HPC & Clustering
Today’s datacenters require solutions that can meet the ever increasing demand for reliability, performance and ease of use. Furthermore, clients want solutions that offer scalable IT solutions that grow with their business and offer the flexibility to deploy new technology at a reasonable price. Clustering is a real solution for this challenge.
The concept is simple and proven. It is a cost effective way to provide high performance computing for a wide variety of applications in different fields of engineering and science.
Possible workloads include CFD & FEM modeling, Bioinformatics, Geo data analysis, Business intelligence, Financial analysis and Digital media & visualization.
Clustering is the practice of connecting multiple servers or processors together into a single, unified computing resource. Processors can be of the same architecture (homogenous) or different architecture (heterogenous). Because it behaves like a single large resource, a clustered system offers many valuable benefits to a modern datacentre environment including:
Central management Melrow offers a variety of tools that offer monitoring, system inventory and deployment facilities from a single point of control.
Consolidation Workloads from around the world can be handled in one central datacenter accessible from all over the world. This makes clusters ideal for license consolidation.
Resource optimization Geographically separated departments can share the resources as daytime is spread across continents.
High speed interconnection Modern day Terminal Server solutions can be used in the cluster to provide access to applications that need high IO capacity, 3D capabilities or both.
High availability Because the workload is spread across different machines downtime of one server doesn’t inflict availability of the application as a whole.
Failover and disaster recovery Data can be spread across multiple sites or even multiple continents to provide extra redundancy and fast disaster recovery procedures.
Melrow Technology designs, develops and manufactures clusters and High Performance Computing (HPC) systems for
customers with applications requiring more computing power than a single commercially available computer can support.
Our integrated and fully tested cluster solutions based on Linux or Microsoft Compute Cluster Server with comprehensive services regarding maintenance, support and remote management.
Melrow’s partnerships with the main third-party application providers guarantees clients a reliable and above all fast cluster solution that accelerate in scalability, energy efficiency, price/performance ratio and manageability. Melrow’s portfolio ranges from Intel Xeon and AMD Opteron processors to Intel Itanium or IBM Cell processers. Operating system support includes Red Hat enterprise Linux, SUSE Linux Enterprise and Microsoft Compute Cluster Server. Complemented by a complete storage portfolio and leading fibre and Infiniband technologies We can design, specify and integrate a wide range of technologies involved in developing a cluster or HPC including:
- Compute node specifications
- Data traffic and storage management
- Power management, sequencing and backup
- Operating systems, application loading tools, software image management, health monitoring, and performance analysis and tuning
- Ergonomics, serviceability and shipping
- Acoustic, vibration, power and thermal characterization
For further information about our HPC Clusters please do contact us.
If you would like to have a look at the possibilities for a turn-key cluster with installation, configuration and/or maintenance, use our configurator.