered set of ALU, and other CPU-type operations (in particular virtual memory loads and stores). The PPT application model mimics the loop structure of the application and replaces the computational kernels with a call to the hardware model's time_compute() function giving tasklists as input that model the compute kernel. A PPT application model thus consists of tasklists representing kernels and the high-er level loop structure that we like to think of as pseudo code. The key challenge for the hardware model's time_compute-function is to translate virtual memory accesses into actual cache hierarchy level hits and misses.PPT also contains another CPU core level hardware model, Analytical Memory Model (AMM). The AMM solves this challenge soundly, where our previous alternatives explicitly include the L1,L2,L3 hit-rates as inputs to the tasklists. Explicit hit-rates inevitably only reflect the application modeler's best guess, perhaps informed by a few small test problems using hardware counters; also, hard-coded hit-rates make the hardware model insensitive to changes in cache sizes. Alternatively, we use reuse distance distributions in the tasklists. In general, reuse profiles require the application modeler to run a very expensive trace analysis on the real code that realistically can be done at best for small examples.},
url = {https://www.osti.gov//servlets/purl/1401959},
doi = {},
year = {Mon Sep 25 00:00:00 EDT 2017},
month = {Mon Sep 25 00:00:00 EDT 2017},
note =
}

Software:
To order this software, request consultation services, or receive further information, please fill out the following request.

Save / Share:
  • The microcomputer program ESPVI 4.0, Electrostatic Precipitation V-I and Performance Model, was developed to provide a user-friendly interface to an advanced model of electrostatic precipitation (ESP) performance. The program is capable of modeling standard ESP configurations as well as those that might be proposed for improved performance. It incorporates many of the latest developments in prediction of ESP performance, including electrical waveform effects, non-rapping reentrainment, and electrode misalignment. The program is organized by a series of menu screens with increasing levels of detail provided as the menus become more specific. A Users Manual (PB92-169614) that is available, provides the documentationmore » needed to load the program from its disk, set up the computer configuration for optimal operation, and introduces the operation of the program.« less
  • This paper reports on the development of a software for determination of thermal conditions of the winding and at the middle of the stator cross-section based on a novel hybrid thermal model and performance prediction using equivalent circuit parameters at fundamental and harmonic voltages of three-phase cage-rotor induction motors when operated at distorted waveforms of the supply system. Static power converters and nonlinear loads used by industries for various purposes change the sinusoidal nature of the system voltages, resulting in injection of harmonic voltages in the power system and when three phase cage-rotor induction motors are connected to this powermore » system, the performances of the motors will not be satisfactory. Due to harmonic currents, the total losses in the motor will be more resulting in higher temperature rise of the motor which may not be allowed due to faster deterioration of the properties of the insulating materials thereby reducing the life expectancy of the motor.« less
  • The Hospital Energy Analysis Toolkit (HEAT) is a menu-driven microcomputer software program designed to help facility managers of existing hospitals evaluate the cost effectiveness of specific Energy Saving Methods (ESMs). The program estimates the energy savings and cost effectiveness of specific ESMs in a user-defined hospital environment. Hospitals are defined in terms of an unlimited number of actual functional space zones, each modeled after one of 21 prototype zones. For each defined zone, the user specifies the floor area, and for some zones the user specifies the types of heating and cooling systems, the percentage of space being actively used,more » and the window orientation. The user also defines an Energy Saving Plan by specifying the current and planned status of each energy parameter applicable to the particular zone. HEAT then computes and reports the energy and economic savings resulting from the Plan. A benchmarking function lets users compare energy use of the entire hospital with national norms to see whether further analysis is warranted. HEAT offers easy-to-use menus and function keys, on-line help screens, and data validation.« less
  • Controls for accelerators and associated systems at the Holifield Radioactive Ion Beam Facility (HRIBF) at Oak Ridge National Laboratory have been migrated from 197Os-vintage minicomputers to a modern system based on Vista and EPICS toolkit software. Stability and capabilities of EPICS software have motivated increasing use of EPICS for accelerator controls. In addition, very inexpensive subsystems based on EPICS and the EPICS portable CA server running on Linux PCs have been implemented to control an ion source test facility and to control a building-access badge reader system. A new object-oriented, extensible display manager has been developed for EPICS to facilitatemore » the transition to EPICS and will be used in place of MEDM. EPICS device support has been developed for CAMAC serial highway controls.« less
  • Analyzing vast quantities of data from diverse information sources is an increasingly important element for nonproliferation and arms control analysis. Much of the work in this area has used human analysts to assimilate, integrate, and interpret complex information gathered from various sources. With the advent of fast computers, one now has the capability to automate this process thereby shifting this burden away from humans. In addition, there now exist huge data storage capabilities which have made it possible to formulate large integrated databases comprising many terabytes of information spanning a variety of subjects. The authors are currently designing a Softwaremore » Toolkit for Analysis Research (STAR) to address these issues. The goal of STAR is to produce a research tool that facilitates the development and interchange of algorithms for locating phenomena of interest to nonproliferation and arms control experts. One major component deals with the preparation of information. The ability to manage and effectively transform raw data into a meaningful form is a prerequisite for analysis by any methodology. The relevant information to be analyzed can be either unstructured text (e.g. journal articles), structured data, signals, or images. Text can be numerical and/or character, stored in raw data files, databases, streams of bytes, or compressed into bits in formats ranging from fixed, to character-delimited, to a count followed by content. The data can be analyzed in real-time or batch mode. Once the data are preprocessed, different analysis techniques can be applied. Some are built using expert knowledge. Others are trained using data collected over a period of time. Currently, the authors are considering three classes of analyzers for use in the software toolkit: (1) traditional machine learning techniques, (2) the purely statistical system, and (3) expert systems.« less

To initiate an order for this software, request consultation services, or receive further information, fill out the request form below. You may also reach us by email at: .

OSTI staff will begin to process an order for scientific and technical software once the payment and signed site license agreement are received. If the forms are not in order, OSTI will contact you. No further action will be taken until all required information and/or payment is received. Orders are usually processed within three to five business days.

Software Request

(required)
(required)
(required)
(required)
(required)
(required)
(required)
(required)