Lenovo ThinkSystem SR670 V2 with NVIDIA A100 PCIe 80GB GPU (Legacy)
Brand: Lenovo | Category: GPUs
SKU: 7D9DA00LNA | Part #: 7D9DA00LNA | MPN: 7D9DA00LNA
Contact for Pricing — Request a Quote
About the Lenovo ThinkSystem SR670 V2 with NVIDIA A100 PCIe 80GB GPU (Legacy)
At 19.5 TFLOPS FP32, 156 TFLOPS TF32, and up to 312 TFLOPS with sparsity enabled, the Lenovo ThinkSystem SR670 V2 with NVIDIA A100 PCIe 80GB GPU delivers enterprise-class AI and HPC acceleration built on the proven NVIDIA Ampere architecture. This 2U rack server pairs dual Intel Xeon Scalable (3rd Gen, Ice Lake) processors with up to 10 NVIDIA A100 PCIe GPUs per system, making it ideal for organizations scaling machine learning training, inference workloads, and scientific computing at datacenter scale. Each GPU provides 80 GB of HBM2e memory with 600 GB/s bandwidth—a combination that minimizes data movement bottlenecks and accelerates complex neural networks across multiple precision formats.
Built for demanding multi-tenant and high-utilization environments, the system supports Multi-Instance GPU (MIG) technology, allowing each A100 to be partitioned into up to 7 independent instances for workload isolation and improved resource efficiency. Lenovo backs the SR670 V2 with enterprise-grade systems management via the Lenovo XClarity Controller (XCC2) and XClarity Administrator, redundant hot-swap Platinum-efficiency power supplies, and broad OS support spanning Red Hat Enterprise Linux, SUSE Linux Enterprise Server, Ubuntu, VMware vSphere, and Windows Server. For IT procurement teams and AI infrastructure architects evaluating GPU-accelerated servers for production deployment, the ThinkSystem SR670 V2 combines density, reliability, and performance in a single 2U form factor. Contact Omnixon Global via RFQ to discuss configuration options, pricing, and availability for part number 7D9DA00LNA.
Key Specifications
- Manufacturer: Lenovo
- Manufacturer Part Number: 7D9DA00LNA
- Form Factor: 2U Rack Server
- GPU Model: NVIDIA A100 PCIe 80GB
- GPU Memory: 80 GB HBM2e with 600 GB/s bandwidth
- Processor Platform: Intel Xeon Scalable (3rd Gen, Ice Lake) with 2 sockets
- Maximum GPU Capacity: Up to 10 NVIDIA A100 PCIe GPUs per server
- Multi-Instance GPU (MIG): Up to 7 instances per GPU
Technical Specifications
| Brand | Lenovo |
| Category | GPUs |
| SKU | 7D9DA00LNA |
| Part Number | 7D9DA00LNA |
| Condition | New |
| Manufacturer Part Number | 7D9DA00LNA |
| Product Family | ThinkSystem SR670 V2 |
| Form Factor | 2U Rack Server |
| GPU Model | NVIDIA A100 PCIe 80GB |
| GPU Architecture | NVIDIA Ampere |
| GPU Memory | 80 GB HBM2e |
| GPU Memory Bandwidth | 600 GB/s (HBM2e) |
| GPU Tensor Core Generation | 3rd Generation (TF32, FP64, FP32, FP16, BF16, INT8) |
| GPU FP64 Peak Performance | 9.7 TFLOPS |
| GPU FP32 Peak Performance | 19.5 TFLOPS |
| GPU TF32 Tensor Performance | 156 TFLOPS (sparsity: 312 TFLOPS) |
| Multi-Instance GPU (MIG) | Up to 7 MIG instances per GPU |
| Maximum GPU Capacity | Up to 10 NVIDIA A100 PCIe GPUs per server |
| Processor Platform | Intel Xeon Scalable (3rd Gen, Ice Lake) |
| Processor Sockets | 2 |
| PCIe Generation | PCIe Gen 4 |
| Systems Management | Lenovo XClarity Controller (XCC2), Lenovo XClarity Administrator |
| Power Supply | Redundant hot-swap (Platinum efficiency) |
| Operating System Support | Red Hat Enterprise Linux, SUSE Linux Enterprise Server, Ubuntu, VMware vSphere, Windows Server |
Frequently Asked Questions about Lenovo ThinkSystem SR670 V2 with NVIDIA A100 PCIe 80GB GPU (Legacy)
What server platforms accept the Lenovo ThinkSystem SR670 V2 with NVIDIA A100 PCIe 80GB GPU (Legacy)?
Reference servers include Dell PowerEdge XE9680 / XE9712, HPE Cray XD670, Lenovo ThinkSystem SR685a / SR675 V3, Supermicro AS-A21GE / SYS-821GE, Gigabyte G593 / G894, ASUS ESC. Share your target platform in the RFQ and we will confirm chassis-to-GPU compatibility and recommended NIC pairing.
How long is the lead time on AI GPUs?
Highly model-dependent. L40S / RTX-class: typically 3-6 weeks. H100/H200/B200 in SXM form factor: 12-16 weeks for whole-platform allocations. We quote genuine-channel ETAs only — no grey-market promises.
Do you supply matched networking (Quantum InfiniBand / Spectrum-X)?
Yes — Omnixon stocks the full NVIDIA networking lineup (Quantum-2 / Quantum-X InfiniBand, Spectrum-X Ethernet, ConnectX NICs, BlueField DPUs) so we can quote a complete training-cluster BOM, not just the GPUs.
Can you help with NVIDIA AI Enterprise licensing?
Yes. We hold genuine-channels for NVIDIA AI Enterprise software subscriptions. Add it to your RFQ and we quote node-aligned licensing along with the hardware.