Bare Metal Servers do away with hypervisors and virtualization: The computing resources are only equipped with a host operating system, offering users the maximum scope to design their own systems. They can freely design customized flavors using virtualization technologies. Bare Metal Servers can also be used without virtualization: Because they do not come with a hypervisor, they offer more consistent performance than virtual machines, making them an ideal solution for big data analyses. Furthermore, Bare Metal Servers also offer better license conditions for the software used.


The allocated resources run in a GDPR-compliant environment and fulfil all European requirements to run secure workloads with the scalable options of a cloud. Non-European authorities have no access.

BMS enables the use of dedicated computing resources without hypervisor overhead. This customized hardware offers an increased level of security and performance while avoiding the “noisy neighbour” effect (negative influences on the performance of the allocated resources caused by other users).

Bare Metal Service provides high integration with other services to provide computing, storage, and network resources. Linked services include Virtual Private Cloud (VPC), Image Management Service (IMS), Elastic Volume Service (EVS), Scalable File Service (SFS), Volume Backup Service (VBS), Cloud Eye (CES), and many more services to meet business needs in a cloud.

The BMS service enables the isolation of physical resources and cannot be influenced by other tenants.

Some BMS support InfiniBand at up to 100 Gbit/s between BMS and provide low latency connectivity to meet high-speed interconnection requirements.
Compute Optimized
The compute optimized flavor is designed for general business purposes that require high performance, higher memory ratio and transactional workloads.
Scenarios:
- Regulatory requirements for physical resource isolation and selfdetermination of hypervisors.
- Online transaction processing (OLTP) databases to run transactional workloads.
Supported Flavor | physical.o2 |
Sockets x Cores per Socket | 2 x 8 |
CPU | Intel® Xeon® E5-2667 v4 |
Memory | 256 GB |
Local Disk | 2 x 800GB (RAID 1) |
Configuration | 2 x 10GE |
High-Performance Computing
The High-Performance computing BMS type provides a large number of CPU cores, a large memory size, and high throughput. These BMSs use Intel Skylake V5 CPUs, InfiniBand NICs and can be quickly provisioned.
Scenarios:
- High-performance computing with EVS disks and lower network latency.
- HPC cluster environment with High-Speed-Connectivity between data nodes.
Supported flavor | physical.o2 |
Sockets x Cores per Socket | 2 x 18 |
CPU | Intel® Skylake® 6151 V5 |
Memory | 192 GB |
Local Disk | 1 x 1.6 TB NVMe SSD |
Configuration | 1 x 100G IB, 2 x 10GE |
Memory Optimized
Memory-optimized BMS use processors optimized for high and stable computing performance. They offer a memory size of up to 4 TB based on DDR4 for memory-intensive applications with high demands on network bandwidth and data throughput in Packets per Second (PPS).
Scenarios:
- High-performance relational databases (e.g., MySQL) and NoSQL databases (e.g., MongoDB or Cassandra).
- Distributed web-based caches are used to improve the performance of dynamic websites with database connections (e.g., Memcached or Redis).
- Applications for processing big unstructured data in real-time (e.g., financial services or Hadoop/Spark clusters).
- High-performance computing (HPC) and electronic design automation (EDA).
Supported flavor | physical.m2 |
Sockets x Cores per Socket | 4 x 24 |
CPU | Intel® Xeon E7-8890 v4 |
Memory | 2 - 4 TB |
Local Disk | 2 x 600 GB (RAID 1) |
Configuration | 2 x 10GE |
GPU Accelerated
Graphic-accelerated BMS use processors optimized for high and stable computing performance. They offer a memory size of up to 4 TB based on DDR4 for memory-intensive applications with high demands on network bandwidth and data throughput in Packets per Second (PPS).
Scenarios:
- AI training, scientific research with Nvidia P100 or V100 GPU
- HPC environment with GPU functions and InfiniBand for low latency requirements between nodes
Supported flavor | physical.p1 | physical.p2 |
Sockets x Cores per Socket | 2 x 14 |
CPU | Intel® Xeon E5-2690 v4 |
Memory | 512 GB |
Local Disk | 2 x 600GB (RAID 1) |
Configuration | 1 x 100G IB |
GPU | 8 x Tesla P100 | 8 x Tesla V100 |
High-Performance Computing (HPC)
BMS enables high-performance computing for various industries such as scientific research, banking, or telecommunications.

Advantages:
- Latest CPUs for high performance
- Network latency in μs and up to 100 Gbit/s InfiniBand network support
- Compatability and free accessible OpenStack APIs
Example Scenario
Multi-cluster HPC environment with low latency requirements and high network performance.
Databases
Bare Metal Server support Elastic Volume Service (EVS) to meet performance and security requirements.

Advantages:
- Scalable Capacity
- Reliable data storage
- BYOL
- Large in-memory capability
Example Scenarios
Online transaction processing (OLTP) databases | In-memory databases like SAP HANA® | Mission-critical database services of governments and financial institutions
Big Data
BMS offers memory-optimised flavors with up to 4 TB memory to meet Big Data analysis requirements.

Advantages
- Dedicated local disks for data analysis
- High-Speed-Connectivity between data nodes
- Flexible, scalable and extendable infrastructure
Example Scenario
Big Data environment with Hadoop cluster
New Features

Pricing overview

Documentation


Profiteer van ons adviesaanbod!
* Voucher is inwisselbaar tot 31.12.2023. Neem contact met ons op voor het bedrag van de voucher bij de boeking. Het kortingsvolume is alleen geldig voor klanten met een factuuradres in Duitsland en vervalt 2 maanden na het afsluiten van het contract. Het tegoed wordt verrekend met de geldige catalogusprijzen volgens de servicebeschrijving. Een uitbetaling is uitgesloten.