https://gke-ai-labs.dev/tags/mig/ Recent content in MIG on GKE AI Labs Hugo en-us https://gke-ai-labs.dev/docs/tutorials/dynamic-resource-allocation/gpu-dynamic-mig/ Mon, 01 Jan 0001 00:00:00 +0000 https://gke-ai-labs.dev/docs/tutorials/dynamic-resource-allocation/gpu-dynamic-mig/ <h2 id="background"><strong>Background</strong><a class="td-heading-self-link" onclick="copyHeadingLink(this, 'background'); event.preventDefault();" onmousedown="this.classList.add('active-link');" onmouseup="this.classList.remove('active-link');" onmouseleave="this.classList.remove('active-link');" href="#background" aria-label="Heading self-link" title="Copy link to this section: Background"><i class="fa-solid fa-link"></i></a> <style> .td-heading-self-link { transition: background 0.2s; border-radius: 4px; padding: 2px; } .td-heading-self-link.active-link { background: #e0e7ef; transition: background 0.2s; } </style> <script> function copyHeadingLink(anchor, anchorId) { navigator.clipboard.writeText(location.origin + location.pathname + '#' + anchorId); let popup = document.getElementById('global-copy-popup'); if (!popup) { popup = document.createElement('span'); popup.id = 'global-copy-popup'; popup.className = 'copy-popup'; popup.textContent = 'Copied to clipboard'; popup.setAttribute('style', 'position:fixed; bottom:24px; left:24px; background:#222; color:#fff; padding:8px 18px; border-radius:6px; font-size:0.8em; white-space:nowrap; z-index:9999; box-shadow:0 2px 8px rgba(0,0,0,0.15); transform:translateY(100px); opacity:0; transition:transform 0.3s cubic-bezier(.4,0,.2,1), opacity 0.3s cubic-bezier(.4,0,.2,1);'); document.body.appendChild(popup); } popup.style.display = 'block'; void popup.offsetWidth; popup.style.transform = 'translateY(0)'; popup.style.opacity = '1'; clearTimeout(window.__copyPopupTimeout); window.__copyPopupTimeout = setTimeout(() => { popup.style.transform = 'translateY(100px)'; popup.style.opacity = '0'; setTimeout(() => { popup.style.display = 'none'; }, 300); }, 1200); } </script></h2> <p>This tutorial guides you through how to do device sharing of NVIDIA GPUs with Dynamic Resource Allocation on Google Kubernetes Engine (GKE). This guide covers Multi-Instance GPU (MIG), with separate guides covering the other device sharing modes, time slicing and Multi-Process Service (MPS).</p>