{"id":83,"date":"2020-01-22T19:19:43","date_gmt":"2020-01-22T19:19:43","guid":{"rendered":"https:\/\/ni.cmu.edu\/computing\/?post_type=ht_kb&#038;p=83"},"modified":"2026-07-06T18:47:36","modified_gmt":"2026-07-06T18:47:36","slug":"mind-cluster-nodes","status":"publish","type":"ht_kb","link":"https:\/\/ni.cmu.edu\/computing\/knowledge-base\/mind-cluster-nodes\/","title":{"rendered":"Mind Cluster Nodes"},"content":{"rendered":"<style>\n    \/* Clean, highly responsive system for the MIND Cluster Specifications page *\/\n    .kb-spec-container {\n        font-family: -apple-system, BlinkMacSystemFont, \"Segoe UI\", Roboto, Helvetica, Arial, sans-serif;\n        color: #2d3748;\n        line-height: 1.6;\n        max-width: 100%;\n        margin: 0 auto;\n        padding: 0 4px;\n    }<\/p>\n<p>    \/* Clean dropdown elements matching the KB dashboard styling *\/\n    .kb-spec-container details {\n        background: #ffffff;\n        border: 1px solid #e2e8f0;\n        border-radius: 6px;\n        padding: 14px 16px;\n        margin-bottom: 12px;\n        box-shadow: 0 1px 2px rgba(0,0,0,0.02);\n    }\n    .kb-spec-container details summary {\n        font-weight: 600;\n        cursor: pointer;\n        color: #4a5568;\n        outline: none;\n        user-select: none;\n    }\n    .kb-spec-container details[open] summary {\n        margin-bottom: 14px;\n        border-bottom: 1px solid #edf2f7;\n        padding-bottom: 10px;\n        color: #1a202c;\n    }<\/p>\n<p>    \/* RESPONSIVE TABLE FIX: Handles overflow by offering crisp horizontal scroll without breaking page flow *\/\n    .kb-responsive-table-wrapper {\n        width: 100%;\n        overflow-x: auto;\n        -webkit-overflow-scrolling: touch; \/* Smooth mobile momentum scrolling *\/\n        margin: 14px 0;\n        border: 1px solid #e2e8f0;\n        border-radius: 6px;\n        background: #ffffff;\n    }\n    .kb-responsive-table {\n        width: 100%;\n        border-collapse: collapse;\n        text-align: left;\n        font-size: 0.88rem;\n    }<\/p>\n<p>    \/* Table Header Layout *\/\n    .kb-responsive-table th {\n        background-color: #f7fafc;\n        color: #4a5568;\n        font-weight: 600;\n        padding: 12px 14px;\n        border-bottom: 2px solid #e2e8f0;\n        white-space: nowrap; \/* Prevents text splitting across headers *\/\n    }<\/p>\n<p>    \/* Strict Column Control to enforce readable horizontal lines *\/\n    .kb-responsive-table td {\n        padding: 12px 14px;\n        border-bottom: 1px solid #edf2f7;\n        color: #4a5568;\n        white-space: nowrap; \n    }<\/p>\n<p>    .kb-responsive-table tr:last-child td {\n        border-bottom: none;\n    }\n    \/* Subtle zebra contrast striping *\/\n    .kb-responsive-table tr:nth-child(even) td {\n        background-color: #fcfdfd;\n    }\n    \/* Highlight focal row on cursor hover *\/\n    .kb-responsive-table tr:hover td {\n        background-color: #f7fafc;\n    }<\/p>\n<p>    \/* List & Formatting adjustments *\/\n    .kb-spec-container ul {\n        padding-left: 20px;\n        margin: 8px 0;\n    }\n    .kb-spec-container li {\n        margin-bottom: 6px;\n    }\n    .node-monospace-tag {\n        font-family: SFMono-Regular, Consolas, \"Liberation Mono\", Menlo, monospace;\n        font-weight: 600;\n        color: #2d3748;\n        background-color: #f1f5f9;\n        padding: 2px 5px;\n        border-radius: 4px;\n        font-size: 0.85rem;\n    }\n<\/style>\n<div class=\"kb-spec-container\">\n<p>The MIND Cluster is a heterogeneous high-performance computing environment supporting computational neuroscience, neuroimaging, artificial intelligence, and data-intensive research. It includes CPU-only and GPU-accelerated compute nodes spanning multiple hardware generations. All nodes are managed via SLURM and connected through a 100 Gbps InfiniBand network.<\/p>\n<details>\n<summary>Cluster Overview<\/summary>\n<ul>\n<li><strong>Total compute nodes:<\/strong> 14<\/li>\n<li><strong>GPU-enabled nodes:<\/strong> 11<\/li>\n<li><strong>CPU-only nodes:<\/strong> 3<\/li>\n<li><strong>Total logical CPU cores:<\/strong> 1,056<\/li>\n<li><strong>Total system memory:<\/strong> 7.38 TB<\/li>\n<li><strong>Total GPUs:<\/strong> 64<\/li>\n<\/ul>\n<\/details>\n<details open>\n<summary>Node Specification Table<\/summary>\n<div class=\"kb-responsive-table-wrapper\">\n<table class=\"kb-responsive-table\">\n<thead>\n<tr>\n<th>Node<\/th>\n<th>CPU Model<\/th>\n<th>CPU Year<\/th>\n<th>Logical Cores<\/th>\n<th>RAM (GB)<\/th>\n<th>GPU Model<\/th>\n<th>GPU Count<\/th>\n<th>GPU Memory (GB)<\/th>\n<th>GPU Year<\/th>\n<th>Architecture<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td><span class=\"node-monospace-tag\">mind-0-15<\/span><\/td>\n<td>AMD EPYC 7443<\/td>\n<td>2021<\/td>\n<td>96<\/td>\n<td>1007.4<\/td>\n<td>None<\/td>\n<td>0<\/td>\n<td>\u2014<\/td>\n<td>\u2014<\/td>\n<td>CPU-only<\/td>\n<\/tr>\n<tr>\n<td><span class=\"node-monospace-tag\">mind-0-16<\/span><\/td>\n<td>AMD EPYC 7443<\/td>\n<td>2021<\/td>\n<td>96<\/td>\n<td>1007.4<\/td>\n<td>None<\/td>\n<td>0<\/td>\n<td>\u2014<\/td>\n<td>\u2014<\/td>\n<td>CPU-only<\/td>\n<\/tr>\n<tr>\n<td><span class=\"node-monospace-tag\">mind-0-17<\/span><\/td>\n<td>AMD EPYC 7443<\/td>\n<td>2021<\/td>\n<td>96<\/td>\n<td>1007.4<\/td>\n<td>None<\/td>\n<td>0<\/td>\n<td>\u2014<\/td>\n<td>\u2014<\/td>\n<td>CPU-only<\/td>\n<\/tr>\n<tr>\n<td><span class=\"node-monospace-tag\">mind-0-18<\/span><\/td>\n<td>Intel Xeon E5-2620 v4<\/td>\n<td>2016<\/td>\n<td>32<\/td>\n<td>125.6<\/td>\n<td>RTX 2080 Ti<\/td>\n<td>4<\/td>\n<td>11<\/td>\n<td>2018<\/td>\n<td>Turing<\/td>\n<\/tr>\n<tr>\n<td><span class=\"node-monospace-tag\">mind-0-20<\/span><\/td>\n<td>Intel Xeon E5-2620 v4<\/td>\n<td>2016<\/td>\n<td>32<\/td>\n<td>125.6<\/td>\n<td>RTX 2080 Ti<\/td>\n<td>4<\/td>\n<td>11<\/td>\n<td>2018<\/td>\n<td>Turing<\/td>\n<\/tr>\n<tr>\n<td><span class=\"node-monospace-tag\">mind-0-22<\/span><\/td>\n<td>Intel Xeon E5-2620 v4<\/td>\n<td>2016<\/td>\n<td>32<\/td>\n<td>125.6<\/td>\n<td>RTX 2080 Ti<\/td>\n<td>4<\/td>\n<td>11<\/td>\n<td>2018<\/td>\n<td>Turing<\/td>\n<\/tr>\n<tr>\n<td><span class=\"node-monospace-tag\">mind-0-24<\/span><\/td>\n<td>Intel Xeon E5-2620 v4<\/td>\n<td>2016<\/td>\n<td>32<\/td>\n<td>125.6<\/td>\n<td>RTX 2080 Ti<\/td>\n<td>4<\/td>\n<td>11<\/td>\n<td>2018<\/td>\n<td>Turing<\/td>\n<\/tr>\n<tr>\n<td><span class=\"node-monospace-tag\">mind-0-26<\/span><\/td>\n<td>Intel Xeon Silver 4208<\/td>\n<td>2019<\/td>\n<td>32<\/td>\n<td>125.3<\/td>\n<td>RTX 3090<\/td>\n<td>2<\/td>\n<td>24<\/td>\n<td>2020<\/td>\n<td>Ampere<\/td>\n<\/tr>\n<tr>\n<td><span class=\"node-monospace-tag\">mind-0-28<\/span><\/td>\n<td>Intel Xeon Silver 4310<\/td>\n<td>2021<\/td>\n<td>48<\/td>\n<td>250.6<\/td>\n<td>RTX A5000<\/td>\n<td>4<\/td>\n<td>24<\/td>\n<td>2021<\/td>\n<td>Ampere<\/td>\n<\/tr>\n<tr>\n<td><span class=\"node-monospace-tag\">mind-1-11<\/span><\/td>\n<td>Intel Xeon E5-2620 v4<\/td>\n<td>2016<\/td>\n<td>32<\/td>\n<td>125.6<\/td>\n<td>TITAN Xp<\/td>\n<td>4<\/td>\n<td>12<\/td>\n<td>2017<\/td>\n<td>Pascal<\/td>\n<\/tr>\n<tr>\n<td><span class=\"node-monospace-tag\">mind-1-15<\/span><\/td>\n<td>AMD EPYC 7513<\/td>\n<td>2021<\/td>\n<td>128<\/td>\n<td>1007.7<\/td>\n<td>L40S<\/td>\n<td>8<\/td>\n<td>48<\/td>\n<td>2023<\/td>\n<td>Ada Lovelace<\/td>\n<\/tr>\n<tr>\n<td><span class=\"node-monospace-tag\">mind-1-19<\/span><\/td>\n<td>AMD EPYC 9554<\/td>\n<td>2022<\/td>\n<td>128<\/td>\n<td>1133.4<\/td>\n<td>RTX PRO 6000<\/td>\n<td>10<\/td>\n<td>96<\/td>\n<td>2025<\/td>\n<td>Blackwell<\/td>\n<\/tr>\n<tr>\n<td><span class=\"node-monospace-tag\">mind-1-24<\/span><\/td>\n<td>Intel Xeon E5-2687W v4<\/td>\n<td>2016<\/td>\n<td>48<\/td>\n<td>251.7<\/td>\n<td>TITAN RTX<\/td>\n<td>10<\/td>\n<td>24<\/td>\n<td>2018<\/td>\n<td>Turing<\/td>\n<\/tr>\n<tr>\n<td><span class=\"node-monospace-tag\">mind-1-29<\/span><\/td>\n<td>AMD EPYC 9554<\/td>\n<td>2022<\/td>\n<td>224<\/td>\n<td>1133.4<\/td>\n<td>L40S<\/td>\n<td>10<\/td>\n<td>48<\/td>\n<td>2023<\/td>\n<td>Ada Lovelace<\/td>\n<\/tr>\n<\/tbody>\n<\/table><\/div>\n<\/details>\n<details>\n<summary>CPU-Only Nodes<\/summary>\n<ul>\n<li><strong>mind-0-15 \/ 0-16 \/ 0-17<\/strong>\n<ul>\n<li>CPU: AMD EPYC 7443<\/li>\n<li>Cores: 48 physical \/ 96 logical<\/li>\n<li>RAM: ~1 TB<\/li>\n<li>GPU: None<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<p><strong>Use cases:<\/strong> Preprocessing, statistical modeling, memory-intensive CPU workloads, neuroimaging pipelines.<\/p>\n<\/details>\n<details>\n<summary>GPU Architecture Summary<\/summary>\n<ul>\n<li><strong>Blackwell (2025):<\/strong> RTX PRO 6000 (96 GB)<\/li>\n<li><strong>Ada Lovelace (2023):<\/strong> L40S (48 GB)<\/li>\n<li><strong>Ampere (2020\u20132021):<\/strong> RTX 3090, RTX A5000<\/li>\n<li><strong>Turing (2018):<\/strong> RTX 2080 Ti, TITAN RTX<\/li>\n<li><strong>Pascal (2017):<\/strong> TITAN Xp<\/li>\n<\/ul>\n<\/details>\n<details>\n<summary>Usage Guidance<\/summary>\n<ul>\n<li><strong>CPU nodes:<\/strong> Preprocessing, statistical analysis, memory-heavy workloads.<\/li>\n<li><strong>RTX 2080 Ti \/ TITAN Xp:<\/strong> Legacy pipelines, imaging, lightweight GPU tasks.<\/li>\n<li><strong>RTX 3090 \/ A5000:<\/strong> General-purpose GPU computing and ML development.<\/li>\n<li><strong>TITAN RTX:<\/strong> Balanced GPU research workloads.<\/li>\n<li><strong>L40S:<\/strong> Large-scale deep learning training and inference.<\/li>\n<li><strong>Blackwell:<\/strong> Next-generation AI and high-memory model training.<\/li>\n<\/ul>\n<\/details>\n<details>\n<summary>Notes<\/summary>\n<ul>\n<li>Compute totals include compute nodes only.<\/li>\n<li>Login, storage, and backup systems are not included.<\/li>\n<li>All nodes are scheduled via SLURM.<\/li>\n<li>All systems connected via 100 Gbps InfiniBand.<\/li>\n<li>Storage systems are separate from compute infrastructure.<\/li>\n<\/ul>\n<\/details>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>The MIND Cluster is a heterogeneous high-performance computing environment supporting computational neuroscience, neuroimaging, artificial intelligence, and data-intensive research. It includes CPU-only and GPU-accelerated compute nodes spanning multiple hardware generations. All nodes are managed via SLURM and connected through a 100 Gbps InfiniBand network. Cluster Overview Total compute nodes: 14 GPU-enabled&#8230;<\/p>\n","protected":false},"author":1,"comment_status":"closed","ping_status":"closed","template":"","format":"standard","meta":{"footnotes":""},"ht-kb-category":[7],"ht-kb-tag":[],"class_list":["post-83","ht_kb","type-ht_kb","status-publish","format-standard","hentry","ht_kb_category-cluster"],"jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/ni.cmu.edu\/computing\/wp-json\/wp\/v2\/ht-kb\/83","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/ni.cmu.edu\/computing\/wp-json\/wp\/v2\/ht-kb"}],"about":[{"href":"https:\/\/ni.cmu.edu\/computing\/wp-json\/wp\/v2\/types\/ht_kb"}],"author":[{"embeddable":true,"href":"https:\/\/ni.cmu.edu\/computing\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/ni.cmu.edu\/computing\/wp-json\/wp\/v2\/comments?post=83"}],"version-history":[{"count":48,"href":"https:\/\/ni.cmu.edu\/computing\/wp-json\/wp\/v2\/ht-kb\/83\/revisions"}],"predecessor-version":[{"id":741,"href":"https:\/\/ni.cmu.edu\/computing\/wp-json\/wp\/v2\/ht-kb\/83\/revisions\/741"}],"wp:attachment":[{"href":"https:\/\/ni.cmu.edu\/computing\/wp-json\/wp\/v2\/media?parent=83"}],"wp:term":[{"taxonomy":"ht_kb_category","embeddable":true,"href":"https:\/\/ni.cmu.edu\/computing\/wp-json\/wp\/v2\/ht-kb-category?post=83"},{"taxonomy":"ht_kb_tag","embeddable":true,"href":"https:\/\/ni.cmu.edu\/computing\/wp-json\/wp\/v2\/ht-kb-tag?post=83"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}