@inproceedings {icl:230, title = {Extending the MPI Specification for Process Fault Tolerance on High Performance Computing Systems}, journal = {Proceedings of ISC2004 (to appear)}, year = {2004}, month = {2004-06}, address = {Heidelberg, Germany}, keywords = {ftmpi, lacsi}, author = {Graham Fagg and Edgar Gabriel and George Bosilca and Thara Angskun and Zizhong Chen and Jelena Pjesivac{\textendash}Grbovic and Kevin London and Jack Dongarra} } @conference {icl:130, title = {Experiences and Lessons Learned with a Portable Interface to Hardware Performance Counters}, booktitle = {PADTAD Workshop, IPDPS 2003}, year = {2003}, month = {2003-04}, publisher = {IEEE}, organization = {IEEE}, address = {Nice, France}, abstract = {The PAPI project has defined and implemented a cross-platform interface to the hardware counters available on most modern microprocessors. The interface has gained widespread use and acceptance from hardware vendors, users, and tool developers. This paper reports on experiences with the community-based open-source effort to define the PAPI specification and implement it on a variety of platforms. Collaborations with tool developers who have incorporated support for PAPI are described. Issues related to interpretation and accuracy of hardware counter data and to the overheads of collecting this data are discussed. The paper concludes with implications for the design of the next version of PAPI.}, keywords = {lacsi, papi}, isbn = {0-7695-1926-1}, author = {Jack Dongarra and Kevin London and Shirley Moore and Phil Mucci and Dan Terpstra and Haihang You and Min Zhou} } @conference {icl:15, title = {End-user Tools for Application Performance Analysis, Using Hardware Counters}, booktitle = {International Conference on Parallel and Distributed Computing Systems}, year = {2001}, month = {2001-08}, address = {Dallas, TX}, abstract = {One purpose of the end-user tools described in this paper is to give users a graphical representation of performance information that has been gathered by instrumenting an application with the PAPI library. PAPI is a project that specifies a standard API for accessing hardware performance counters available on most modern microprocessors. These counters exist as a small set of registers that count \"events\", which are occurrences of specific signals and states related to a processor{\textquoteright}s function. Monitoring these events facilitates correlation between the structure of source/object code and the efficiency of the mapping of that code to the underlying architecture. The perfometer tool developed by the PAPI project provides a graphical view of this information, allowing users to quickly see where performance bottlenecks are in their application. Only one function call has to be added by the user to their program to take advantage of perfometer. This makes it quick and simple to add and remove instrumentation from a program. Also, perfometer allows users to change the \"event\" they are monitoring. Add the ability to monitor parallel applications, set alarms and a Java front-end that can run anywhere, and this gives the user a powerful tool for quickly discovering where and why a bottleneck exists. A number of third-party tools for analyzing performance of message-passing and/or threaded programs have also incorporated support for PAPI so as to be able to display and analyze hardware counter data from their interfaces.}, keywords = {papi}, author = {Kevin London and Jack Dongarra and Shirley Moore and Phil Mucci and Keith Seymour and T. Spencer} } @conference {icl:16, title = {The PAPI Cross-Platform Interface to Hardware Performance Counters}, booktitle = {Department of Defense Users{\textquoteright} Group Conference Proceedings}, year = {2001}, month = {2001-06}, address = {Biloxi, Mississippi}, abstract = {The purpose of the PAPI project is to specify a standard API for accessing hardware performance counters available on most modern microprocessors. These counters exist as a small set of registers that count \"events,\" which are occurrences of specific signals and states related to the processor{\textquoteright}s function. Monitoring these events facilitates correlation between the structure of source/object code and the efficiency of the mapping of that code to the underlying architecture. This correlation has a variety of uses in performance analysis and tuning. The PAPI project has developed a standard set of hardware events and a standard cross-platform library interface to the underlying counter hardware. The PAPI library has been implemented for a number of Shared Resource Center platforms. The PAPI project is developing end-user tools for dynamically selecting and displaying hardware counter performance data. PAPI support is also being incorporated into a number of third-party tools.}, keywords = {papi}, author = {Kevin London and Shirley Moore and Phil Mucci and Keith Seymour and Richard Luczak} } @article {icl:20, title = {Review of Performance Analysis Tools for MPI Parallel Programs}, journal = {European Parallel Virtual Machine / Message Passing Interface Users{\textquoteright} Group Meeting, Lecture Notes in Computer Science 2131}, year = {2001}, month = {2001-09}, pages = {241-248}, publisher = {Springer Verlag, Berlin}, address = {Greece}, abstract = {In order to produce MPI applications that perform well on today{\textquoteright}s parallel architectures, programmers need effective tools for collecting and analyzing performance data. A variety of such tools, both commercial and research, are becoming available. This paper reviews and evaluations the available cross-platform MPI performance analysis tools.}, keywords = {papi}, doi = {https://doi.org/10.1007/3-540-45417-9_34}, author = {Shirley Moore and David Cronk and Kevin London and Jack Dongarra} } @conference {icl:11, title = {Using PAPI for Hardware Performance Monitoring on Linux Systems}, booktitle = {Conference on Linux Clusters: The HPC Revolution}, year = {2001}, month = {2001-06}, publisher = {Linux Clusters Institute}, organization = {Linux Clusters Institute}, address = {Urbana, Illinois}, abstract = {PAPI is a specification of a cross-platform interface to hardware performance counters on modern microprocessors. These counters exist as a small set of registers that count events, which are occurrences of specific signals related to a processor{\textquoteright}s function. Monitoring these events has a variety of uses in application performance analysis and tuning. The PAPI specification consists of both a standard set of events deemed most relevant for application performance tuning, as well as both high-level and low-level sets of routines for accessing the counters. The high level interface simply provides the ability to start, stop, and read sets of events, and is intended for the acquisition of simple but accurate measurement by application engineers. The fully programmable low-level interface provides sophisticated options for controlling the counters, such as setting thresholds for interrupt on overflow, as well as access to all native counting modes and events, and is intended for third-party tool writers or users with more sophisticated needs. PAPI has been implemented on a number of platforms, including Linux/x86 and Linux/IA-64. The Linux/x86 implementation requires a kernel patch that provides a driver for the hardware counters. The driver memory maps the counter registers into user space and allows virtualizing the counters on a perprocess or per-thread basis. The kernel patch is being proposed for inclusion in the main Linux tree. The PAPI library provides access on Linux platforms not only to the standard set of events mentioned above but also to all the Linux/x86 and Linux/IA-64 native events. PAPI has been installed and is in use, either directly or through incorporation into third-party end-user performance analysis tools, on a number of Linux clusters, including the New Mexico LosLobos cluster and Linux clusters at NCSA and the University of Tennessee being used for the GrADS (Grid Application Development Software) project. }, keywords = {papi}, author = {Jack Dongarra and Kevin London and Shirley Moore and Phil Mucci and Dan Terpstra} } @techreport {icl:226, title = {A Portable Programming Interface for Performance Evaluation on Modern Processors}, journal = {University of Tennessee Computer Science Technical Report, UT-CS-00-444}, year = {2000}, month = {2000-07}, author = {Shirley Browne and Jack Dongarra and Nathan Garner and Kevin London and Phil Mucci} } @inproceedings {icl:32, title = {A Scalable Cross-Platform Infrastructure for Application Performance Tuning Using Hardware Counters}, journal = {Proceedings of SuperComputing 2000 (SC{\textquoteright}00)}, year = {2000}, month = {2000-11}, address = {Dallas, TX}, keywords = {papi}, author = {Shirley Browne and Jack Dongarra and Nathan Garner and Kevin London and Phil Mucci} }