NVIDIA/HP Tesla V100 PG500-216 32GB HBM2 PCIe 3.0 x16 Refurbished Data Center AI Deep Learning HPC Accelerator GPU
NVIDIA/HP Tesla V100 PG500-216 32GB HBM2 PCIe 3.0 x16 Refurbished Data Center AI Deep Learning HPC Accelerator GPU
How We Test
Every unit is diagnostic-tested and verified healthy before shipment. Learn about our full testing process ↓
About this item
| Type | Enterprise |
|---|---|
| Brand | NVIDIA/HP |
| Series | Nvidia Tesla V100 |
| Board Name | PG500-216 |
| Nvidia Part Number | 699-2G500-0216-400 |
| HPE Part Number | P44861-001/P44899-001 |
| GPU Memory | 32 GB |
| Memory Bandwidth | 900GB/sec |
| GPU Architecture | NVIDIA Volta |
| NVIDIA Tensor Cores | 640 |
| NVIDIA CUDA Cores | 4608 |
| Double-Precision Performance | 7 TFLOPS |
| Single-Precision Performance | 14 TFLOPS |
| Tensor Performance | 112 TFLOPS |
| ECC | Yes |
| Interconnect Bandwidth | 32GB/sec |
| System Interface | PCIe Gen3 |
| Max Power Consumption | 250 W |
| Thermal Solution | Passive |
| Compute APIs | CUDA, DirectCompute, OpenCL, OpenACC |
| Weight | 2.681 lb |
| Dimensions | 1.37 x 10.5 x 4.37 in |
Warranty
|
Condition - Seller Refurbished : 1 Year Limited Period Warranty
|
|
| Satisfaction Guarantee: Return for refund within 30 days of package reception date. | |
| This item is covered by serverpartdeals.com Return Policy | |
| To initiate a return please fill out the Return Form | |
| For frequently asked questions about warranties and returns please visit our FAQs |
Disclaimer
Some references may appear on this website or product descriptions to trademarks belonging to others not affiliated with ServerPartDeals.com (SPD), including references which may not display the ® symbol. All third-party trademarks (including logos and icons) referenced by SPD remain the property of their respective owners. Unless specifically identified as such, SPD’s use of third-party trademarks does not indicate any relationship, sponsorship, or endorsement between SPD and the owners of these trademarks. Any references by SPD to third party trademarks is to identify the corresponding third-party goods and/or services and shall be considered nominative fair use under trademark law.This animation is a visual aid only and does not represent the actual testing methodology, sequence, or results for this drive.
We test every graphics card before you do.
Every graphics card Server Part Deals sells passes an internal 4-stage testing protocol before it ships. While many resellers stop at a basic power-on check, we validate video output, GPU and VRAM performance, thermal behavior, and stability under sustained load. Each card is tested to confirm that it meets or exceeds its rated specifications.
Stage 1: Physical Inspection
- PCIe connector and contact inspection
- PCB, component, and solder check
- Fan, heatsink, and shroud inspection
- Display-port and power-connector check
Stage 2: Hardware Validation
- GPU model and specification verification
- VRAM capacity and type confirmation
- BIOS and device-ID validation
- Sensor and fan-control check
Stage 3: Performance Testing
- Graphics and compute benchmarking
- GPU and VRAM load testing
- Clock-speed and power validation
- Multi-round stress-test cycles
Stage 4: Stability & Integrity Testing
- Artifact and rendering-error scan
- VRAM error detection
- Temperature and thermal-throttling review
- Final display-output verification