====== NVIDIA NCP-AIN Training Plan ====== **Certification:** NVIDIA-Certified Professional: AI Networking (NCP-AIN) \\ **Format:** 70-75 multiple-choice questions, 120 minutes total \\ **Cost:** $400 USD, online proctored \\ **Recommended experience:** 2–3 years of operational experience working in a data center with NVIDIA hardware solutions. The candidate should be able to deploy and manage NVIDIA AI networking infrastructure in support of AI workloads. \\ **Plan length:** ~6 weeks \\ **Exam Overview Link:** https://www.nvidia.com/en-us/learn/certification/ai-networking-professional/ ^ **Topics Areas** ^ **% of Exam** ^ **Topics Covered** ^ | AI Data Center Design and Optimization | 5% | * Describe an AI factory networking architecture and its components (e.g., GPUs, NVIDIA BlueField, scalable units, switches).\\ * Describe rail-optimized topologies for high-performance AI workloads.\\ * Describe GPU-to-GPU communications. | | NVIDIA Spectrum Networking | 30% | * Configure NVIDIA Spectrum-X switches for RDMA over Converged Ethernet (RoCE) to enable high-speed, low-latency communication.\\ * Enable and verify QoS, explicit congestion notification (ECN), and priority flow control (PFC), advanced features like adaptive routing, and telemetry.\\ * Configure multi-tenancy Border Gateway Protocol Ethernet VPN (BGP-EVPN) to isolate tenant workloads.\\ * Use NVIDIA Air to simulate network environments and identify potential issues.\\ * Diagnose congestion or packet loss using in-band telemetry and NVIDIA What Just Happened® (WJH) services.\\ * Use NetQ for real-time network monitoring, including congestion detection and latency measurements.\\ * Install NVIDIA DOCA.\\ * Configure NVIDIA SuperNIC™ functionality for advanced packet processing and congestion control. | | NVIDIA InfiniBand Networking | 30% | * Perform initial configuration and provisioning, including high availability (HA).\\ * Configure partition keys (PKeys) to ensure secure multi-tenancy in InfiniBand networks.\\ * Configure QoS and adaptive routing to dynamically adjust paths based on congestion.\\ * Use UFM to monitor InfiniBand link status and bandwidth utilization. | | Kubernetes Integration | 5% | * Deploy the NVIDIA Network Operator to manage RDMA interfaces and InfiniBand networks within Kubernetes clusters.\\ * Verify NVIDIA Network Operator functionality.\\ | | Troubleshooting Tools | 20% | * Use tools like cl-resource-query to check resource allocation in Spectrum-X environments.\\ * Use WJH services for real-time event analysis.\\ * Verify low-latency interconnects between GPUs, CPUs, and storage systems.\\ * Use UFM system health to diagnose InfiniBand issues.\\ * Use commands like ib_write_lat, ib_write_bw, ibping, ibstat, ibdiagnet, ibnodes, and iblinkinfo to diagnose connectivity issues. | | Automation and Configuration | 10% | * Manage Spectrum-X switch configurations through NVUE templates.\\ * Write Ansible playbooks to automate network setup tasks like VLAN creation or RoCE configuration.| ---- This exam does not have a lab section like the NCP-AIO, and the questions are generally a bit easier. So while a deep understanding of NVIDIA Networking concepts will still be required to pass this exam, it will not necessarily require the depth of knowledge and same level of hands-on experience that was needed for the NCP-AIO. \\ \\ ===== Template Stories ===== **Link to NCP-AIN Jira Epic:** https://siriuscloudservices.atlassian.net/browse/AITB-3738 \\ Below are the links to the individual template stories that I used to pass the exam. While I was able to pass only while only using these materials, there were many questions based on commands and more specific knowledge than this material tends to cover. So if you have any doubts about knowledge gaps when studying this material, you are encouraged to utilize tools like NVIDIA Air or creating labs in the cloud to help gain more hands-on experience. \\ \\ **NVIDIA Courses:**\\ * [[https://siriuscloudservices.atlassian.net/browse/AITB-4085?atlOrigin=eyJpIjoiMTFlN2Q3ZjdiYjUxNGM3ZGExODlkNmRiYzkyNjcxYWQiLCJwIjoiaiJ9|Cumulus Linux Essentials + InfiniBand Essentials Courses]] * [[https://siriuscloudservices.atlassian.net/browse/AITB-4086?atlOrigin=eyJpIjoiZTc1ZjVhZTgxN2ZmNDM1Mzk0NTk3NzcwNGQ4ODkyOGUiLCJwIjoiaiJ9|InfiniBand Network Administration Course]] * [[https://siriuscloudservices.atlassian.net/browse/AITB-4090?atlOrigin=eyJpIjoiZjA0YTBlNTA5ZmI1NGI3ZDhkZmJjYTdmODc5NWRhNzIiLCJwIjoiaiJ9|Cumulus Linux Administration Course]] **NVIDIA Study Guide:**\\ * [[https://siriuscloudservices.atlassian.net/browse/AITB-4088?atlOrigin=eyJpIjoiMWJhYjNiZjBiYWEzNDE5OGFmM2QyYTA1NTljYTcyZjUiLCJwIjoiaiJ9|NCP-AIN NVIDIA Study Guide Part 1]] * [[https://siriuscloudservices.atlassian.net/browse/AITB-4089?atlOrigin=eyJpIjoiZTZiZDkwMjA2OGJkNGIyMTkyZjk4NDE0Njg5OWNmMTkiLCJwIjoiaiJ9|NCP-AIN NVIDIA Study Guide Part 2]] **Core Topics and Troubleshooting Scenarios:**\\ * [[https://siriuscloudservices.atlassian.net/browse/AITB-4091?atlOrigin=eyJpIjoiZTI0NzJkYTllY2YzNDU1M2IxOTJlYjkwZDMxMjdkZWIiLCJwIjoiaiJ9|NCP-AIN Core Topics Review]] * For this story, I copy/pasted the story description into Claude and had it quiz me on each topic until I knew it by heart. I started with having Claude tell me the answers for each service, then I would ask it to just prompt me with the topic and the questions and I would answer by memory. * [[https://siriuscloudservices.atlassian.net/browse/AITB-4092?atlOrigin=eyJpIjoiNWMyMWMyOGRjNmY0NDlhZjkxMzYzNmIyOWQ1MzViZjciLCJwIjoiaiJ9|NCP-AIN Troubleshooting Scenarios]] * For this story, I also copy/pasted the description into Claude, but made sure to do so in the same conversation that I used for the previous story so that it would keep the knowledge and focus of the core topics. Then I used it as interactive troubleshooting practice where I would think through possible answers to the questions in the description, and Claude would nudge me towards the correct answers if I was off base. I found this and the previous story to provide some of the best learning throughout my exam preparation. **Practice Exams:**\\ * [[https://siriuscloudservices.atlassian.net/browse/AITB-4087?atlOrigin=eyJpIjoiMzZmYTRlYjBiOWEwNGU2ZWI2ZTQzYTAxNjUwNDFiZmMiLCJwIjoiaiJ9|NCP-AIN Practice Exams]] * These Udemy practice exams were generally much easier than actual exam. They usually have one obvious answer and three other choices that do not make any sense in the scenario, or the question very obviously points towards the correct answer by using terms that are found in the answer. I also found that the explanations for the right and wrong answers were not very helpful. Very rarely does it actually describe the service and why it's correct. The "explanations" usually boil down to saying the answer is correct because it's correct, and the others are wrong because they're wrong. Overall, I felt like these exams did help me prepare a bit and overall improved my score on the exam, but it might also be worth trying other practice exams in case they are more helpful. \\ \\ \\ [[home-page|AI Cloud Managed Services Policies and Procedures]]