Content for the HPC community and innovation enthusiasts: tutorials, news and press releases for users, engineers and administrators
- All News
- Aerospace & Defence
- Artificial Intelligence
- Cloud Platform
- E4 Various
- European Projects
- Kubernetes Cluster
With E4 Computer Engineering, the future of HPC goes through containers
The company’s solutions respond to the demand for a modern and effective workload management. Among customers ITS Umbria, Academy of Applied Sciences and Technologies.
E4 Computer Engineering, one of the main Italian companies for high-performance solutions for HPC, artificial intelligence and containerization, offers a complete range of container-based proposals that allow you to optimize workloads ensuring benefits in terms of development and management of IT environments.
In the constantly evolving world of High Performance Computing (HPC), advances and algorithmic improvements are achievable only if you have appropriate technological infrastructures to which you can entrust the development of new and ongoing projects. Data analytics, machine learning and deep learning require extremely high computing power, scalable and available almost everywhere.
The complexity of planning, implementation and balancing of infrastructure load grows exponentially. Companies are increasingly relying on containers to respond to these challenges, increase automation and be able to organize infrastructure, development teams and processes and elaborate complex data models thanks to GPU (Graphic Processing Units) computing.
Containers allow you to isolate applications, grouping libraries, dependencies and files necessary for their execution, allowing developers to work and move applications between different environments – from production, to development, to testing – without security problems and keeping intact all the application functionalities. They require minimal processing resources, are efficient and simple to implement, can be distributed in clusters, with individual containers for single components of complex applications. Thanks to this separation, developers can update application components individually, without the need to modify them entirely, ensuring efficient use of the underlying infrastructure.
To allow proper management and effective application of containers, E4 Computer Engineering provides two advanced solutions – GAIA and KAPTAIN – which offer maximum flexibility and infrastructure customization to adapt to specific needs.
GAIA (GPU Artificial Intelligence Appliance) is a scalable, high-performance hardware platform, developed for Data Analytics and Machine Learning workloads on GPUs, able to successfully face the computational challenge of training the most complex Deep Learning models. The solution consists of a powerful multi-GPU server, on which E4 Analytics Studio is installed, a modern, complete and fully containerized software stack that integrates the main tools and frameworks used for the development of advanced data analysis applications. KAPTAIN is a turnkey Kubernetes cluster, guaranteed for ease of use and high performance, suitable for hosting the most demanding computing and storage workloads. Easy to use and intuitive, it has a graphical interface that allows you to simplify the most complex operations. It allows to deal with the analysis of huge volumes of data with the paradigm of high-performance Cloud Native Data Science, offering on the same system, platforms and frameworks that, in other scenarios, would require dedicated infrastructures – physical and/or virtual.
ITS Umbria, Academy in Applied Sciences and Technologies that offers training courses to young graduates who wish to acquire high technological skills for a more agile and qualified entry into the world of work, has recently relied on the functionality and performance of KAPTAIN to implement an infrastructure that could host large-scale data analysis platforms, and allow maximum flexibility in the use of the cluster’s computing and storage resources.
“We needed a“ turnkey ”hardware and software platform, performing but also easy to use and maintain. E4’s KAPTAIN represented the perfect meeting point: a hardware platform that met our computing needs, a customizable software stack, flexible resource management and a large and modern application catalog. The implementation was quick and allowed us to take advantage of the power of Kaptain very quickly”, said Emilio Di Giacomo and Fabrizio Montecchiani, professors of ITS Umbria.
“Today, companies need highly performing and scalable infrastructures for complex projects, which increasingly include the analysis and processing of large volumes of data”, said Mario Rosati, Chief Innovation Officer of E4 Computer Engineering. “Our commitment is aimed at the design of high-performance infrastructures capable of hosting the platforms, services and frameworks necessary for the typical workflows of analyzing large amounts of data, combining flexibility and high availability with ease of use, to simplify the development and management of container-based applications, capable of supporting the most demanding Data Science projects”.
To know more contact us