8 GPU Server with up to 8 Tesla P100, K80, or M40 GPUs
Microway Octoputer is our densest GPU server offering. Ideal for well ported, massively parallel GPU codes, it supports up to 8 NVIDIA GPUs in a 4U footprint.
Each Octoputer arrives fully integrated with the NVIDIA CUDA Toolkit. It passes rigorous stress tests for over 48 hours prior to shipment.
Need more compute power? Link multiple Octoputer servers together for highly parallel CUDA cluster.
- Up to 8 NVIDIA Tesla P100, K80, or M40 GPUs
- NVIDIA CUDA Toolkit installed and configured – Ready to run CUDA jobs!
- Up to 56 processor cores
- Up to 3 TB memory
- Up to 28 TB storage built-in
- (2) Intel Xeon Xeon Scalable Processor “Skylake-SP” CPUs (clock speed: up to 3.6 GHz)
- Six-Channel DDR4 2666 MHz ECC/Registered Memory (24 slots)
- Eight PCI-Express 3.0 x16 double-width slots
- One PCI-Express 3.0 x16 single-width slot
- Four PLX 8747 PCI-Express 3.0 Switches (switched architecture)
- (14) Hot-Swap 2.5” 6Gbps drive bays
- Front and rear USB 3.0 ports
- Two integrated Intel 10G Ethernet Ethernet ports
- IPMI 2.0 with Integrated LAN Support
- 3200W 2+1 Redundant High-Efficiency Power Supplies
- NVIDIA Quadro GPUs for visualization
- Select NVIDIA GeForce GPUs for low-cost code development
- Additional mezzanine slot for Mellanox ConnectX-4 56Gb FDR or 100Gb EDR InfiniBand
- PGI Accelerator Compilers for GPUs
- Intel compilers, libraries and tools
Supported for Life
Our technicians and sales staff consistently ensure that your entire experience with Microway is handled promptly, creatively, and professionally.
Telephone support is available for the lifetime of your server(s) by Microway’s experienced technicians. After the initial warranty period, hardware warranties are offered on an annual basis. Out-of-warranty repairs are available on a time & materials basis.
System Price: $7,500 to $50,000
Each Microway system is customized to your requirements. Final pricing depends upon configuration and any applicable educational or government discounts.