Lenovo ThinkSystem SR650 V2 with NVIDIA A30 24GB PCIe GPU (Legacy)
Brand: Lenovo | Category: GPUs
SKU: 4X97A68625 | Part #: 4X97A68625 | MPN: 4X97A68625
Contact for Pricing — Request a Quote
About the Lenovo ThinkSystem SR650 V2 with NVIDIA A30 24GB PCIe GPU (Legacy)
PCIe Gen4 x16 interface and 165 W TDP determine server chassis and power supply compatibility—critical specifications for any data center infrastructure team evaluating GPU acceleration hardware. The Lenovo ThinkSystem SR650 V2 with NVIDIA A30 24GB PCIe GPU (part number 4X97A68625) integrates enterprise-grade tensor acceleration into a purpose-built rack server platform. This Lenovo solution pairs the NVIDIA A30 Tensor Core GPU—built on NVIDIA Ampere architecture—with Intel Xeon Scalable 3rd Generation (Ice Lake) processors, delivering 24 GB of HBM2 memory and 933 GB/s memory bandwidth for demanding AI, machine learning, and high-performance compute workloads.
The GPU accelerator supports mixed-precision inference and training across multiple numerical formats: 10.3 TFLOPS for both FP64 and FP32, 165 TFLOPS for FP16 Tensor Core operations, and 330 TOPS for INT8 workloads. Multi-Instance GPU (MIG) technology enables partitioning into up to 4 independent GPU instances using the MIG 4g.6gb profile, maximizing resource utilization across concurrent applications. Lenovo XClarity Administrator management integration simplifies deployment and monitoring. The server supports Windows Server, Red Hat Enterprise Linux, Ubuntu, and VMware vSphere operating systems per the SR650 V2 compatibility matrix. IT procurement teams and AI infrastructure architects seeking a production-ready, fully integrated GPU acceleration solution should evaluate this legacy platform for compatibility with existing deployments.
Enterprise Deployment Scenarios
- Machine learning inference serving with mixed-precision model optimization across FP32 and INT8 pipelines
- High-performance computing clusters requiring 24 GB GPU memory per node and 933 GB/s bandwidth
- Virtual GPU workloads leveraging MIG technology to partition resources among up to 4 independent instances
- Data center consolidation projects integrating GPU acceleration into existing Intel Xeon Scalable infrastructure
- Enterprise analytics and scientific computing requiring PCIe Gen4 x16 GPU connectivity within a managed XClarity environment
Contact Omnixon Global to request an RFQ for part number 4X97A68625.
Technical Specifications
| Brand | Lenovo |
| Category | GPUs |
| SKU | 4X97A68625 |
| Part Number | 4X97A68625 |
| Condition | New |
| Manufacturer Part Number | 4X97A68625 |
| Product Name | ThinkSystem SR650 V2 with NVIDIA A30 24GB PCIe GPU (Legacy) |
| GPU Model | NVIDIA A30 Tensor Core GPU |
| GPU Architecture | NVIDIA Ampere |
| GPU Memory | 24 GB HBM2 |
| Memory Bandwidth | 933 GB/s |
| FP64 Performance | 10.3 TFLOPS |
| FP32 Performance | 10.3 TFLOPS |
| FP16 (Tensor Core) Performance | 165 TFLOPS |
| INT8 Performance | 330 TOPS |
| Interface | PCIe Gen4 x16 |
| MIG Support | Yes – up to 4 GPU instances (MIG 4g.6gb profile) |
| Form Factor | Rack server integrated GPU option (PCIe add-in card) |
| Server Platform | Lenovo ThinkSystem SR650 V2 |
| Server Processor Support | Intel Xeon Scalable 3rd Generation (Ice Lake) |
| Thermal Design Power (TDP) | 165 W |
| Management | Lenovo XClarity Administrator compatible |
| Operating System Support | Windows Server, Red Hat Enterprise Linux, Ubuntu, VMware vSphere (per SR650 V2 compatibility matrix) |
| Product Category | GPU Accelerator – Enterprise / Data Center |
Frequently Asked Questions about Lenovo ThinkSystem SR650 V2 with NVIDIA A30 24GB PCIe GPU (Legacy)
What server platforms accept the Lenovo ThinkSystem SR650 V2 with NVIDIA A30 24GB PCIe 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.