我們接受加密貨幣付款 — 讓您持有的幣種也能買到真正的硬體設備。以加密貨幣付款,直接購入真正的硬體設備。 全球訂單滿 US$399,即享免費保險運送。簡約外箱,不印任何品牌標誌,從德國出貨。訂單滿 US$399,即享免費保險運送。 結帳一開始,您的匯率即鎖定 30 分鐘。匯率鎖定 30 分鐘。 每件商品皆原廠密封、序號經過核對,並享有 24 個月保固。原廠密封,保固 24 個月。

追蹤清單 登入

DELL

DELL 490-BHVP NVIDIA A2 Tensor Core 16GB GDDR6 60W FHHL GPU

SKU CH-8P4T-JXPKM

NVIDIA A2 Tensor Core GPU with 16GB GDDR6 memory and 60W power draw for Dell PowerEdge R640, R650, R660, R740, R750, R760 and compatible servers

  • TypeRack
  • 16 GB GDDR6 memory accelerates AI inference workloads
  • 60 W power envelope fits dense rack deployments
  • Full-height half-length form factor installs in standard PCIe slots
  • Tensor Core architecture speeds matrix operations for deep learning
  • Compatible with PowerEdge R640 through R7625 servers

Dell NVIDIA A2 16GB Tensor GPU

The Dell 490-BHVP is a full-height, half-length PCIe add-in card built around the NVIDIA A2 Tensor Core GPU with 16 GB of GDDR6 memory. It draws 60 W and targets inference workloads in compatible Dell PowerEdge servers.

Factory firmware and Dell qualification

The card ships with updated firmware and carries the MG5JX Dell part number, meaning it has been validated for the listed PowerEdge chassis. No additional software licence or driver bundle is mentioned in the supplied data.

Suited for dense inference in qualified Dell racks

Buyers running AI inference on supported PowerEdge platforms such as the R640, R750 or XR12 can install this card directly. It is not intended for workstations, non-Dell servers or systems that require a different power envelope or form factor.

商品特色

  • 16 GB GDDR6 memory accelerates AI inference workloads
  • 60 W power envelope fits dense rack deployments
  • Full-height half-length form factor installs in standard PCIe slots
  • Tensor Core architecture speeds matrix operations for deep learning
  • Compatible with PowerEdge R640 through R7625 servers

商品規格

BrandDELL
Model490-BHVP
TypeRack

商品問答

Will the 16 GB GDDR6 memory on the A2 need replacement during the server's life?

The 16 GB GDDR6 is soldered to the board and not user-replaceable; plan for a full GPU swap if memory capacity becomes a bottleneck.

What thermal limit appears first when the 60 W A2 runs inference continuously in a rack?

The 60 W power envelope is the binding constraint; sustained loads will hit the GPU's thermal throttle point before the PCIe slot or chassis airflow limits are reached.

How much fan noise and heat does the FHHL card add when it draws 60 W all day in a PowerEdge R750?

At 60 W the card relies on the server's existing fans; expect a measurable but modest rise in inlet-to-exhaust delta-T and corresponding fan-speed increase.

同類商品

顧客也比較過

同一分類,同一套結帳流程 — 支援八種加密貨幣,匯率鎖定 30 分鐘。