In line with our commitment to meet the surging demand for GPUs for AI and HPC acceleration, we're excited to announce that we've added a fleet of on-demand GPUs to service customer needs globally. These additions ensure that our customers have immediate access to the most powerful, efficient, and versatile computing resources available anytime and affordably.
New GPU additions
NVIDIA A40 and A6000 GPUs are now available on-demand on our platform. Both GPUs are built on NVIDIA’s Ampere architecture, which is a significant improvement over its predecessors in terms of energy efficiency, computational power, and AI training and inference.
NVIDIA A40
The NVIDIA RTX A40 delivers 37.4 teraflops of FP32 (single-precision floating-point) performance, with 10,752 CUDA cores that facilitate parallel computing. Additionally, it includes 336 Tensor Cores designed for deep learning performance. It is also equipped with 48 GB of GDDR6 memory, offering a bandwidth of 696 GB/s. This vast memory pool and high bandwidth efficiently handle large datasets and complex AI models.
| Specification | NVIDIA RTX A40 |
|---|---|
| FP32 Performance | 37.4 TFLOPS |
| CUDA Cores | 10,752 |
| Tensor Cores | 336 |
| Memory Type | 48 GB GDDR6 |
| Memory Bandwidth | 696 GB/s |
The RTX A40 is currently available on demand from $0.79 /hour. Configure Now.
NVIDIA A6000
The NVIDIA RTX A6000, on the other hand, is a versatile GPU that can be used for AI, HPC, and graphics-intensive applications, like video editing (VFX) and 3D rendering. Its Ampere architecture, 10,752 CUDA processing cores, and 48GB of GDDR6 memory provide a broad bandwidth of 768.0 GB/s, enabling it to handle large datasets and intensive AI workloads.
| Specification | NVIDIA RTX A6000 |
|---|---|
| CUDA Cores | 10,752 |
| Memory Type | 48 GB GDDR6 |
| Memory Bandwidth | 768.0 GB/s |
| Ray Tracing Cores | 84 (Next-Generation) |
The RTX A6000 is also equipped with 84 next-generation Ray Tracing (RT) cores, which makes it excellent for VFX and complex 3D models.
You can get started using the RTX A6000 now from $0.79 /hour. Start now.
NVIDIA V100
The NVIDIA Tesla V100 is designed with a strong emphasis on AI and deep learning applications. It delivers up to 14 TFLOPS of FP32 performance with 5,120 CUDA cores. Furthermore, it features 640 Tensor Cores, specifically engineered for accelerated deep learning performance. The V100 is also equipped with 16 GB or 32 GB of HBM2 memory, providing a bandwidth of up to 900 GB/s. This extensive memory capacity and superior bandwidth are adept at efficiently managing large datasets and complex AI models.
| Specification | NVIDIA Tesla V100 |
|---|---|
| FP32 Performance | Up to 14 TFLOPS |
| CUDA Cores | 5,120 |
| Tensor Cores | 640 |
| Memory | 32 GB HBM2 |
| Memory Bandwidth | Up to 900 GB/s |
These additions underscore our commitment to delivering cutting-edge technology that meets the needs of new and existing customers.
Get access to the NVIDIA V100 now from $0.49 /hour. Start now.
More A5000 GPUs coming soon
We are thrilled to announce that we’ll be adding more NVIDIA A5000 GPUs to our extensive roster. The A5000 is designed to tackle a wide array of professional graphics and compute tasks, ensuring versatility and reliability. These forthcoming additions will further boost our users' capabilities. Stay tuned!
US Locations Update
We're delighted to share that we're expanding our network with new strategic locations while bolstering current hubs to elevate our service, availability, and performance.
New Los Angeles Data Center Now Live!
Our new data center, us-losangeles-1, is now fully operational and strategically situated to bolster the capabilities of our US-based customers.
Equipped with 16x Tesla V100 GPUs and 10 dedicated CPU compute hosts, us-losangeles-1 guarantees robust computing resources primed for the most demanding tasks.
The LA data center brings several advantages to our users in the United States, including reduced latency and increased overall capacity to handle peak demand scenarios.
GPUs added to New York, Santa Clara, and Irvne Locations
Lastly, we have added more GPUs to our New York and Santa Clara locations which are now live. We will also be adding more GPUs to our Irvine location this month. This expansion ensures that our customers can access high-performance computing resources whenever and wherever they need them, all while maintaining the reliability and efficiency for which we are known.
Get in touch with us today!
Learn more: LinkedIn , Twitter , YouTube , Get in touch .
