AIME T504 - Deep Learning Performance Workstation

The elaborated cooling system of the AIME T504 covers the CPU with liquid cooling and the GPUs with a multi GPU compatible air cooling concept of turbo blower style fan GPUs.

The case is designed for maximum air intake supported by powerfull and temperature controlled high air flow fans. The CPU is cooled by an closed AIO water loop to reduce its impact as additional heat source.

CPU and GPUs directly exhausts the hot air outside of the case to prevent building up heat inside the case.

This prevents overheating of the GPU array and maintane high performance under full load in 24/7 scenarios.

4 times NVIDIA RTX 3090 - Ampere power

Due to the air cooling concept a high packing of GPUs is possible: With the AIME T504 4 high-end GPUs are packed into a system with the dimensions of a standard midi tower.

The AIME T504 relies on NVIDIA RTX 3090 high performance GPUs for Deep Learning.

Each NVIDIA RTX 3090 trains AI models with 10496 CUDA cores and 328 tensor cores of the third generation and 24 GB of ultra-fast GDDR6X memory. The AIME T504 combines the power of 4 of those adding up to more than 2000 Trillion Tensor FLOPS of AI performance.

Supported by NVIDIA’s CUDA-X AI SDK, including cuDNN, TensorRT, and more than 15 other libraries. Which works with all popular deep learning frameworks.

4 times NVIDIA RTX A6000 - even more Ampere power

Alternatively, the AIME T504 is available with four NVIDIA RTX A6000 GPUs.

The NVIDIA RTX A6000 is the Ampere based refresh of the Quadro RTX 6000. It features the same GPU processor (GA-102) as the RTX 3090 but with all processor cores enabled. It even outperforms the RTX 3090 with its 10752 CUDA and 336 Third-generation Tensor Cores. With the double amount of GPU memory: 48 GB GDDR6 ECC.

The NVIDIA RTX A6000 is currently the fastest available workstation GPU, best suited for the most compute and memory demanding tasks.

With the AIME T504 four of those can be packed into the most powerful workstation.

Threadripping CPU Performance

The high-end 3d generation AMD Threadripper CPU designed for workstations and servers delivers up to 64 cores with a total of 128 threads per CPU with an unbeaten price performance ratio.

The available 88 PCI 4.0 lanes of the AMD Threadripper CPU allow highest interconnect and data transfer rates between the CPU and the GPUs.

A large amount of available CPU cores can improve the performance dramatically in case the CPU is used for prepossessing and delivering of data to optimal feed the GPUs with workloads.

Up to 8 TB High-Speed SSD Storage

Deep Learning is most often linked to high amount of data to be processed and stored. A high throughput and fast access times to the data are essential for fast turn around times.

The AIME T504 can be configured with two NVMe SSDs, which are connected by PCIe lanes directly to CPU and main memory. We offer following 3 class types of SSD to be configured:

• QLC Type: high read rates, average write speed - best suitable for reading of static data libraries or archieves
• TLC Type: highest read and high write speed - best suitable for fast read/write file access
• MLC Type: highest read and write speed - best suitable for high performance databases, data streaming and virtualization

Well Balanced Components

All of our components have been selected for their energy efficiency, durability, compatibility and high performance. They are perfectly balanced, so there are no performance bottlenecks. We optimize our hardware in terms of cost per performance, without compromising endurance and reliability.

Tested with Real Life Deep Learning Applications

The AIME T504 was first designed for our own deep learning application needs and evolved in years of experience in deep learning frameworks and customized PC hardware building.

Our machines come with preinstalled Linux OS configured with latest drivers and frameworks like Tensorflow, Keras, PyTorch and mxnet. Just login and start right away with your favourite Deep Learning framework.