Skip to content
Menu
  • Contact Us
  • Support

Let’s Talk

GigaIO
  • Gryf
  • Solutions
  • Resources
  • News
    • News
    • Press Releases
    • Events
  • Company
    • Leadership
    • Careers
  • Partners
    • Become a Partner
  • Let’s Talk
Close Menu
  • Contact Us
  • Support

Let’s Talk

Featured Product

Fabric Card

For Rack-Scale Composable Fabric Computing

GigaIO Fabric Card
Talk to an Expert

Highlights

High Performance

Up to 512Gb/s speed | 128GB/s bandwidth

Lowest Latency

Less than 10ns

Easy To Use Cabling

Dual QSFP-DD connections for copper or optical

Low Profile

Fits in full- or half- height PCIe slots

Ideal For

  • AI/ML training and inferencing clusters
  • High-performance computing environments
  • Data analytics acceleration
  • Composable infrastructure deployments
  • Scale-up computing architectures

The GigaIO™ Fabric Card enables non-blocking low-latency 512Gb/s rack-scale composable fabric computing.

It gives AI/ML, HPC, and Data Analytics users the flexibility to create exactly the system they need for optimized performance and reduced total cost of ownership.

GigaIO’s Fabric Card supports the high-performance, cabled interface to rack-scale cluster subsystems across our revolutionary AI fabric hyper-performance network. The card provides two modes:

  • Host Mode for installing into a host / head-node server
  • Target Mode for installing into an Accelerator Pooling Appliance (APA) / resource box.

The Fabric Card enables the creation of shared pools of vendor agnostic PCIe devices from GPUs to FPGA to Storage or memory, any PCIe compliant device. It also enables unique features such as auto-discovery and management of disaggregated resource pools.

For full specifications on the Fabric Card please download the PDF below.

GigaIO Data Sheet for Fabric Card
Download the PDF

Sign up for GigaIO News

"*" indicates required fields

This field is for validation purposes and should be left unchanged.

Back To Top
GigaIO Logo

© 2026 GigaIO
GigaIO and FabreX are trademarks of GigaIO Networks, Inc., all rights reserved.

Privacy
Policy Terms & Conditions

  • Gryf
  • Solutions
  • Resources
  • News
    • Press Releases
    • Events
  • Company
  • Partners
  • Contact Us
  • Support
© 2019 GigaIO
GigaIO and FabreX are trademarks of GigaIO Networks, Inc., all rights reserved.

Privacy Policy
Terms & Conditions

Envelope

Contact Us

  • This field is for validation purposes and should be left unchanged.

Contact Our Team Today

"*" indicates required fields

This field is for validation purposes and should be left unchanged.

Speak to Our Experts

"*" indicates required fields

This field is for validation purposes and should be left unchanged.
Country*

Schedule a chat about SuperNODE

  • This field is for validation purposes and should be left unchanged.

Learn More

"*" indicates required fields

This field is for validation purposes and should be left unchanged.
How can we help you?

Curious to see how it works? Check out this 15-minute demo on how to liberate your compute resources.

Great! We’ll send you the information you need to make the best decision for your EDU environment.

Name*

Request More Information

  • This field is for validation purposes and should be left unchanged.

Schedule a Meeting

  • This field is for validation purposes and should be left unchanged.

Subscribe to GigaIO News

"*" indicates required fields

This field is for validation purposes and should be left unchanged.

Speak to a GigaIO HPC Expert

  • This field is for validation purposes and should be left unchanged.

Become a Partner

  • This field is for validation purposes and should be left unchanged.

Contact Us

  • This field is for validation purposes and should be left unchanged.

Niraj Mathur

Niraj has over 20 years of industry experience in strategic and product marketing, product management, business development, customer applications and advanced silicon engineering. He has held senior leadership roles and led global, cross-functional teams to support these disciplines. Niraj was instrumental in driving numerous successful networking products at Nortel Networks, Quake Technologies, AppliedMicro, Snowbush, Gennum, Semtech and Rambus. He has defined, developed and supported carrier grade hardware and software for the world’s leading telecom, enterprise and cloud customers. His past projects include Ethernet PHYs, core Internet switches, metro optical routers, high-speed silicon IPs and PCI Express products. Niraj holds a Bachelor of Computer Engineering from McGill University and an MBA from Cornell University.

Download Your Resource

After submitting the form, your resource will be delivered to your inbox.

  • This field is for validation purposes and should be left unchanged.
  • This field is hidden when viewing the form
  • This field is hidden when viewing the form
  • This field is hidden when viewing the form
  • This field is hidden when viewing the form

Request a demo

  • This field is for validation purposes and should be left unchanged.