The ideal HPC Cluster
Medooza is an HPC cluster designed specifically to meet a client’s particular requirements. It is characterised by a carefully engineered optimisation of hardware and software components, with custom or “standardised” configurations for the X86, GPU, storage and high-speed interconnects.
To manage cluster flexibility and redundancy, the solution uses the OpenStack set of software components, which provides common services for infrastructure management.
The Medooza solution includes integrated tools and standardized installation procedures to ensure the release of stable and optimized operating environments that allow fully autonomous operational management by the customer.
Engineered, not assembled
Medooza is based on a suite of hardware and software components suitably harmonised by our expert engineers, lending itself to the continuous integration of new modules and modern features.
To guarantee maximum reliability, Medooza is put through a long-lasting stress test during production.
Prior to delivery, we produce an output report with test and real-time performance data for our clients.
Medooza 2.0 is available in Entry, Premium and Mission critical versions.
The MEDOOZA approach
Medooza offers an intuitive and easy-to-manage approach to HPC clusters typical of traditional infrastructures but with the flexibility and robustness of a modern system.
New modules and features are released on a regular basis to ensure state-of-the-art performance and functionality that seamlessly integrates with your existing infrastructure.
AGNOSTIC
For both hardware and software, Medooza supports the widest possible range of high-performance CPUs (ARM, x86_64), accelerators (GPU, FPGA), networks (IB, RoCE) and disks (NVMe, SSD) as well as the most popular operating systems (Red Hat Enterprise Linux, Rocky Linux)
MODULAR
The various development workspaces accessible to users are flexible and reconfigurable. Their management takes place in environment modules created by E4 according the client’s needs.
MULTICLUSTER
Medooza is natively multicluster. The same infrastructure can be used to serve multiple clusters within the same local network.
FUTURE PROOF
It is possible to expand the HPC system by adding new nodes (scale-out), check the state of the hardware and recondition old clusters where appropriate.
Designed to be the best, always
Advanced Workload Computing Engineering processes require in-depth data analysis, complex visualizations, and sophisticated modeling and simulation. For them, we use high-performance components such as GPU and FPGA server accelerators, as well as networking and storage.
Based on twenty years of experience working with the academic and industrial sectors, E4 engineers configure and implement HPC clusters using a suite of components suitably harmonised by Medooza.
Discover the advantages
Solution Layout
The MEDOOZA 2.0 solution consists of two main modules:
- the computing infrastructure, consisting of high-performance components such as server accelerators (GPUs and FPGAs), networking and high-performance storage
- the operating platform (E4HPC-PLATFORM), which enables the services and functionalities of the HPC cluster, as well as its full and effective management.
INFRASTRUCTURE + PLATFORM
Medooza 2.0 – ENTRY – Single Bare-metal Control Node
- Medooza contains all the services required to operate the cluster, through the use of containers
- It is possible to activate additional E4HPC-PLATFORM components such as ICE4HPC and others that are sold separately
- Ideal for small HPC batches that do not require bare-metal reconfiguration over time
SMALL ENVIRONMENT / STANDARD APPLICATIONS | BATCH WORKFLOWS
Medooza 2.0 – PREMIUM – Single OpenStack Control Node
- Medooza contains all the services necessary for the cluster to function, through the use of Virtual Instances
- OpenStack, active only on the Control Node (on CEPH storage), takes care of bare-metal deployment of compute nodes
- Includes monitoring and management features such as Talos, CubeView and others that are sold separately
- It is possible to activate E4HPC-PLATFORM add-ons such as Talos, CubeView, ICE4HPC, and others that are sold separately
- With cloud architecture, ideal for HPC customers who require flexibility to autonomously adapt infrastructure to changes
PRIVATE CLOUD ENVIRONMENT / STANDARD APPLICATIONS | BATCH WORKFLOWS
Medooza 2.0 – MISSION CRITICAL – Solution with at least 3 OpenStack Control Nodes
- Medooza contains all the services necessary for the cluster to function, through the use of Virtual Instances
- OpenStack takes care of the As-a-Service deployment of the infrastructure, through preconfigured images, CEPH storage, expandable on demand, is on dedicated and redundant switches
- Includes monitoring and management features such as Talos, CubeView and others that are sold separately
- It is possible to activate E4HPC-PLATFORM add-ons such as Talos, CubeView, ICE4HPC, and others that are sold separately
- Ideal for customers who require a high degree of robustness and flexibility
HIGH AVAILABILITY ENVIRONMENTS | BATCH / INTERACTIVE WORKFLOWS
Technical features
INFRASTRUCTURAL FEATURES
MANAGED
HPC resource manager & scheduler configured to obtain maximum efficiency from the cluster and have full control over workload execution.
MONITORED
Centralized monitoring and alerting thanks to a single and intuitive control point that allows you to check the status of the cluster and receive alerts in case of anomalies.
STACK SOFTWARE
The software stack includes the main opensource tools for developing and running high-performance applications: compilers, scientific and numerical libraries, MPI, OpenMP.
ADDITIONAL COMPONENTS
Medooza can be enriched with a series of additional features according to the specific needs of each project:
- Optimized compilers and libraries
- Critical services delivered in HA
- Parallel and/or ephemeral file system
GUIDELINES
Since the architecture is extremely flexible and adaptable to any environment, here are some guidelines to keep in mind when designing a Medooza solution.
E4HPC-PLATFORM | CONTROL NODE VERSIONS
| ENTRY | PREMIUM | MISSION CRITICAL |
| HPC Scheduler (Add-On) | HPC resource manager & scheduler configured with resource usage accounting for operational reporting of cluster resource utilization | HPC resource manager & scheduler configured with resource usage accounting for operational reporting of cluster resource utilization |
| Stack Software (Add-On) | Software Stack, which includes the main opensource tools: compilers, scientific and numerical libraries, MPI and OpenMP | Extended software stack that includes the Advanced version components plus libraries optimized by hardware vendors, Python computing environments, and common computing applications |
| Ephemeral file system (Add-On) | Ephemeral file system (Add-On) | Ephemeral file system, for handling Scratch I/O intensive workloads (BeeOnd) |
| Singularity (Add-On) | Singularity (Add-On) | Singularity, Container Platform that enables the creation and execution of containers that package parts of software in a portable and reproducible manner |
| IPA | Identity Manager, for managing users and the access to resources | IPA | Identity Manager, for managing users and the access to resources | IPA | Identity Manager, for managing users and the access to resources |
| NFS | Shared Filesystem NFS, for data sharing within the cluster | NFS | Shared Filesystem NFS, for data sharing within the cluster | NFS | Shared Filesystem NFS, for data sharing within the cluster |
| ZABBIX | Centralized Web UI for collecting, manipulating, analyzing, and visualizing data related to the operation of the entire computing infrastructure | ZABBIX | Centralized Web UI for collecting, manipulating, analyzing, and visualizing data related to the operation of the entire computing infrastructure | ZABBIX | Centralized Web UI for collecting, manipulating, analyzing, and visualizing data related to the operation of the entire computing infrastructure |
| FOG (Add-On) | A free open-source network computer cloning and management solution; can capture, deploy, and manage various distributions, that can be securely managed with FOG Project | FOG (Add-On) | A free open-source network computer cloning and management solution; can capture, deploy, and manage various distributions, that can be securely managed with FOG Project | FOG (Add-On) | A free open-source network computer cloning and management solution; can capture, deploy, and manage various distributions, that can be securely managed with FOG Project |
| N/A | TALOS (Add-On) | Centralised CLI for monitoring, configuring and administrating the entire Hardware and Software infrastructure | TALOS | Centralised CLI for monitoring, configuring and administrating the entire Hardware and Software infrastructure |
| N/A | EXAMON (Add-On) | a data collection and analysis platform oriented to the management of big data | EXAMON (Add-On) | a data collection and analysis platform oriented to the management of big data |
| N/A | ICE4HPC (Add-On) | Provides web access to computing resources to be used interactively only through the workload manager services (SLURM) | ICE4HPC (Add-On) | Provides web access to computing resources to be used interactively only through the workload manager services (SLURM) |
Architectural advantages
OPEN SOURCE
The E4 HPC cluster is built using Open Source technology, validated by both our R&D laboratories and by our years of experiences with clients from both academic and corporate spheres.
BUILDING BLOCKS
Medooza’s design is based on hardware building blocks validated by our engineers, and configured and managed by internal tools developed in opensource.
SCALABLE
Medooza’s scaleout architecture allows it to respond transparently and natively to growing needs.
EFFECTIVE FLEXIBILITY
With a simple but appropriate reconfiguration of the environment, it is possible to add a variety of new nodes whilst leaving the systm architecture unchanged.
Why choose the E4 solution
UNIQUE
E4 engineers always design a solution by first understanding the client’s specific requirements and expectations.
Here is a summary of the principles with uphold throughout the design process:
• LISTENING: understand the client’s needs
• EXPLORATION: Collaborate with domain-specific ISVs / partners to optimize every single aspect of the design
• UNRESTRICTED: research and select the best infrastructure and implementation processes for the solution
• VERSATILE: provide of on-site and cloud-based solutions for production, testing and evaluation
VALIDATED
All the systems included in a solution must first pass firmware, homogeneity, sanity and setup checks before configuration. Next, the following general performance tests are performed: HPL (High Performance Linpack) to test the computing power of a machine, measured in FLOPs; STREAM to test the memory access bandwidth, measured in MB/s; FIO to test disk access speed, measured in MB/s and IOPS. Once the solution is ready, further tests are carried out in agreement with the client for the verification of system function and comparison with expected performance.
TESTED
Individual components are tested through burn in tests (developed by E4) lasting up to 120 hours to guarantee a unique, perfectly engineered and functioning system, reducing both the DoA (Dead on Arrival) and the ” early failure” rate after release. This significantly improves the overall reliability of E4’s HPC solution.
SERVICED
E4 is currently amongst only a few companies that delivers high-end services not only to both the academic and industrial sectors, but also to research centres of national and international importance, with which it has collaborated on the design, configuration and execution of solutions containing thousands of nodes for high performance and complex processing.
E4 is therefore able to provide clients with whatever level of support and advice they need, regardless of the size or complexity of their system, guaranteeing them perfect efficiency indefinitely:
• Level 1 & Level 2 Support, onsite service, 24/7 support
• Onsite specialists, infrastructure evolution, on-demand performance tuning.
Add-on Modules | New features coming soon
In addition to the features currently offered by its basic configuration, Medooza can be enriched with a series of add-ons. Some are integrable and others are available on demand. Contact us for their release date.
ADDITIONAL OPTIONS – ON DEMAND
- Advanced Slurm for accounting, billing, QoS, FairShare, etc.
- Backup: implementation of solutions based on FOG/Bacula
- Disaster Recovery: implementation of solutions based on StorWare
- ICE4HPC: provides web-based access to compute resources that can be used interactively through workload manager services (SLURM) configured to optimize the user experience for both interactive users and those submitting jobs to batch queues from the command line.
Click here to learn more
E4HPC-PLATFORM | ADDITIONAL MODULES
EXAMON
ExaMon (Exascale Monitoring) is a data collection and analysis platform designed to manage large volumes of information. It easily manages heterogeneous data and, through a series of plugins, makes it possible to control data from the logs of the various services and from the sensors of all the equipment in the datacenter
TALOS
System for centralized management of the cluster. CLI interface for centralized management of Cluster nodes for configuration, monitoring and diagnostics activities.
HIGH PERFORMANCE COMPUTING
