pr\u00e9sentent<\/p>\n\n\n\n
Autoris\u00e9 \u00e0 pr\u00e9senter ses travaux en vue de l\u2019obtention du Doctorat de l’Institut Polytechnique de Paris, pr\u00e9par\u00e9 \u00e0 T\u00e9l\u00e9com SudParis en :<\/p>\n\n\n\n
le LUNDI 28 NOVEMBRE 2022 \u00e0 14h00<\/p>\n\n\n\n
Amphi 3 – T\u00e9l\u00e9com SudParis
19 place Marguerite Perey 91120 Palaiseau<\/p>\n\n\n\n
Membres du jury :<\/strong><\/p>\n\n\n\n M. Denis\u00a0CONAN<\/strong>, Ma\u00eetre de conf\u00e9rences, T\u00e9l\u00e9com SudParis, FRANCE – Directeur de th\u00e8se Les supports d’ex\u00e9cution sont couramment utilis\u00e9s par les applications parall\u00e8les afin d’exploiter efficacement les ressources mat\u00e9rielles sous-jacentes. Un support d’ex\u00e9cution dissimule la complexit\u00e9 de la gestion du mat\u00e9riel et expose aux d\u00e9veloppeurs d’applications une interface de haut niveau. \u00c0 cette fin, il prend des d\u00e9cisions en s’appuyant sur des heuristiques qui estiment le comportement futur de l’application. Nous proposons Pythia, une biblioth\u00e8que utilisable comme oracle capable de pr\u00e9dire le comportement futur d’une application afin que les supports d’ex\u00e9cutions puissent prendre des d\u00e9cisions plus \u00e9clair\u00e9es. Pythia s’appuie sur la nature d\u00e9terministe de nombreuses applications HPC : en enregistrant une trace d’ex\u00e9cution, Pythia capture le comportement g\u00e9n\u00e9ral de l’application. Cette trace est ensuite utilis\u00e9e lors des ex\u00e9cutions suivantes de l’application et le support d’ex\u00e9cution peut demander des pr\u00e9dictions sur le comportement futur du programme. Nous \u00e9valuons Pythia sur 13 applications MPI et nous montrons que Pythia peut pr\u00e9dire pr\u00e9cis\u00e9ment le futur de la plupart de ces applications, m\u00eame en variant la taille des donn\u00e9es. Nous montrons comment les pr\u00e9dictions de Pythia peuvent guider l’optimisation d’un support d’ex\u00e9cution en impl\u00e9mentant une strat\u00e9gie de parall\u00e9lisme adaptatif dans le support d’ex\u00e9cution GNU OpenMP. L’\u00e9valuation montre que, gr\u00e2ce aux pr\u00e9dictions de Pythia cette strat\u00e9gie adaptative peut r\u00e9duire le temps d’ex\u00e9cution d’une application jusqu’\u00e0 38%.<\/p>\n\n\n\n Runtime systems are commonly used by parallel applications in order to efficiently exploit the underlying hardware resources. A runtime system hides the complexity of the management of the hardware and exposes a high-level interface to application developers. To this end, it makes decisions by relying on heuristics that estimate the future behavior of the application. We propose Pythia, a library that serves as an oracle capable of predicting the future behavior of an application, so that the runtime system can make more informed decisions. Pythia builds on the deterministic nature of many HPC applications: by recording an execution trace, Pythia captures the application main behavior. The trace can be provided for future executions of the application, and a runtime system can ask for predictions of future program behavior. We evaluate Pythia on 13 MPI applications and show that Pythia can accurately predict the future of most of these applications, even when varying the problem size. We demonstrate how Pythia predictions can guide a runtime system optimization by implementing an adaptive thread parallelism strategy in GNU OpenMP runtime system. The evaluation shows that, thanks to Pythia prediction, the adaptive strategy reduces the execution time of an application by up to 38%.<\/p>\n","protected":false},"excerpt":{"rendered":" L’Ecole doctorale : Ecole Doctorale de l’Institut Polytechnique de Pariset le Laboratoire de recherche SAMOVAR – Services r\u00e9partis, Architectures, MOd\u00e9lisation, Validation, Administration des R\u00e9seaux pr\u00e9sentent l\u2019AVIS DE SOUTENANCE de Monsieur Alexis COLIN Autoris\u00e9 \u00e0 pr\u00e9senter ses travaux en vue de l\u2019obtention du Doctorat de l’Institut Polytechnique de Paris, pr\u00e9par\u00e9 \u00e0 T\u00e9l\u00e9com SudParis en : Informatique […]<\/p>\n","protected":false},"author":4,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"ocean_post_layout":"","ocean_both_sidebars_style":"","ocean_both_sidebars_content_width":0,"ocean_both_sidebars_sidebars_width":0,"ocean_sidebar":"0","ocean_second_sidebar":"0","ocean_disable_margins":"enable","ocean_add_body_class":"","ocean_shortcode_before_top_bar":"","ocean_shortcode_after_top_bar":"","ocean_shortcode_before_header":"","ocean_shortcode_after_header":"","ocean_has_shortcode":"","ocean_shortcode_after_title":"","ocean_shortcode_before_footer_widgets":"","ocean_shortcode_after_footer_widgets":"","ocean_shortcode_before_footer_bottom":"","ocean_shortcode_after_footer_bottom":"","ocean_display_top_bar":"default","ocean_display_header":"default","ocean_header_style":"","ocean_center_header_left_menu":"0","ocean_custom_header_template":"0","ocean_custom_logo":0,"ocean_custom_retina_logo":0,"ocean_custom_logo_max_width":0,"ocean_custom_logo_tablet_max_width":0,"ocean_custom_logo_mobile_max_width":0,"ocean_custom_logo_max_height":0,"ocean_custom_logo_tablet_max_height":0,"ocean_custom_logo_mobile_max_height":0,"ocean_header_custom_menu":"0","ocean_menu_typo_font_family":"0","ocean_menu_typo_font_subset":"","ocean_menu_typo_font_size":0,"ocean_menu_typo_font_size_tablet":0,"ocean_menu_typo_font_size_mobile":0,"ocean_menu_typo_font_size_unit":"px","ocean_menu_typo_font_weight":"","ocean_menu_typo_font_weight_tablet":"","ocean_menu_typo_font_weight_mobile":"","ocean_menu_typo_transform":"","ocean_menu_typo_transform_tablet":"","ocean_menu_typo_transform_mobile":"","ocean_menu_typo_line_height":0,"ocean_menu_typo_line_height_tablet":0,"ocean_menu_typo_line_height_mobile":0,"ocean_menu_typo_line_height_unit":"","ocean_menu_typo_spacing":0,"ocean_menu_typo_spacing_tablet":0,"ocean_menu_typo_spacing_mobile":0,"ocean_menu_typo_spacing_unit":"","ocean_menu_link_color":"","ocean_menu_link_color_hover":"","ocean_menu_link_color_active":"","ocean_menu_link_background":"","ocean_menu_link_hover_background":"","ocean_menu_link_active_background":"","ocean_menu_social_links_bg":"","ocean_menu_social_hover_links_bg":"","ocean_menu_social_links_color":"","ocean_menu_social_hover_links_color":"","ocean_disable_title":"default","ocean_disable_heading":"default","ocean_post_title":"","ocean_post_subheading":"","ocean_post_title_style":"","ocean_post_title_background_color":"","ocean_post_title_background":0,"ocean_post_title_bg_image_position":"","ocean_post_title_bg_image_attachment":"","ocean_post_title_bg_image_repeat":"","ocean_post_title_bg_image_size":"","ocean_post_title_height":0,"ocean_post_title_bg_overlay":0.5,"ocean_post_title_bg_overlay_color":"","ocean_disable_breadcrumbs":"default","ocean_breadcrumbs_color":"","ocean_breadcrumbs_separator_color":"","ocean_breadcrumbs_links_color":"","ocean_breadcrumbs_links_hover_color":"","ocean_display_footer_widgets":"default","ocean_display_footer_bottom":"default","ocean_custom_footer_template":"0","ocean_post_oembed":"","ocean_post_self_hosted_media":"","ocean_post_video_embed":"","ocean_link_format":"","ocean_link_format_target":"self","ocean_quote_format":"","ocean_quote_format_link":"post","ocean_gallery_link_images":"off","ocean_gallery_id":[],"footnotes":""},"categories":[286,549],"tags":[],"class_list":["post-5039","post","type-post","status-publish","format-standard","hentry","category-fractualites-ennews-fr","category-seminaire-acmes","entry"],"_links":{"self":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/5039","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/comments?post=5039"}],"version-history":[{"count":2,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/5039\/revisions"}],"predecessor-version":[{"id":5041,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/5039\/revisions\/5041"}],"wp:attachment":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/media?parent=5039"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/categories?post=5039"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/tags?post=5039"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
Mme Camille\u00a0COTI<\/strong>, Professeure, Universit\u00e9 du Qu\u00e9bec, CANADA – Rapporteure
M. Samuel\u00a0THIBAULT<\/strong>, Professeur des universit\u00e9s, INRIA, Universit\u00e9 de Bordeaux, FRANCE – Rapporteur
M. Fran\u00e7ois\u00a0TRAHAY<\/strong>, Ma\u00eetre de conf\u00e9rences, T\u00e9l\u00e9com SudParis, FRANCE – Co-encadrant de th\u00e8se
M. Arnaud\u00a0LEGRAND<\/strong>, Directeur de recherche, IMAG, FRANCE – Examinateur
M. Patrick\u00a0CARRIBAULT<\/strong>, Directeur de recherche, CEA\/DAM, FRANCE – Examinateur
Mme Amel\u00a0BOUZEGHOUB<\/strong>, Professeure, T\u00e9l\u00e9com SudParis, FRANCE – Examinatrice<\/p>\n\n\n\n
R\u00e9sum\u00e9 :<\/strong><\/p>\n\n\n\n
Abstract : \u00ab\u00a0From trace collection to the prediction of the behaviour of parallel applications\u00a0\u00bb<\/strong><\/p>\n\n\n\n