The Trump Administration has fired a number of National Science Foundation employees who had been handpicked for their expertise in AI, threatening the agency’s ability to sustain key AI research, Bloomberg reported. One of the affected departments inside NSF, called the Directorate for Technology, Innovation, and Partnerships, was instrumental in funneling government grants focused on […]

© 2024 TechCrunch. All rights reserved. For personal use only.

Read more

Three key investors have left General Catalyst amidst a series of recent changes at the firm, which now describes itself as an “investment and transformation company,” TechCrunch has learned. The departed managing directors include Deep Nishar and Kyle Doherty, who co-led General Catalyst’s late-stage strategy known as Endurance, and Adam Valkin, one of the three […]

© 2024 TechCrunch. All rights reserved. For personal use only.

Read more

Google Gemini users can now access the AI chatbot directly from the iPhone’s lock screen, thanks to an update released on Monday first spotted by 9to5Google. Users can now call up Gemini Live, Google’s relatively real-time voice feature for its AI chatbot, before they unlock their phone by adding a Gemini widget to their lock […]

© 2024 TechCrunch. All rights reserved. For personal use only.

Read more

Last week, the Florida-based company Lonestar Data Holdings launched a shoebox-size device carrying data from internet pioneer Vint Cerf and the government of Florida, among others, on board Intuitive Machines’ Athena lander. When its device lands on the moon later this week, the company will be the first to explicitly test out a question that has been on some technologists’ minds of late: Maybe it’s time to move data centers off Earth?

After all, energy-guzzling data centers are springing up like mushrooms all over the world, devouring precious land, straining our power grids, consuming water, and emitting noise. Building facilities in orbit or on or near the moon might help ameliorate many of these issues. 

For Steve Eisele, Lonestar’s president and chief revenue officer, a big appeal of putting data storage on the moon is security. “Ultimately, the moon can be the safest option where you can have a backup for your data,” Eisele says. “It’s harder to hack; it’s way harder to penetrate; it’s above any issues on Earth, from natural disasters to power outages to war.”

Lonestar’s device is equipped with eight terabytes of storage, about as much as a high-end laptop. It will last for just a couple of weeks before lunar night descends, temperatures plummet, and solar power runs out. But the company expects that to be enough time to test practicalities like downloading and uploading data and verifying secure data transfer protocols.

And it has bigger plans. As early as 2027, the company aims to launch a commercial data storage service using a bunch of satellites placed in the Earth-moon Lagrange point L1, a gravitationally stable point 61,350 kilometers above the moon’s surface. There, the spacecraft would have a constant view of Earth to allow continuous data access.

Other companies have similar aspirations. The US space company Axiom, best known for organizing short trips to the International Space Station for private astronauts, intends to launch a prototype server to the station in the coming months. By 2027, the firm wants to set up a computing node in low Earth orbit aboard its own space station module. 

A company called Starcloud, based in Washington state, is also betting on the need to process data in space. The company, which raised an $11 million round in December and more since then, wants to launch a small data-crunching satellite fitted with Nvidia GPUs later this year. 

Axiom sees an urgent need for computing capacity in space beyond simply providing an untouchable backup for earthly data. Today’s growing fleets of Earth- and space-observing satellites struggle with bandwidth limitations. Before users can glean any insights from satellite observations, the images must be downlinked to ground stations sparsely scattered around the planet and sent over to data centers for processing, which leads to delays.

“Data centers in space will help expedite many use cases,” says Jason Aspiotis, the global director of in-space data and security at Axiom. “The time from seeing something to taking action is very, very important for national security and for some scientific applications as well. A computer in space would also save costs that you need to bring all the data to the ground.”

But for these data centers to succeed, they must be able to withstand harsh conditions in space, pull in enough solar energy to operate, and make economic sense. Enthusiasts say the challenges are more tractable than they might appear—especially if you take into account some of the issues with data centers on Earth.

Better in space?

The current boom in AI and crypto mining is raising concerns about the environmental impact of computing infrastructure on Earth. Currently, data centers eat up around 1% or 2% of the world’s electricity. This number could double by 2030 alone, according to a Goldman Sachs report published last year. 

Space-tech aficionados think orbiting data centers could solve the problem.

“Data centers on Earth need a lot of power to operate, which means they have a high carbon footprint,” says Damien Dumestier, a space systems architect at the European aerospace conglomerate Thales Alenia Space. “They also produce a lot of heat, so you need water to cool them. None of that is a problem in space, where you have unlimited access to solar power and where you can simply radiate excess heat into space.”

Dumestier, who led an EU-funded study on the feasibility of placing large-scale IT infrastructure in Earth’s orbit, also sees space as a more secure option than Earth for data transportation and storage. Subsea fiber-optic cables are vulnerable to sabotage and natural disasters, like the undersea volcanic eruption that cut Tonga off from the web for two weeks.

High above Earth, data centers connected with unhackable laser links would be much harder to cut off or penetrate. Barring antisatellite missiles, space-based nuke explosions, or interceptor robots, these computing superhubs would be nigh untouchable. That is, except for micrometeorites and pieces of space debris, which spacecraft can dodge and, to some extent, be engineered to withstand. 

Outside of Earth’s protective atmosphere, the electronic equipment would also be exposed to energetic particles from the sun, which could damage it over time. Axiom plans to tackle the problem by using hardened military equipment, which Aspiotis says survives well in extreme environments. Lonestar thinks it could avoid the harsh radiation near the moon by ultimately placing its data centers in lava tubes under the lunar surface.

Then there is the matter of powering these facilities. Although solar power in Earth’s orbit is free and constantly available, it’s never previously been harvested in amounts needed to power data infrastructure at the scale existing on Earth. 

The Thales Alenia Space study, called ASCEND (an acronym for “advanced space cloud for European net zero emission and data sovereignty”), envisions orbiting data platforms twice as large as the International Space Station, the largest space structure built to date. The server racks at the heart of the ASCEND platforms would be powered by vast solar arrays producing a megawatt of power, equivalent to the electricity consumption of about 500 Western households. In comparison, the solar panels on the ISS produce only about one-quarter that amount—240 kilowatts at full illumination.

Launch costs—and the environmental effects of rocket launches—also complicate the picture. For space-based data centers to be an environmental win, Dumestier says, the carbon footprint of rocket flights needs to improve. He says SpaceX’s Starship, which is designed to carry very large loads and so could be cheaper and more efficient for each kilogram launched, is a major step in the right direction—and might pave the way for the deployment of large-scale orbital data centers by 2030. 

Aspiotis echoes those views: “There is a point in the not-too-distant future where data centers in space are as economical as they are on the ground,” he says. “In which case do we want them on the ground, where they are consuming power, water, and other kinds of utilities, including real estate?”

Domenico Vicinanza, an associate professor of intelligent systems and data science at Anglia Ruskin University in the UK, tempers the optimism, however. He says that moving data centers to space en masse is still a bit of a moonshot. Robotic technologies that could assemble and maintain such large-scale structures do not yet exist, and hardware failures in the harsh orbital environment would increase maintenance costs. 

“Fixing problems in orbit is far from straightforward. Even with robotics and automation, there are limits to what can be repaired remotely,” Vicinanza says. “While space offers the benefit of 24-7 solar energy, solar flares and cosmic radiation could damage sensitive electronic equipment and current electronics, from mainstream microchips to memories that are not built and tested to work in space.”

He also notes that any collisions could further crowd Earth orbit with space debris. “Any accidental damage to the data center could create cascading debris, further complicating orbital operations,” he says.

But even if we don’t move data centers off Earth, supporters say it’s technology we will need to expand our presence in space. 

“The lunar economy will grow, and within the next five years we will need digital infrastructure on the moon,” Eisele says. “We will have robots that will need to talk to each other. Governments will set up scientific bases and will need digital infrastructure to support their needs not only on the moon but also for going to Mars and beyond. That will be a big part of our future.”

Read more

Technological advances continue to move at breakneck speeds. While companies struggle through their digital transformation journeys, even more new technologies emerge, with promises of opportunity, cost savings—and added complexity. Many companies have yet to fully adopt AI and ML technologies, let alone figure out how newer technologies like generative AI might fit into their programs.

A 2024 IDC survey revealed 22% of tech leaders said their organizations haven’t yet reached full digital maturity, and 41% of respondents said the complexity of integrating new technologies and approaches with existing tech stacks is the biggest challenge for tech adoption.

To fuel successful technology adoption and maximize outcomes, companies need to focus on simplifying infrastructure architecture rather than how to make new technologies fit into existing stacks. “When it comes to digital transformation, choosing an architectural approach over a purely technology-driven one is about seeing the bigger picture,” says Rajarshi Purkayastha, the VP of solutions at Tata Communications. “Instead of focusing on isolated tools or systems, an architectural approach connects the dots—linking silos rather than simply trying to eliminate them.”

Establishing the robust global network most companies need to connect these dots and link their silos requires more capability and bandwidth than traditional networks like multiprotocol label switching (MPLS) circuits can typically provide in a cost-effective way. To keep pace with innovation, consumer demands, and market competition, today’s wide area networks (WANs) need to support flexible, anywhere connectivity for multi-cloud based services, remote locations and users, and edge data centers.

Understanding hybrid WAN

Traditional MPLS became the gold standard for most WAN architectures in the early 2000s to address the mounting challenges brought by the rapid growth of the internet and subsequent rapid expansions of enterprise networks. Today, as technological advances continue to accelerate, however, the limitations of MPLS are becoming apparent: MPLS networking is expensive; hard-wired connectivity is difficult to scale; and on its own, it doesn’t fit well with cloud computing adoption strategies.

In 2014, Gartner predicted hybrid WANs would be the future of networking. Hybrid WANs differ from traditional WANs in that the hybrid architecture facilitates multiple connection points: private network connections for mission-critical business, usually via the legacy MPLS circuits; and public network connections, typically utilizing internet connections such as 5G, LTE, or VPN, for less critical data traffic; and dedicated internet access (DIA) for somewhat critical traffic.

In 2025, we are seeing signs Gartner’s hybrid WAN prediction might be coming to fruition. At Tata Communications, for example, hybrid WAN is a key component of its network fabric—one facet of its digital fabric architecture, which weaves together networking, interaction, cloud, and IoT technologies.

“Our digital fabric simplifies the complexity of managing diverse technologies, breaks down silos, and provides a secure, unified platform for hyper-connected ecosystems,” explains Purkayastha. “By doing so, it ensures businesses have the agility, visibility, and scalability to succeed in their digital transformation journey—turning challenges into opportunities for innovation and growth.”

Hybrid WAN provides the flexible, real-time data traffic channeling an architectural approach requires to create a programmable, performant, and secure network that can reduce complexities and ease adoption of emerging technologies. “It’s not just about solving today’s challenges—it lays the groundwork for a resilient, scalable future,” says Purkayastha.

Benefits of hybrid WAN

Hybrid networking architectures support digital transformation journeys and emerging tech adoption in several ways.

More efficient, even intelligent, data trafficking. A hybrid architecture brings together multiple avenues of data flow from MPLS and internet connectivity, which provides highly flexible, resilient architecture along with increased bandwidth to decrease network congestion. It also allows companies to prioritize critical data traffic. Hybrid WANs can also combine the hyper-secure connectivity of MPLS with software-defined WAN (SD-WAN) technology, which allows for intelligent switching across a company’s information highways. If, for instance, one route encounters latency or malfunctions, that traffic will be automatically re-routed, helping to maintain continuous connectivity and reduce downtime.

Increased scalability. The agility and flexibility of a hybrid WAN allows companies to dynamically scale bandwidth up or down as application needs change. An agile WAN architecture also paves the way for scaling business operations.

Less complex cloud migration and easier adoption of new technologies. Adding internet connectivity to MPLS circuits allows for seamless data trafficking to the cloud, providing a more direct way for companies to transition to cloud-first strategies. Easing cloud migration also opens doors for emerging technologies like AI, generative AI, and machine learning, enabling companies to innovate to remain relevant in their markets.

Improved productivity. The internet speed and connectivity of a hybrid WAN keeps geographically separated company locations and remote workers connected, increasing efficiency and collaboration.

Easier integration with legacy systems. A hybrid approach allows legacy MPLS connections to remain, while offloading less sensitive data traffic to the internet. The ability to incorporate legacy applications and processes into a hybrid architecture not only eases integration and adoption, but helps to maximize returns on network investments.

Network cost savings. Many of the benefits on this list translate into cost savings, as internet bandwidth is considerably cheaper than MPLS networking. A reduction in downtime reduces expenses companywide, and the ability to customize bandwidth usage at scale gives companies more control over network expenses while maximizing connectivity.

Deploying a hybrid WAN

A recent collaboration between Air France-KLM Group and Tata Communications highlights the benefits a hybrid WAN can bring to a global enterprise.

Air France looked to increase its network and application performance threefold without incurring additional costs—and while ensuring the security and integrity of their network. A hybrid WAN solution—specifically, using MPLS and internet services from Tata Communications and other third-party providers—afforded the flexibility, resilience, and continuous connectivity they needed.

According to Tata Communications, the hybrid architecture increased Air France’s network availability to more than 99.94%, supporting its global office locations as well as their customer-facing applications, including passenger and cargo bookings and operating service centers.

“However, which connectivity to choose based on location type and application is complex, given the fact that networks vary by region, and one has to also take into account regulations, for e.g., in China,” says Purkayastha. “This is what Tata Communications helps customers with—choosing the right type of network, resulting in both cost savings and a better user experience.”

Enabling business for success

Innovating and expanding enterprise operations in today’s era of increasingly complex technology evolutions requires businesses to find agile and cost-effective avenues to stay connected and competitive.

As emerging machine learning and AI technologies aren’t likely to slow, hybrid network architectures likely are going to become necessary infrastructure components for companies of all sizes. The flexibility, resiliency, and configurability of a hybrid WAN provides a relatively straightforward, lightweight network upgrade to allow companies to focus on business objectives with less time and expense worrying about network reliability and reach. “At the end of the day, it isn’t just about technology—it’s about enabling your business to stay agile, competitive, and ready to innovate, no matter how the landscape shifts,” says Purkayastha.

Read more
1 … 788 789 790 791 792 … 3,372