{"id":1773,"date":"2023-02-02T10:56:40","date_gmt":"2023-02-02T10:56:40","guid":{"rendered":"https:\/\/hpc.grnet.gr\/?page_id=1773"},"modified":"2026-06-15T10:27:55","modified_gmt":"2026-06-15T10:27:55","slug":"preparatory-access","status":"publish","type":"page","link":"https:\/\/www.hpc.grnet.gr\/en\/preparatory-access\/","title":{"rendered":"Call for Preparation\/Development Projects in the ARIS System"},"content":{"rendered":"<h2 class=\"p1\">General<\/h2>\n<p class=\"p2\">The National Infrastructures for Research and Technology (GRNET S.A.) provides resources on the ARIS supercomputing system for the execution of preparatory computational projects.<\/p>\n<p class=\"p2\">The call for preparatory projects on the ARIS system is continuously open for proposals. These proposals are reviewed by GRNET&#8217;s technical committee.<\/p>\n<p class=\"p2\">Note: The evaluation of proposals is a continuous process and researchers should expect results within 10 working days from the date of submission of their application.<\/p>\n<h2 class=\"p1\">Start date<\/h2>\n<p class=\"p2\">Selected projects begin up to one month after their evaluation date.<\/p>\n<h2 class=\"p1\">Objectives of the call<\/h2>\n<p class=\"p2\">The call for preparatory projects aims to give researchers working at Greek educational or research institutions access to the ARIS system for the purpose of:<\/p>\n<ul class=\"ul1\">\n<li class=\"li2\">Conducting application scaling tests and potentially resolving related issues.<\/li>\n<li class=\"li2\">Supporting porting and optimization tests of parallel application code.<\/li>\n<\/ul>\n<p class=\"p2\">The primary objective of providing access for preparatory projects is to allow researchers to test their codes and applications in order to strengthen their applications for production project calls.<\/p>\n<p class=\"p2\">Types of preparatory projects:<\/p>\n<ul class=\"ul1\">\n<li class=\"li2\">Type A preparatory projects<\/li>\n<li class=\"li2\">Type B preparatory projects<\/li>\n<\/ul>\n<p class=\"p2\"><b>Type A: Code scaling tests<\/b><\/p>\n<p class=\"p2\">Type A preparatory projects aim to create performance graphs of code and data on the ARIS system, allowing researchers to use them in their applications for production projects.<\/p>\n<p class=\"p2\">The maximum access duration for Type A projects is 2 months, while the maximum number of core hours provided is 50,000 core hours on the Thin node island, 25,000 core hours on the GPU island, and 25,000 core hours on the Fat Node Island. The total number of core hours that can be requested must not exceed 50,000 core hours. On all islands, core hours are counted at the <span class=\"s1\">CPU level<\/span>.<\/p>\n<p class=\"p2\">After the 2-month period, and within 30 days, researchers must send GRNET a report on the execution and results of the preparatory project, based on the procedure described <a href=\"https:\/\/www.hpc.grnet.gr\/access_policy\/progress-reports\/\"><span class=\"s2\">here<\/span><\/a>.<\/p>\n<p class=\"p2\"><b>Type B: Code porting and optimization<\/b><\/p>\n<p class=\"p2\">Type B preparatory projects aim to give users access to the system for the purpose of porting and optimizing their code so that it operates efficiently on the ARIS system. Users must include in their application a detailed development plan, including information about the human resources available to the project and their technical skills.<\/p>\n<p class=\"p2\">The maximum access duration for Type B projects is 4 months, while the maximum number of core hours provided is 50,000 core hours on the Thin node island, 25,000 core hours on the GPU island, and 25,000 core hours on the Fat Node Island. The total number of core hours that can be requested must not exceed 50,000 core hours. On all islands, core hours are counted at the CPU level.<\/p>\n<p class=\"p2\">After the 4-month period, and within 30 days, researchers must send GRNET a development, execution, and results report for the preparatory projects, based on the procedure described <a href=\"https:\/\/www.hpc.grnet.gr\/access_policy\/progress-reports\/\"><span class=\"s2\">here<\/span><\/a>.<\/p>\n<h2 class=\"p1\">Eligibility<\/h2>\n<p class=\"p2\">Eligible candidates are scientists and researchers working at academic or research institutions in Greece.<\/p>\n<p class=\"p2\">Candidates must commit to using the resources allocated on the system, as well as to providing a report of their work in accordance with the proposed schedules (see above). In addition, users must acknowledge the use of GRNET&#8217;s ARIS system in all publications related to results obtained from the use of the allocated resources.<\/p>\n<p class=\"p2\">An indicative acknowledgement format is the following:<\/p>\n<p class=\"p2\">&#8220;<b><i>We acknowledge GRNET S.A. for awarding us access to the Greek national high performance systems, ARIS<\/i><\/b>.&#8221;<\/p>\n<p class=\"p2\">Users acknowledge GRNET&#8217;s right to publish the project summary as well as the technical performance results of the applications on the ARIS system.<\/p>\n<h2 class=\"p1\">HPC Computing System &#8211; ARIS<\/h2>\n<p class=\"p2\">The ARIS infrastructure has been expanded and now consists of three islands of computing systems based on AMD x86 architecture, interconnected in a unified NVIDIA Infiniband HDR 100 Gbps network that offers multiple processing capabilities and architectures. In detail, the infrastructure consists of:<\/p>\n<ul class=\"ul1\">\n<li class=\"li2\">A thin node island consisting of 48 Dell PowerEdge R6525 servers. Each server offers 2 AMD EPYC 7763 processors and 512 GB of main memory.<\/li>\n<li class=\"li2\">A fat node island (large memory nodes) consisting of 16 Dell PowerEdge R6525 servers. Each server offers 2 AMD EPYC 7763 processors and 1 TB of main memory.<\/li>\n<li class=\"li2\">A GPU accelerator node island consisting of 3 Dell PowerEdge XE8545 servers. Each server contains 2 AMD EPYC 7763 processors, 512 GB of memory, and 4 NVIDIA Ampere A100 cards.<\/li>\n<li class=\"li2\">To optimize system usage, one of the three GPU servers with 4 NVIDIA A100 GPUs has been configured in MIG (Multi-Instance GPU) mode. This provides more but smaller GPU partitions for users who need less GPU memory, with the aim of serving more jobs simultaneously.<\/li>\n<\/ul>\n<p class=\"p4\">Available GPU configurations:<\/p>\n<ul class=\"ul1\">\n<li class=\"li2\">1g.10gb \u2014 up to 10 GB GPU memory, 4 CPU cores and 15.5 GB memory<\/li>\n<li class=\"li2\">2g.20gb \u2014 up to 20 GB GPU memory, 8 CPU cores and 31 GB memory<\/li>\n<li class=\"li2\">3g.40gb \u2014 up to 40 GB GPU memory, 16 CPU cores and 62 GB memory<\/li>\n<li class=\"li2\">a100 \u2014 up to 80 GB GPU memory, 32 CPU cores and 124 GB memory<\/li>\n<\/ul>\n<p class=\"p5\">In addition, the system offers 2.5 PB of storage space via IBM&#8217;s General Parallel File System (GPFS). The infrastructure is completed with an IBM TS4500 tape library with a maximum storage capacity of approximately 6 PB.<\/p>\n<h2 class=\"p1\">Application procedure<\/h2>\n<p class=\"p2\">The application for access to the ARIS system for preparatory projects is available at the following link:<\/p>\n<p class=\"p6\"><div class=\"blox_element_center\"><a class='blox_elem_button blox_animation_animate_before blox_animation_appear blox_elem_button_default blox_elem_color_background blox_elem_border_ blox_elem_button_small light center ' style='background-color:#f7941e;' href='https:\/\/survey.grnet.gr\/index.php?r=survey\/index&amp;sid=748281&amp;lang=en' target='_self'>Online Form<\/a><\/div><\/p>\n<h2 class=\"p1\">Evaluation procedure<\/h2>\n<p class=\"p2\">The evaluation of proposals for preparatory projects includes their technical assessment by GRNET&#8217;s specialized staff. The evaluation results are forwarded to researchers within 10 days from the evaluation date.<\/p>\n<p class=\"p2\">In the event of resource unavailability for project execution, ARIS resources will be allocated on a priority basis.<\/p>\n<h2 class=\"p1\">Contact<\/h2>\n<p class=\"p7\">hpc-access at lists.grnet.gr<\/p>\n","protected":false},"excerpt":{"rendered":"<p>General The National Infrastructures for Research and Technology (GRNET S.A.) provides resources on the ARIS supercomputing system for the execution of preparatory computational projects. The call for preparatory projects on the ARIS system is continuously open for proposals. These proposals are reviewed by GRNET&#8217;s technical committee. Note: The evaluation of proposals is a continuous process [&hellip;]<\/p>\n","protected":false},"author":2,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-1773","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.hpc.grnet.gr\/en\/wp-json\/wp\/v2\/pages\/1773","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.hpc.grnet.gr\/en\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.hpc.grnet.gr\/en\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.hpc.grnet.gr\/en\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.hpc.grnet.gr\/en\/wp-json\/wp\/v2\/comments?post=1773"}],"version-history":[{"count":7,"href":"https:\/\/www.hpc.grnet.gr\/en\/wp-json\/wp\/v2\/pages\/1773\/revisions"}],"predecessor-version":[{"id":5230,"href":"https:\/\/www.hpc.grnet.gr\/en\/wp-json\/wp\/v2\/pages\/1773\/revisions\/5230"}],"wp:attachment":[{"href":"https:\/\/www.hpc.grnet.gr\/en\/wp-json\/wp\/v2\/media?parent=1773"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}