Cloud-based AI and Edge Computing Advancements

Cloud-based AI and Edge Computing Advancements

The integration of Artificial Intelligence (AI) into edge systems is revolutionizing data management and creating a surge in demand for flash memory.

KIOXIA, a company specializing in hardware that supports AI, is at the forefront of this transformation. The company's hardware plays a crucial role in data management, enabling quick and reliable local data storage and access for edge AI applications.

According to Axel Stoermann, the Chief Technology Officer and VP at KIOXIA Europe, the role of supporting infrastructure technologies like flash memory is becoming increasingly important due to the expansion of AI into edge systems. In an interview on the Inside Electronics Podcast, Stoermann discussed this trend and shared insights about the industry.

The podcast, part of a series called "TechXchange," explored the growing demand for AI in both central server and edge applications in the cloud. The episode was titled "AI on the Edge," and focused on the latest developments in the electronics industry.

Edge AI, which embeds intelligence directly on or near data sources, is enabling real-time, context-aware decision-making without relying on cloud connectivity. This reduces latency, enhances privacy and security, and mitigates bandwidth bottlenecks—key requirements in safety-critical and resource-constrained environments.

By 2025, over 55% of deep neural network data analysis is expected to occur at the edge, a substantial increase from less than 10% in 2021. This shift reflects exponential growth in data generated at edge locations, projected to reach 2,763 exabytes by 2026. Processing data locally requires sophisticated hardware capable of managing complex AI workloads within power and size constraints, often supported by specialized components like FPGAs and AI solution stacks optimized for embedded systems.

This surge in edge AI activity intensifies demand for flash memory, which is crucial for fast, reliable local data storage and access in these devices. Flash enables quick data retrieval and persistence for real-time analytics, training subsets, and inferencing, all within the limited hardware footprint of edge systems.

The industrial automation, automotive, manufacturing, energy, and healthcare sectors are among the most active adopters of edge AI solutions that rely heavily on robust flash memory subsystems to support AI workloads and data storage needs. From a data management perspective, edge AI facilitates smarter data processing strategies by filtering, summarizing, and analyzing data locally, sending only essential insights to the cloud. This reduces network load and operational costs while maintaining high responsiveness and security commitments.

Explainable AI (XAI) techniques are also being integrated at the edge to ensure outputs are interpretable and compliant with emerging regulations, adding another layer to data management complexity.

In summary, the growing integration of AI into edge systems is drastically transforming how data is processed and stored, driving substantial growth in flash memory demand, and enabling more intelligent, autonomous, and secure devices closer to where data is captured. This trend is accelerating across multiple industries and is expected to continue expanding through 2025 and beyond. The demand for next-generation flash memory to store this valuable data is growing, and companies like KIOXIA are at the forefront of this exciting development.

[1] "Edge AI: The Future of Artificial Intelligence," KIOXIA Corporation, 2021. [2] "Edge AI Market and Technology Forecast 2021-2026," ABI Research, 2021. [3] "Explainable AI: The Need for Transparency in AI Systems," Forbes, 2020. [4] "Data Management Strategies for Edge AI," TechTarget, 2021.

1. KIOXIA's hardware, which supports AI, is integral to the surge in demand for flash memory due to the integration of Artificial Intelligence into edge systems, a phenomenon discussed in a podcast episode titled "AI on the Edge" from the series "TechXchange."
2. With over 55% of deep neural network data analysis expected to occur at the edge by 2025, and the growing demand for edge AI solutions across several sectors, the need for flash memory for fast, reliable local data storage and access in these devices is increasing, as seen with companies like KIOXIA specializing in hardware that aids in data and cloud computing, and technology like artificial intelligence.

Read also: