In a world where the International Space Station is welcoming private astronauts and Elon Musk is launching satellites into space—promising to bring internet access to people around the world—it feels as though anything is possible. However, as space becomes increasingly accessible beyond government and research contexts, ensuring the functionality of the technology that makes space exploration possible is even more important. What’s more, it turns out the solution to transforming the space industry isn’t up in the atmosphere; rather, it will begin right here on Earth.
In Extreme Environments, Reliability is King
Anyone who has ever dropped a laptop knows how fragile computers can be. Even the most advanced computers created by humans still rely on delicate electronic components. So, to survive in space’s harsh environment, computers need specialized parts.
When electronics journey into space, they’re exposed to numerous variables that threaten their integrity. Solar winds, temperature fluctuation, cosmic radiation, and Van Allen radiation belts all impact how computers behave in space.
Due to the technologies currently available, spacecraft are typically designed with two environments. One of these is relatively protected from the above elements, and temperature-controlled, and thus electronics are restricted to these areas. There are clear advantages to freeing up electronics to be used more liberally around a spacecraft. However, with temperatures ranging from –200 to +200°C, temperature tolerance and material fatigue due to temperature changes are a major obstacle to achieving high reliability.
Radiation effects can be temporary, non-destructive disturbances wherein a computer chip’s internal bits get flipped. However, total ionizing dose and single-event latch-ups spell destruction for chips.
While non-destructive effects can amount to hiccups during routine tasks like playing a video or establishing basic communication channels in space, the stakes become far higher when a computer is responsible for gathering data. In these situations, the data might become completely unreliable, leading to a loss of money, time, and knowledge.
The data needs to be absolutely ironclad to guard against false conclusions on which further, flawed research may be carried out. It’s vital that any errors or faults that can happen are detected and recovered to ensure that the collected data is uncorrupted.
Rad-Hard Chips Build a Foundation for the Future
Radiation-hardened chips, also known as rad-hard chips, can withstand the severe environment in space and have the potential to bring the convenience of modern computing to the galaxy’s most unforgiving frontiers.
As a well-established technology, radiation hardening primarily reduces faults from occurring to improve chip reliability. That doesn’t mean they don’t present their own specific challenges. Rad-hard technology lowers the faults caused by extreme environments via special technology process libraries that can withstand radiation effects. However, compared to commercial technology, the cost of development is higher in addition to compromising on key metrics such as power, performance, and area.
Resiliency techniques offer a different path forward, where faults are embraced but the system recovers from it. Though this means lower system availability, there’s also no need for specialized process libraries. The question then becomes: How and where should we deploy resiliency techniques instead of full-blown rad-hard technology? The answer very much depends on mission profile and features present in the system.
Putting Commercial Components to Work in Space
In the most optimistic scenario, we can use the parts we have. In 2017, NASA sent an HP supercomputer into space to test how standard hardware and software would hold up on a Mars mission. Eleven months later, the computer still functioned, at least without any major glitches.
The reason for the seeming durability isn’t just luck—similar results have been observed in fault-injection experiments simulated in research labs. Most of the faults injected don’t cause system failures largely because of masking behavior and application characteristics. But later when faced with complex workloads, failures show up and they can be catastrophic.
Both rad-hard technology and resiliency techniques help with reducing failure. Resiliency techniques are better positioned to tackle the bit flip problem. That’s because the system recovery process will enable us to monitor the health of the system eliminating concerns about data corruption.
Exploring a Challenging Atmosphere
Most of today’s research takes place in low-Earth-orbit space applications like communication, observation, and commercial travel. At this level of orbit, radiation exposure is lower than it would be on Mars, but far more severe than it would be just a stratosphere away. With aspirations for satellite and transportation infrastructure to last longer than a single space mission, researchers are dedicating themselves to long-term chip dependability.
That means modern researchers are focusing on hardening for total-ionizing-dose radiation, otherwise known as the amount of radiation the device should be able to withstand in its lifetime before something fails. Usually, a good benchmark is 100 kilorads of total-dose radiation hardness. Radiation hardening creates more robust electronics that, when coupled with system resiliency techniques like Arm Triple-core lockstep, can lead to more reliable electronics.
A number of experimental programs have explored the viability of rad-hard Arm-based chips. These trials set about to prove whether modern CMOS technology could cope with single-event upset, total ionization dose, and related latch-up.
The ongoing High-Performance Spaceflight Computing (HPSC) program at the University of Michigan, funded by NASA and the U.S. Air Force, seeks to dramatically advance the state of the art for spaceflight computing, developing highly power-efficient and fault-tolerent systems. As part of this program, a Next Generation Space Processor (NGSP) analysis program engaged industry to define and benchmark future multicore processor architectures. The reference design features eight radiation-hardened Arm Cortex-A53 64-bit processor cores described as “chiplets”—building blocks for a space-capable computer. Following a competitive procurement, the contract was awarded to Boeing, with deliverables due in April 2021.
Race to Make Better Semiconductors
As new demand for radiation-hardened technology emerges in the coming years, researchers are studying ways to build sturdier chips that can withstand the harsh environment in space.
Some enhancements occur as the technology improves for standard chips. For example, as the average technology node size decreases from 90 to 14 nm, its increased tolerance for total-ionizing-dose radiation follows. But at the same time, physically smaller chips are more vulnerable to single-event upsets at the gate and transistor level. The key to solving this new problem depends on new breakthroughs in the competitive semiconductor industry.
On Earth, semiconductors like silicon are an essential building block in any computer chip. In the rad-hard industry, researchers are experimenting with new materials like gallium nitride, silicon carbide, diamond, and graphene to significantly increase reliable conductivity in harsh environments. These emerging materials are inherently radiation-hardened, but costly and far less in demand than the standard silicon that consumer products like iPhones rely on.
Resiliency + Radiation Hardening = Better Electronics in Space and on Earth
It’s possible that changing conditions on Earth may one day make rad-hard technology as applicable to our daily lives. Due to factors like heightened solar radiation, there’s more radiation on Earth today than our ancestors dealt with. In fact, there are already harsh environments on Earth that could benefit from rad-hard chips. For instance, they could help easily monitor the Fukushima radioactivity zone and other disaster areas, or be applied to radiation imaging in the healthcare industry for longer-lasting X-ray equipment that would require less maintenance and upkeep.
Exploring rad-hard chip applications on Earth actually will allow us to use them to their highest potential in space. Exploiting the endless number of uses can increase demand, making innovation and mass production cheaper. This, in turn, would make it lucrative for researchers to improve their low-Earth orbit reliability, and make space applications more dependable. With transistors only a few nanometers across, these diminutive chips can transform the universe—and our homes.
Balaji Venu is a Staff Research Engineer and Eric Van Hensbergen is a Fellow in the Research Division at Arm.