Recent News

Micro-Data Centers Enable Big Data, AI for Edge Computing – EE Times

Edge computing plays an essential role in the efficient implementation of several embedded applications, such as artificial intelligence (AI), machine learning, deep learning, and the Internet of Things (IoT). Today’s data centers, however, cannot currently meet the requirements of these types of applications. This is where edge micro-data centers (EMDCs) come into play.

By moving intelligence closer to the embedded system—that is, to the edge—it is possible to create systems with a high degree of autonomy and decision-making capacity. In this way, dependence on the cloud (typical of centralized systems) is reduced, with consequent benefits in terms of energy savings, reduced latency, and lower costs.

Autonomous vehicles, robotic surgery, augmented reality in manufacturing, and drones are a few examples of edge computing’s early uses. As of today, the requirements for these applications cannot be met by current data centers (hyperscale, big, and co-location) with “cloud services,” necessitating the need for complementing edge infrastructure, such as EMDCs and “edge services.”

This edge infrastructure, hardware, and edge services must meet the following requirements:

  • high computing speed, which is required to locally process data as much as possible (i.e., on the edge)
  • highly resiliency
  • high efficiency

One EMDC, born from a collaboration between the companies HIRO and Vicor, can bring edge computing into multiple smart applications.

Importance of edge computing

By delegating greater autonomy and decision-making power to intermediate levels, edge computing reduces response times in mission-critical applications, which can thus operate in real time. The cloud is used as a memory unit for data storage, while data processing is performed as much as possible on the peripheral nodes, also improving data security and sensitivity.

Edge computing technology allows you to significantly expand the range of possible applications and services, thanks to the ability to natively support AI rather than relying on AI in the cloud.

This approach is particularly suitable for applications such as Industry 4.0, smart manufacturing, 5G, IoT, self-driving vehicles, smart cities, smart hospitals, robotics, machine vision, and more.


Based in the Netherlands, HIRO-MicroDataCenters BV specializes in the development of innovative edge infrastructures (hardware and software) that can provide intelligent Edge-as-a-Service for industrial and other end users. Edge computing is an approach that reduces dependence on the cloud by making peripheral nodes more autonomous, better performing, and efficient.

Edge computing requires compact and energy-efficient solutions that can operate even in harsh environments with space constraints, bringing computing power as close as possible to sensors and other data sources. From a hardware standpoint, efficient power systems with high power density and a small form factor are required.

Since human contact with edge devices and sensors occurs everywhere, it is necessary to implement the supporting edge infrastructure as a distributed mesh of EMDCs and edge servers, even in the most distant places, challenging environmental conditions, and confined spaces. This brings computer resources as near as possible to the data producers and users, which results in very compact form factors with high power densities that are previously unheard of, posing new technical difficulties for energy efficiency, electrical signal integrity, and high dependability.

“Our hardware designs and technological choices aim for excellence in three areas: thermal management, small form factor modularity, and power conversion,” said Fred Buining, founder and CTO of HIRO.

The solutions developed by HIRO, specifically the company’s EMDC, are scalable and compact edge computing systems that can operate even in confined environments, or even outdoors. The EMDC, shown in Figure 1 below, can be configured with a custom mix of any type and number of CPUs, GPUs, FPGAs, and NVMe (Non-Volatile Memory Express) media in compact packages—enabling the EMDC to deliver power from 1.5 kW up to 500 kW.

Figure 1: Three examples of EMDC. From left to right: 1.8 kW, 3.6 kW, 5.4 kW. (Source: HIRO/Vicor)

The EDMC hardware includes a dual-switching fabric (PCIe and Ethernet), creating high bandwidth and configuration flexibility. The switching fabric also enables the creation of large clusters of FPGAs, GPUs, and NVMes attached to a single CPU. Made entirely from solid-state components, these platforms require little maintenance and no active cooling systems.

As shown in Figure 1, heat can be dissipated via a fan-supported dry-cooler or a completely passive dry-cooler, making the EMDC passive. The solution proposed by HIRO, compared to similar products, reduces power absorption by 40%.

“We are developing small edge data centers with a maximum power up to 5.4 kW, with the size of a shoebox that can be wall mounted. That’s an incredible density that doesn’t exist anywhere,” Buining said.

Vicor’s Power Module

To achieve higher efficiency, HIRO has chosen to power the EMDC with a voltage of 48 VDC, instead of the classic 12 VDC. Widely used in the telecommunications and industrial sectors, this relatively higher voltage has the advantage of reducing I2R losses across the power supply network (PDN).

HIRO used Vicor’s high-density, high-efficiency power modules to meet this requirement, providing energy-efficient and high-density power conversion. HIRO chose Vicor’s DCM modules to perform the first conversion from 48 VDC to 12 VDC, as shown in Figure 2.

Figure 2: Vicor DCM 48VDC – 12VDC power modules contribute to HIRO energy-efficient EMDC designs. (Source: HIRO/Vicor)

Next will come support for chips like FPGAs, converting down from 48 VDC to less than 1 VDC at the point-of-load. HIRO’s EMDCs can also be installed in a solar farm or wind farm, receiving power from renewable energy sources.

Compared to other solutions based on discrete components or standard power supplies, which are bulkier and complex to design, Vicor’s modules offer a reliable and efficient high-power density solution.

Hiro plans to leverage the scalability of Vicor’s EMDC to develop a containerized solution based on six racks capable of around 200 kW each.

“Many organizations are moving their legacy applications into containerized applications and onto a cloud-service environment often in a remote data center,” Buining said. “HIRO started working with the early adopters of edge technology that are seeking an edge-service environment to be on premise.”

Building an edge infrastructure distributed throughout academic hospitals in Europe is a specific commitment made by HIRO. Hospitals are required to maintain data on site, but in order to train AI models that can help detect and treat complicated conditions like cancer, tumors, and cardiovascular diseases, they require vast data sets that go beyond their own data.

“We enable medical experts and researchers to effectively collaborate across diverse and distributed data sets and make significant progress on cardiovascular diseases, cancer, genetic diseases (Alzheimer’s, ALS, Parkinson’s), kidney disease, and etcetera by building a distributed, federated, and highly secure infrastructure across hospitals,” Buining said.

Without moving or disclosing the data beyond the hospital, HIRO is establishing the affordable infrastructure that will enable medical experts to train their models using data from other hospitals.

“Edge data centers are also supported by the European Commission, who see them as a sort of European independence from the big hyperscalers,” Buining said. “If we can catch the data at the edge, it doesn’t have to go to the cloud.”

HIRO is also involved in the €16 million (about $16.6 million) BRAINE project, which was funded by the EU’s ECSEL Joint Undertaking with €8.5 million (about $8.8 million). Multiple test applications targeting smart cities, smart hospitals, smart manufacturing and robotics, and smart supply chains will be supported by four test-bed locations across the EU: two in the Netherlands, one in Italy, and one development platform in Hungary.

This UrIoTNews article is syndicated fromGoogle News

About Post Author