Open App →
Back to News Feed
Nvidia Developer Blog November 10, 2026 neutral

Start and Restore GPU Workloads in Seconds with NVIDIA Snapshot

NVIDIAAI / LLMNVIDIA / GPUSupply Chain
<img width="768" height="432" src="https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-768x432.jpg" class="webfeedsFeaturedVisual wp-post-image" alt="Green cube with a neural-network diagram, surrounded by circular arrows beneath descending data blocks on a purple background." style="display: block; margin-bottom: 5px; clear:both;max-width: 100%;" link_thumbnail="" decoding="async" loading="lazy" srcset="https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-768x432.jpg 768w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-179x101.jpg 179w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-300x169.jpg 300w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-625x352.jpg 625w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-1536x864.jpg 1536w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-645x363.jpg 645w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-660x370.jpg 660w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-500x281.jpg 500w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-160x90.jpg 160w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-362x204.jpg 362w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-196x110.jpg 196w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-1024x576.jpg 1024w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-960x540.jpg 960w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore.webp 1920w" sizes="auto, (max-width: 768px) 100vw, 768px" title="NGC - Promo - Day 0 Nemotron Tech Blog Image - 3326282" />NVIDIA Snapshot, a new open-source component of the NVIDIA DSX AI Factory Platform, makes initialized GPU workloads reusable across inference...<img width="768" height="432" src="https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-768x432.jpg" class="webfeedsFeaturedVisual wp-post-image" alt="Green cube with a neural-network diagram, surrounded by circular arrows beneath descending data blocks on a purple background." style="display: block; margin-bottom: 5px; clear:both;max-width: 100%;" link_thumbnail="" decoding="async" loading="lazy" srcset="https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-768x432.jpg 768w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-179x101.jpg 179w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-300x169.jpg 300w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-625x352.jpg 625w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-1536x864.jpg 1536w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-645x363.jpg 645w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-660x370.jpg 660w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-500x281.jpg 500w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-160x90.jpg 160w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-362x204.jpg 362w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-196x110.jpg 196w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-1024x576.jpg 1024w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore-960x540.jpg 960w, https://developer-blogs.nvidia.com/wp-content/uploads/2026/11/nvidia-snapshot-gpu-checkpoint-restore.webp 1920w" sizes="auto, (max-width: 768px) 100vw, 768px" title="NGC - Promo - Day 0 Nemotron Tech Blog Image - 3326282" /><p>NVIDIA Snapshot, a new open-source component of the NVIDIA DSX AI Factory Platform, makes initialized GPU workloads reusable across inference replicas—speeding startup and recovery. When inference demand spikes or a replica fails, available GPUs don’t translate immediately into serving capacity. Initializing an LLM workload—a process known as a cold start—can take minutes…</p> <p><a href="https://developer.nvidia.com/blog/start-and-restore-gpu-workloads-in-seconds-with-nvidia-snapshot/" rel="nofollow" data-wpel-link="internal" target="_self">Source</a></p>
Read original article ↗

Related Articles

LITEON présente ses innovations en matière d'IA du cloud à la périphérie et la plateforme NVIDIA MGX à l'electronica 2026

MUNICH, 14 novembre 2026 /PRNewswire/ -- LITEON Technology (2301) présente ses capacités technologiques transversales à

PR Newswire · November 14, 2026

Faster Scientific Simulations with New FP64 Algorithms in NVIDIA cuBLAS

Many scientific simulations, from molecular chemistry to materials science, spend the majority of their execution time c

Nvidia Developer Blog · November 13, 2026

Run GPU-Powered Black Hole Interferometry with NVIDIA cuPhoton

Imaging black holes is challenging because they are distant and appear tiny in the sky—for example, M87*’s observed ring

Nvidia Developer Blog · November 13, 2026

Run Native DLA Inference in Python on NVIDIA Jetson AGX Orin Using cuDLA

AI applications at the edge often need to run multiple neural networks within limited compute and power budgets. NVIDIA

Nvidia Developer Blog · November 13, 2026