AIVRES Visits Brazil Alongside PerfilGlobal and PerfilComp to Advance AI Factory Infrastructure and HPC Liquid Cooling
- 4 days ago
- 5 min read
The growing collaboration between the companies expands opportunities for high-performance projects in Brazil and Latin America, combining AIVRES' global portfolio with PerfilGlobal's and PerfilComp's expertise in HPC and artificial intelligence infrastructure.
Por Perfilcomp, Farroupilha/RS
21 de agosto de 2026
A promising partnership
The evolution of artificial intelligence already transformed the infrastructure required to support it. Environments featuring high-density NVIDIA GPUs, HPC clusters, massive data volumes, and new power and cooling requirements have become increasingly important topics for companies, universities, and research centers.
It is within this context that AIVRES is strengthening its presence and engagement with the Brazilian market.
Representatives from the manufacturer recently visited Brazil and Perfil headquarters as part of a technical and commercial agenda that reinforces the collaboration between AIVRES, PerfilGlobal, and PerfilComp.
The partnership between AIVRES and PerfilGlobal brings to Brazil a company specializing in one of the fastest-growing areas of technology infrastructure: high-performance computing and infrastructure for artificial intelligence. AIVRES is one of NVIDIA's major partners for the development of AI Factory environments, an area in which PerfilGlobal and PerfilComp also have extensive experience, as direct channels for NVIDIA solutions.
Who is Aivres?
AIVRES is a global provider of IT infrastructure products and solutions, operating across cloud computing, artificial intelligence, edge computing, and high-density data centers, with solutions ranging from HPC liquid cooling and air cooling to immersion liquid cooling.
The company currently has a commercial presence in more than 70 countries and regions, in addition to 9 offices, 3 manufacturing facilities, and 2 wo research and development centers.
One of its main areas of focus is precisely the infrastructure required to support artificial intelligence.
AIVRES' Full Stack AI strategy combines computing infrastructure, resource management platforms, and tools designed for large language models. Its architecture encompasses AI computing clusters, parallel storage systems, and RDMA interconnects, as well as tools for managing and optimizing these environments.
This approach allows the AI discussion to extend beyond individual servers or GPUs. In large-scale projects, compute, storage, networking, cooling, and management need to operate as an integrated architecture.
From GPUs to complete Data Centers
Another differentiator of the AIVRES portfolio is the breadth of its solutions.
In addition to accelerated servers for AI and HPC workloads, the manufacturer develops storage infrastructure and liquid cooling solutions designed for high-density environments.
In 2026, for example, the company introduced an end-to-end liquid-cooled data center solution designed to support up to 130 kW per rack, targeting AI, general-purpose computing, and edge computing scenarios. The manufacturer also reports accumulated experience in projects involving more than 1,000 liquid-cooled racks across different regions.
In another project disclosed by the company, AIVRES completed an AI data center in Europe featuring 100 high-density liquid-cooled racks in just four months, including deployment, testing, optimization, and commissioning.
The ability to operate across different infrastructure layers is particularly relevant to AI projects. As GPU quantities and densities increase, infrastructure planning can no longer focus solely on processing capacity. Power, cooling, interconnects, storage, and cluster operations become equally important considerations.
A company built for the new generation of AI
The recent evolution of the portfolio helps illustrate how the company intends to position itself in this new generation of infrastructure.
In May 2026, the company introduced an All-QLC storage solution designed for clusters of up to 10,000 GPUs, aiming to reduce data-feeding bottlenecks in large-scale AI training environments.
The manufacturer also develops MotusAI, a platform designed to manage and orchestrate computing resources for artificial intelligence, and KSManage, focused on the management and monitoring of AI data center infrastructure.
Liquid cooling is another important area of development. In 2026, AIVRES' AI Compute Pod received recognition from both the iF Design Award and the Red Dot Award. The solution combines high-density infrastructure, liquid cooling, and integrated management for AI data center environments.
These developments reflect an important shift in the market: artificial intelligence projects are no longer being evaluated solely based on the capabilities of individual servers. Instead, they are increasingly being designed and deployed as complete infrastructure environments.
Why does this partnership matters to Brazil
This is where the collaboration between AIVRES, PerfilGlobal, and PerfilComp becomes particularly relevant.
AIVRES brings end-to-end development and manufacturing capabilities for high-performance environments, ranging from liquid cooling infrastructure to complete data centers and HPC environments.
PerfilGlobal adds its international reach and its ability to structure projects and supply solutions across different markets. PerfilComp, in turn, contributes its experience in the Brazilian market, particularly in the design, integration, deployment, and support of infrastructure for data centers, HPC, and artificial intelligence.
The combination of these capabilities creates opportunities for joint projects ranging from high-performance servers and GPU-accelerated environments to complete HPC and AI clusters, storage, networking, and data center infrastructure.
The visit to Brazil provided an opportunity to deepen technical discussions, better understand local market requirements, and align opportunities in which the three companies
can work together.
For Brazilian customers, this means broader access to new architectures and technology alternatives at a time when demand for computing capacity is growing rapidly.
A Partnership Focused on AI, HPC, and Large-Scale Projects
Companies are moving from experimental artificial intelligence projects to production environments.
Universities and research centers increasingly require computing capacity for simulations and scientific workloads. At the same time, the growing density of computing environments is creating demand for new approaches to power and cooling. In this context, infrastructure is no longer simply the underlying layer that supports technology. It increasingly determines how far these projects can scale.
The partnership between AIVRES, PerfilGlobal, and PerfilComp emerges precisely at this point of convergence.
On one side is a global manufacturer developing servers, storage, AI platforms, and data center solutions for high-density environments. On the other are companies with close ties to the Brazilian and Latin American markets and experience integrating complex infrastructure projects.
AIVRES' visit to Brazil therefore represents more than an institutional engagement. It marks the beginning of a collaborative effort aimed at transforming global technological capabilities into concrete projects for companies, universities, research centers, and organizations building the next generation of infrastructure for artificial intelligence and HPC.
Talk to a Specialist
Want to learn more about AIVRES solutions for AI and HPC?
PerfilGlobal and PerfilComp, in partnership with AIVRES, are ready to support companies, universities, and research centers in developing infrastructure projects for artificial intelligence, HPC, storage, and high-density data centers.
From architecture design and solution sizing to environment deployment, our team can help identify the infrastructure best suited to each project's specific requirements.
Contact our specialists and discover how we can support your next AI and HPC project.
Comments