One Server - Two GPUs

On-Demand GPU Servers with Dual Intel^® MAX 1100

Dramatically accelerate AI/ML and HPC workloads with minimal investments.

2 x Intel MAX 1100 GPUs
56 Xe cores + 48 GB HBM2E memory per unit
GPU-to-GPU Intel Xe Link Bridge
No licensing fees for GPU features
Provisioning via API, CLI, WebUI, or IaC

Already have an account? Log in

Unlock Breakthrough Performance for Data-Intensive Computing

To get the most out of rising technologies such as AI or ML, you need an IT infrastructure that can handle inferencing, model enhancement, and High-Performance Computing (HPC) with ease. phoenixNAP’s Bare Metal Cloud lets you pair state-of-the-art hardware with a low-latency, high-throughput network.

Provision dedicated servers featuring next-gen GPU technologies in minutes using the platform’s API, CLI, or web portal. Choose pre-configured servers powered by dual Intel^® Max 1100 GPUs and 4^th Gen Intel Xeon^® Scalable CPUs to cost-effectively accelerate end-to-end AI and analytics pipelines with no licensing fees or inconsistencies in CPU and GPU implementation.

OVERVIEW

Explore Intel MAX GPU Benefits with Bare Metal Cloud

3:51

Benefits

The Future of AI Computing. Just a Few Clicks Away.

Cutting-Edge Technology

Streamline and accelerate data-hungry workloads with robust and dedicated compute, storage, and network resources.

Built for AI

Use Intel’s highest-density discrete GPU to streamline AI inference and unlock breakthrough performance for HPC.

Provisioning in Minutes

Quickly spin up workload-optimized dedicated GPU servers at different locations using API, CLI, IaC, or the platform’s WebUI.

Locations

Phoenix, AZ

Ashburn, VA

More Coming Soon...

Phoenix, AZ

Ashburn, VA

More Coming Soon...

We Value
Your Trust

We pride ourselves on being the preferred choice for our clients.

"With Bare Metal Cloud, we were able to provision high-performance physical services in minutes across multiple geographic locations via simple calls to the API. We have the freedom to provision server instances that are optimized for compute or memory-intensive workloads. On top of this, everything is open-source and we don’t have to worry about vendor lock-in."

Matt Russell, Chief Cloud Officer, Namecheap

“Migrating to phoenixNAP’s Bare Metal Cloud yielded a 50% monthly cloud cost reduction compared to our previous home-rolled solution in AWS. And our in-house solution was already 40% cheaper than the AWS-managed OpenSearch! We were able to leverage consistent speeds of over 20 Gbps between Bare Metal Cloud and multiple dedicated storage servers simultaneously, allowing us to separate the processing and data storage aspects of our infrastructure to optimize both speed and costs.”

Ryan McLaughlin, Senior DBA, SpyFu

“We need to analyze huge volumes of data, which requires advanced compute resources to provide insights and value. Intel and phoenixNAP technologies are vital to ensuring efficient and secure data processing. We’ve had a great collaboration with both companies over the years and we are excited to continue exploring the potential of their technologies.”

Nik Schork, Ph.D., Deputy Director of Quantitative Science, TGen

Instances

Meet Your Workload-Optimized GPU Servers

Bare Metal Cloud helps you easily build AI infrastructure clusters running on dedicated resources. Take your pick of the server instances powered by Intel Max 1100 GPUs. Spin up your pre-configured server with a few clicks or a single API call and tap into the raw performance of bare metal with cloud-like flexibility.

Explore other BMC instances »

Instance Type	CPU	Specs	Hourly	Monthly	12 mo	24 mo	36 mo
d3.g2.c1.xlarge		SGX enabled - c x GHz (GB RAM, , ) 2x Intel MAX 1100 GPU
d3.g2.c2.xlarge		SGX enabled - c x GHz (GB RAM, , ) 2x Intel MAX 1100 GPU
d3.g2.c3.xlarge		SGX enabled - c x GHz (GB RAM, , ) 2x Intel MAX 1100 GPU
+ 15 TB Free Bandwidth + On-demand access to terabytes of scale-out Network Storage or S3-compatible Object Storage

Instance Type

d3.g2.c1.xlarge

CPU

SpecsSGX enabled - c x
GHz (GB RAM,
,
)
2x Intel MAX 1100 GPU

Billing Model

Hourly

Monthly

12 mo

24 mo

36 mo
d3.g2.c2.xlarge

CPU

SpecsSGX enabled - c x
GHz (GB RAM,
,
)
2x Intel MAX 1100 GPU

Billing Model

Hourly

Monthly

12 mo

24 mo

36 mo
d3.g2.c3.xlarge

CPU

SpecsSGX enabled - c x
GHz (GB RAM,
,
)
2x Intel MAX 1100 GPU

Billing Model

Hourly

Monthly

12 mo

24 mo

36 mo

+ 15 TB free bandwidth
+ On-demand access to terabytes of scale-out Network Storage or S3-compatible Object Storage

Features

Leverage Intel Max-Powered GPU Servers as a Service

Running your data-intensive computing models in a highly parallel environment has never been this simple or affordable. Access Intel’s future-proof technologies through API-driven bare metal servers optimized for high-performance computing. Optimize your IT spend, enhance existing AI models, and speed up inferencing or real-time analytics without code refactoring or infrastructure management complexities.

Intel Max 1100 GPUs

Consume Intel’s highest performing, highest density discrete GPU as a service. Drive breakthroughs with record efficiency.

Intel X^e Matrix Extensions

Perform up to 256x Int8 operations per clock for faster AI training and inference. Utilize vector and matrix capabilities in a single device.

Stunning GPU Specs

Boost visualization and animation speed with 2x 408 MB L2 cache and 48 GB of high-bandwidth memory + 56 built-in ray tracing units.

Intel X^e Link

Enjoy blazing-fast speeds of PCIe Gen5 and seamless communication between 2 Intel Max GPUs in a single system.

Intel oneAPI

Create multiarchitecture applications that target their hardware of choice. No refactoring. No vendor lock-in.

IaC Integration

Automate server management with Terraform, Ansible, Chef, or Pulumi and optimize DevOps, AIOps, or DataOps pipelines.

Scale-Out Storage

Near-instantly scale storage capacity with terabytes of Network File Storage or S3-compatible Object Storage.

Lightning-Fast Network

Enjoy a fast, redundant, and safe data transfer with a 50 Gbps network (2x25 Gbps bonded) + free 20 Gbps DDoS protection.

Deployment Flexibility

Quickly provision bare metal optimized for AI across locations. Use reservations to optimize costs.

Integration

Intel Max 1100 GPU in Bare Metal Cloud

Through automation, Bare Metal Cloud lets you reap all the benefits of dedicated compute, storage, and network resources without the hassle of manual infrastructure provisioning. With streamlined access to dedicated servers with Intel’s highest-density GPUs and AI accelerator-packed CPUs, you can scale your resources and optimize the performance of your GPU-intensive workloads at will.

AI and HPC Acceleration Powered by Bare Metal Cloud
Intel Max 1100 GPUs	4^th Gen Intel Xeon Scalable CPUs	phoenixNAP’s Bare Metal Cloud
1 server – 2 GPUs Dual Intel Max 1100 GPU cards connected via Intel’s high-speed Xe Link Bridges.	More Cores + Increased Multi-Socket Bandwidth Up to 1.53x gen-on-gen performance gain.	Global and edge deployment in a matter of minutes.
108 MB L2 cache Delivers up to 2x performance gain over competition for HPC and AI workloads.	Intel AMX Accelerates DL inference and AI training workloads.	Dedicated servers equipped with high-performance DDR5 RAM and NVMe storage.
56 GPU Cores Allows for parallel processing of complex models with 2x 56 X^e cores.	Intel Data Streaming Accelerator (DSA) Boosts the performance of storage, networking, and data-intensive workloads.	Direct access to hardware without vendor lock-in or hypervisor overhead.
48 GB of HBM2E Memory Supports large and complex DL models and datasets without memory constraints.	Intel Dynamic Load Balancer (DLB) Dynamically distributes network data across CPU cores.	API-driven provisioning and management via IaC tools (Terraform, Ansible, Pulumi, Chef).
Intel XMX Performs up to 256x Int8 operations per clock for faster AI training and inference.	Intel QuickAssist Technology (QAT) Accelerates encryption, decryption, and data compression.	Easy deployment of terabytes of low-cost network storage or S3-compatible object storage.
Ray Tracing Acceleration Delivers faster scientific visualization and animation with 56 ray tracing units.	Intel In-Memory Analytics Accelerator (IAA) Improves analytics performance and accelerates database query throughput.	50 Gbps network speed. 20 Gbps free DDoS protection. Public and private network options.
Shared Code Investments Optimizes CPU and GPU performance without duplicating your development efforts.	Intel Software Guard Extensions (SGX) Isolates sensitive data in a secure enclave with hardware-based memory protection.	Flexible billing and bandwidth models. Pay-as-you-go options with discounts for reservations.

Use Cases

The Perfect Infrastructure for Data-Intensive Applications

AI and Machine Learning

Improve the efficiency of LLM inferencing or large-scale data analytics. Stay ahead of the AI curve with minimal investments.

Healthcare and Medical Research

Boost the speed and quality of medical imaging analysis, genomics research, drug discovery, or patient data analytics.

Financial Services and Banking

Optimize risk assessment, fraud detection, algorithmic trading, and customer analytics to make your financial operations more effective.

Manufacturing and Industry 4.0

Improve quality control, predictive maintenance, supply chain optimization, and industrial automation with reduced downtime and TCO.

FAQ

Frequently Asked Questions

Intel Data Center GPU MAX 1100 GPU is Intel’s latest and fastest GPU. With over 100 billion transistors, up to 128 Xe Cores, built-in ray tracking acceleration, and Intel’s Xe Matrix Extensions (XMX), it is designed to significantly boost AI inference and HPC workloads.

Depending on the operating system and the server instance you choose during the deployment process, you can deploy a pre-configured Bare Metal Cloud server in anywhere from 60 seconds to 15 minutes.

These server instances are workload-optimized and designed to accelerate AI/ML inference, enhance already trained models, boost high-performance computing (HPC) or data and database analytics processing speed, or streamline E-Commerce or AdTech workloads.

While Bare Metal Cloud instances with Intel MAX 1100 GPUs are currently available in Phoenix (AZ) and Ashburn (VA) only, the platform offers over 50 different server instances deployable across the U.S., Europe, and Asia-Pacific.

All Bare Metal Cloud servers feature enterprise-grade local NVMe drives. Additionally, Bare Metal Cloud lets you easily and near-instantly scale your storage capacity with terabytes of VAST Network File Storage or S3-compatible object storage.

Bare Metal Cloud servers with Intel MAX 1100 GPUs can be consumed through a 1, 12, 24, or 36-month reservation. There are no additional licensing fees for any Intel MAX GPU features you use. Discounts are available for monthly or yearly reservations. Billing and usage info is transparent and accessible via the platform’s self-service portal.

Need more info?

Let's get in touch!

Contact us today and revolutionize the way you process data-intensive workloads! Our Sales team will get back to you within two business days to help you quickly find the right solution for your use case.