In an era where digital security is paramount, BTQ Technologies is at the forefront of revolutionizing the future of quantum-safe cryptography. A key part of their technological innovation is the integration of Processing-In-Memory (PIM), which enables faster, more efficient cryptographic processing. This post explores how BTQ’s CASH architecture, leveraging PIM, is setting new standards for performance, security, and efficiency, especially in the realm of quantum computing.

Introduction to Processing-In-Memory (PIM) and CASH Architecture

BTQ's CASH architecture is a secure, agile, and reconfigurable cryptographic accelerator that depends on Processing-In-Memory (PIM) strategies to achieve its performance. This architecture has some major advantages over other architectures: simplicity of design, low area and power requirements, high throughput, and flexibility to tackle emerging cryptography and even other acceleration tasks.

Processing in memory is a strategy that moves some compute out of the processor and into the memory. Instead of a processor’s more powerful instructions that operate on a small set of data, processing in memory uses simple operations that operate on entire rows of memory at once, and these operations are extremely local, often operating on single bits in parallel. Instead of running a dedicated multiplier circuit on each of a set of operands, for example, the multiplier circuit is executed gate-by-gate on the bits of the input. This dedicated multiplier’s circuitry is mostly not reused; with processing in memory, however, this circuitry is reused in many other operations.

This strategy is very capable for small circuits performed many times in parallel. This applies to any highly parallel computation, but also to the bitwise operations in many cryptographic algorithms.

A hardened microprocessor is able to perform all of these tasks, but its generality limits its throughput, and much of its compute circuitry sits idle. Post-quantum cryptography demands substantially larger working sets in memory, which means much more of the computation is spent waiting for the bus. Processing in memory solves these problems by reducing the volume of data that needs to move into or out of memory, and having simple, bitwise logic that gets used in many ways.

Understanding Processing-In-Memory (PIM)

At its core, Processing-In-Memory aims to minimize the bottleneck caused by transferring large sets of data between the memory and processor. Traditionally, processors carry out complex computations on data in memory. However, the speed of this process is often hindered by the need to move data in and out of memory constantly.

With PIM, the computational logic is placed closer to the memory storage itself. This allows the system to perform simple operations directly within the memory without requiring large-scale data transfers to the processor. In particular, PIM excels in handling highly parallel tasks, where multiple operations are needed simultaneously, such as the bitwise operations fundamental to many cryptographic algorithms.

By processing data in memory, PIM makes computations faster, more energy-efficient, and less reliant on large, complex accelerators. This is particularly important for cryptography, which often involves performing repetitive and parallel bitwise operations.

Advantages of BTQ Technologies’ CASH Architecture

BTQ’s CASH architecture is designed to meet the unique demands of quantum security, where high throughput and low latency are essential. Here’s how CASH stands out:

• Security and Agility: CASH is a secure cryptographic accelerator that is agile –it can adapt to new cryptographic standards, especially other post-quantum algorithms. This makes it well-suited for the evolving landscape of quantum-safe encryption.
• Reduced Area and Power Requirements: Traditional processors require extensive hardware to perform cryptographic operations, consuming more space and power. In contrast, CASH reduces the need for complex dedicated circuits, using memory-centric processing to lower both area and power consumption.
• High Throughput: The processing power of CASH comes from leveraging parallelism and locality. With PIM, numerous simple operations can be performed simultaneously on large datasets, resulting in a dramatic increase in throughput—improving performance on large sets of encrypted data, and improving latency.
• Flexibility for Cryptographic Tasks: CASH isn’t just designed for one type of computation. Thanks to its reconfigurability, it can handle various cryptographic tasks, including post-quantum cryptographic operations that demand substantial memory resources.
• Side-channel resilience: Bitwise parallelism on wide words automatically obfuscates some of the signal from side-channel leakage present in computations on other devices. Simplifying the set of instructions also dramatically simplifies the effort required to blind the logic gates. Focusing on bit-serial computation means that Boolean masking is intrinsically applicable to algorithms written natively for bit-serial computation.

Why Quantum Security Demands PIM

Post-quantum cryptography (PQC) uses fundamentally different hard problems, with fundamentally different computational structures. Quantum computers have the potential to break conventional cryptographic methods, demanding the development of new algorithms that can withstand quantum attacks. These post-quantum algorithms require significantly larger working sets and higher computational power, and they are unlikely to get dramatically more efficient in the near future.

One of the challenges in PQC is the need for larger data sets to be processed quickly. Traditional hardened microprocessors, designed for general-purpose computing, struggle to manage the high demand for parallel processing and the need to handle large volumes of data efficiently.

With PIM, these limitations are addressed. PIM allows for computations to be performed directly within memory, reducing the need to move data back and forth between the processor and memory. This significantly cuts down on bus use and memory bottlenecks. PIM also improves memory use substantially, by sizing the structures in memory to the actual bit length of the parameters in question..

PIM, in essence, directly addresses the problems caused by increasing the size of cryptographic keys, signatures, and ciphertexts. 

Real-World Applications and Industry Impact

The implementation of PIM-enabled quantum security solutions has far-reaching implications across various industries:

• Financial Sector: For financial institutions, where security and efficiency are paramount, PIM accelerates transaction processing, particularly for cryptographic protocols and blockchain-based systems.
• Defense: Military and defense applications require ultra-secure communications. With PIM’s speed and low-power nature, encrypted communication networks can operate more securely and without the overhead of traditional secure systems. With enhanced side-channel hardening, PIM accelerators can deliver highly secure, performant solutions at a much lower cost.
• Telecommunications and Infrastructure: Telecoms rely on secure communications infrastructure, which is increasingly vulnerable to quantum attacks. PIM-based cryptography acceleration offers robust encryption solutions that are scalable for massive data handling.

PIM acceleration can provide these and many other industries with enhanced latency reduction, increased efficiency, and better overall security—transforming how data is handled and protected.

Future Prospects

As quantum security continues to evolve, the potential of PIM solutions to accelerate cryptography is enormous. BTQ Technologies is committed to advancing PIM technology and further optimizing its use in quantum-safe solutions. The roadmap includes:

• Further integration of PIM into more quantum-safe products to ensure the most robust security solutions for businesses and governments.
• Development of next-generation quantum accelerators that can scale with increasing demands for both security and computational power.

BTQ’s continuous innovation in PIM-based quantum security will play a key role in shaping the future of post-quantum cryptography, ensuring industries are equipped to handle the challenges of tomorrow.

Conclusion

Processing-In-Memory (PIM) is a transformative technology that is reshaping how we approach cryptographic computations, especially in the era of quantum computing. BTQ Technologies’ CASH architecture leverages this strategy to deliver secure, high-throughput, and energy-efficient quantum-safe solutions. By integrating PIM into its quantum security products, BTQ is pushing the boundaries of what’s possible in securing data against future quantum threats.

As the world moves toward a quantum future, BTQ remains at the cutting edge, committed to delivering innovative security solutions that will protect critical data for generations to come.

Stay tuned for more updates on BTQ’s advancements in quantum security as they continue to lead the charge in next-generation cryptographic technologies.

Looking to join the team at BTQ Technologies? We’re hiring!

BTQ is a fast-growing technology company led by experienced builders and industry pioneers. We’re looking for individuals who are passionate about the intersection of quantum computing and cryptography to join our world-class team today. Visit our careers page here.