HPE Data Processing Unit InfiniBand NDR200/Ethernet 200Gb 2-port QSFP112 FHHL B3220 Adapter

MPN: P66386-H21
Quick Code: B49586438
(HPE DPU IB NDR200/EN 200G 2P - ADPTR)
🔍
HPE Data Processing Unit InfiniBand NDR200/Ethernet 200Gb 2-port QSFP112 FHHL B3220 Adapter
Note: Image accuracy is not guaranteed. Please do not rely solely on the image for your purchase.

📹 View Product Demo

$15237.13
$18284.56 inc. VAT
All products on this website are available duty-free in the EU
Overseas Customers with EU-VAT number do not pay VAT. - Verify it here
Quick Code: B49586438
MPN: P66386-H21
Manufacturer: HPE
EAN Code: 0190017690599
💰 Get $109.71 Cash Back if paying by Debit Card or Bank Transfer!
📧 Request More Info

Have questions? Need a quote?

Contact Us
📬 Notify me when price drops
NOTE: The information below is provided for your convenience only and we cannot guarantee its accuracy. If necessary, please verify with us before purchasing.

HPE Data Processing Unit InfiniBand NDR200/Ethernet 200Gb 2-port QSFP112 FHHL B3220 AdapterData Processing Unit InfiniBand NDR200/Ethernet 200Gb 2-port QSFP112 FHHL B3220 Adapter

Do you need a data-center-on-chip platform to deliver better performance for your HPC/edge environments? The HPE Data Processing Unit is a cloud infrastructure processor that governs the offloading, acceleration, and isolation of tasks like networking, storage, security and management to improve efficiency and performance. This series features two default operating modes: DPU and SuperNIC. By leveraging DPU mode, a data center-to-business application separation is created, resulting in a zero trust infrastructure with customized operations and an overall reduction in the total cost of ownership. In SuperNIC mode, the DPU too runs as an enhanced solution for east to west traffic in GPU systems, enabling accelerated networking for AI computing. SuperNICs support workloads in power-efficient, low-profile designs and can provide superb AI workload efficiency through proven RoCE. Utilize the HPE Data Processing Unit for consistent and predictable performance for accelerating AI computing.