https://gke-ai-labs.dev/docs/tutorials/dynamic-resource-allocation/ Recent content in Dynamic Resource Allocation on GKE AI Labs Hugo en-us https://gke-ai-labs.dev/docs/tutorials/dynamic-resource-allocation/gpu-time-slicing/ Mon, 01 Jan 0001 00:00:00 +0000 https://gke-ai-labs.dev/docs/tutorials/dynamic-resource-allocation/gpu-time-slicing/ <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 time slicing, with separate guides covering the other device sharing modes, MIG and Multi-Process Service (MPS).</p> 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>