
Dynamic Neural Accelerator from EdgeCortix
Go Back to IP & Solutions Dynamic Neural Accelerator ML Framework EdgeCortix Dynamic Neural Accelerator (DNA), is a flexible IP core for deep learning inference
Looking for an economical 1U server for a single FPGA card? BittWare’s TeraBox 1000S is an excellent solution that doesn’t limit your card options—any of our Achronix-, Intel-, or Xilinx-FPGA equipped PCIe cards are compatible! Front panel PCIe access gives you the best access to QSFPs, with up to 8x 100G or 32x 10/25G in a single rack space!
The TeraBox 1000S supports a single- or dual-width BittWare PCIe FPGA card. Choose from a variety of cards based on Achronix Speedster 7t, Intel Stratix 10/Arria 10, or Xilinx UltraScale+ FPGAs.
Card specs vary depending on the FPGA card you select. For example, some cards support up to 512 GBytes of DDR4. Our latest UltraScale+ cards feature QSFP-DDs, which you can use with a breakout cable for up to 8x 100Gb interfaces.
For system management, BittWare’s FPGA cards are equipped with a Board Management Controller (BMC), which accepts IPMI 2.0 commands. Use it along with BittWare’s BittWorks II Toolkit to program the FPGA over USB, monitor card power and temperature, and re-program the onboard clocks. You’ll also be able to set points to shut down the card when it gets too hot, access JTAG, or access the software tools remotely.
Choosing a TeraBox FPGA server means knowing you are getting a pre-configured and tested solution. This includes setup and installation of your FPGA cards and associated hardware, your choice of operating system, and development tools. Your TeraBox arrives ready for use—giving your team more time for development and deployment.
The TeraBox 1000S supports many of BittWare’s Achronix, Intel and Xilinx FPGA-based PCIe cards. The table below lists system totals when populated with four cards*:
FPGA | Cards in Server | Memory | I/O | Processing | |
S7t-VG6 | Speedster7t | 1 | 8 banks GDDR6 (up to 192 GBytes) | 600 Gbits/sec 6x 100/50/40/25/10 GbE | 692K 6-input lookup tables (LUTs) 189 Mb embedded RAM |
520N-MX | Stratix 10 MX | 1 | 2 banks DDR4 (up to 256 GBytes) 4 banks QDRII+ (up to 1152 Mbits) | 400 Gbits/sec 4x 100/50/40/25/10 GbE | 2.1 million system logic elements 16 GBytes HBM2 |
XUP-VV8 | UltraScale+ VU13P | 1 | 4 banks DDR4 (up to 512 GBytes) 8 banks QDRII+ (up to 2.3 Gbits) | 800 Gbits/sec 8x 100/50/40/25/10 GbE | 3.8 million system logic cells Up to 12,288 DSP slices |
* Subject to availability. Contact BittWare for additional FPGA card options.
Our technical sales team is ready to provide availability and configuration information, or answer your technical questions.
"*" indicates required fields
Go Back to IP & Solutions Dynamic Neural Accelerator ML Framework EdgeCortix Dynamic Neural Accelerator (DNA), is a flexible IP core for deep learning inference
CUSTOM PRODUCT DEVELOPMENT Design + Manufacturing from BittWare, a Molex Company Custom Product Development Design + Manufacturing from BittWare, a Molex Company Build on our
White Paper Comparing FPGA RTL to HLS C/C++ using a Networking Example Overview Most FPGA programmers believe that high-level tools always emit larger bitstreams as
For a data recorder, how many drives are required for a given sustained data rate? Find the answer in this informative white paper.