Quside Blog
Can my code be accelerated by an RPU? Part 2: Verticals

5 de abril de 2023
3.4 min read

Part 2: Six industries where an RPU can boost performance

In our previous post, we introduced you to Quside’s Randomness Processing Unit (RPU) and outlined how you could use it to accelerate your stochastic workloads, whatever type they may be. In this post, we will explore different business verticals and provide some examples of specific workloads that can be accelerated efficiently with an RPU so that you can make the most of this device and its ecosystem.


If you are familiar with the finance industry, you know Monte Carlo simulations – a widely used tool for modeling market behavior and evaluating risk. These simulations involve generating a huge amount of random scenarios and using them to simulate the behavior of financial instruments or portfolios over time.

Having an RPU, you can accelerate these simulations, enabling you to make faster and more informed investment decisions. For example, you could quickly evaluate the risk of a new investment opportunity, determine whether it is a good fit for your portfolio, or assess the impact of different economic scenarios on your portfolio, thus making more informed decisions about resource allocation.


If you work in cryptography and cybersecurity, you know that randomness plays a critical role in many applications, such as generating encryption keys and creating secure passwords. The RPU, as it integrates a high-speed, high-quality Quantum Random Number Generator (QRNG) from Quside, enables fast generation of random numbers, thus allowing you to generate secure encryption keys faster and more efficiently while improving the overall security of your applications.

Additionally, you could use the RPU to accelerate your full cryptographic workload. That way, you can use it to create more secure passwords and reduce the processing time required for cryptographic operations, allowing your applications to run faster and more efficiently.


When working in logistics some of the most difficult challenges are those related to optimizing routing, supply chain management, and demand forecasting. Stochastic models often address these challenges but can be computationally expensive and time-consuming. By accelerating these models with an RPU, you could optimize your logistics operations more efficiently, reducing transportation costs and improving delivery times. For example, you could use accelerated stochastic models to predict demand and inventory levels more accurately, thus improving inventory management and procurement.


Many simulations, such as molecular dynamics simulations or weather forecasting models, use randomness to deal with the inherent uncertainty of these processes. Running these simulations gets expensive very quickly, behaving as a limiting factor for the scope of your research.

Using an RPU, you could use accelerated simulations to more accurately model weather patterns or predict the impact of climate change, improving our understanding of these critical phenomena. Also, you could use these devices for scenario generation in areas such as materials science or biochemistry.


Some ML algorithms involve stochastic or randomized components, such as stochastic gradient descent or randomized feature selection. Running these algorithms can be computationally expensive and time-consuming, limiting the size and complexity of the models you can train.

Using an RPU can potentially enable training larger-scale models or conducting more experiments in parallel, enabling you to explore a broader range of models and achieve better performance, leading to more accurate predictions, and enabling you to develop better products and stay ahead the competition.


Some games depend on processes, such as procedural generation or AI behavior, often involving stochastic or randomized components.

Even in this sector, an RPU can be helpful: combining its high-performance randomness with an accelerated algorithm to generate realistic terrain for an open-world game enables the creation of larger, more detailed environments for players to explore. Additionally, you could use accelerated AI algorithms to create more intelligent and engaging non-player characters, improving the overall gameplay experience.

In summary, if you work in any of these industries and are looking for ways to improve your performance, differentiation, accuracy, or work scope, consider the benefits of using an RPU to accelerate your stochastic workload. By exploring the RPU and its ecosystem, you could gain valuable insights and stay ahead of the competition in any of these fast-paced industries.