CoreWeave: US-EAST-04A Data Center
About US-EAST-04A Data Center
CoreWeave US-EAST-04A Data Center in Virginia is a CoreWeave Availability Zone in the US-EAST super region for GPU cloud, AI infrastructure, CPU compute, and AI Object Storage workloads. The zone is listed as General Access, making it available through CoreWeave’s platform rather than reserved only for select customers. US-EAST-04A supports organizations deploying model training, inference, rendering, high-performance computing, AI pipelines, and cloud-native applications in the Eastern United States.
The zone supports H200 InfiniBand, H100 InfiniBand, RTX Pro 6000 Blackwell Server Edition, L40S, L40, GH200, and A100 GPU instances. CPU options include High Performance AMD Genoa, High Core AMD Genoa, and High Core AMD Genoa with InfiniBand. CoreWeave also lists AI Object Storage for US-EAST-04A, supporting data-intensive AI workflows close to the compute environment. Network ranges include public IP space, NAT egress ranges, node ranges, pod ranges, service ranges, and load balancer ranges for workload deployment and interoperability.
The Virginia placement gives US-EAST-04A an East Coast infrastructure position within CoreWeave’s US-EAST footprint, which includes zones in Virginia, New Jersey, New York, Pennsylvania, Georgia, Ohio, Massachusetts, North Carolina, and Michigan. The zone is built for cloud infrastructure consumption rather than traditional retail colocation, with service access delivered through CoreWeave’s platform regions, instance types, Kubernetes infrastructure, and storage services.
🏢 Facility Highlights
- Region: US-EAST-04
- Availability Zone: US-EAST-04A
- Location: Virginia, USA
- Access Type: General Access
- AI Object Storage: Available
- GPU Instances: H200, H100, RTX Pro 6000 Blackwell Server Edition, L40S, L40, GH200, and A100
- CPU Instances: High Performance AMD Genoa, High Core AMD Genoa, and High Core AMD Genoa with InfiniBand
🧠 AI Infrastructure
- Training GPUs: H200 InfiniBand, H100 InfiniBand, and A100 instances
- Inference GPUs: RTX Pro 6000 Blackwell Server Edition, L40S, L40, and GH200 options
- Networking: InfiniBand instance options for high-performance distributed workloads
- Storage: AI Object Storage available in the zone
🌐 Connectivity & Network Information
- Public IP Ranges: 166.19.10.0/23 and 166.19.16.0/21
- NAT Egress: 166.19.10.192/26 and 166.19.19.192/26
- Node Range: 10.176.192.0/18
- Workload Networking: Pod, service, and load balancer private address ranges listed in CoreWeave documentation
🏭 Who It Serves
- AI Training: Distributed GPU workloads using H200, H100, and A100 infrastructure
- AI Inference: Production inference services using RTX Pro 6000, L40S, L40, and GH200 instances
- HPC: Compute-intensive simulations, rendering, and accelerated workloads
- Cloud-Native Applications: CPU and GPU services deployed through CoreWeave’s platform