Phala Network is a decentralized blockchain protocol that acts as a co-processor for other decentralized applications. It leverages Trusted Execution Environments (TEEs) and integrates them with blockchain technology to improve decentralized applications’ overall efficiency, security, and scalability. The process is achieved through smart contracts that enable trustless off-chain computations known as Phat Contracts. These Phat Contracts allow the developers to offload the more intensive computational tasks like data processing and AI inferences to decentralized TEE nodes (CPUs). Since launching in 2021, Phala has scaled to over 30,000 TEE nodes.

Increased Phala Offerings

The partnership with io.net was first announced in September 2024 and has allowed the Phala Network to expand its offerings into more advanced GPU-driven tasks. Phala is able to tap into over 300,000 decentralized GPUs & CPUs across 130 countries on the io.net network, including the more advanced Nvidia H100 and H200 GPU models. io.net benefits from having another partnership with a highly scalable network that is actively building the infrastructure for future decentralized AI projects. With Phala’s privacy-preserving TEEs, io.net also benefits from integrating its GPU network while allowing AI and ML developers to maintain that same level of data privacy.

More Affordable TEE Services

The other major benefit of the partnership to the Phala Network is the cost savings of having a decentralized cloud provider like io.net. Phala can access highly scalable GPU processing at a fraction of the cost compared to centralized providers. This makes the Phala network more efficient with its resources and opens up the opportunity for offering more advanced cloud compute while passing on more attractive prices to AI and ML developers looking to utilize its decentralized coprocessor model. It ensures that developers looking to access secure decentralized computing are not cost-prohibited.

The Future of Decentralized Coprocessing

Both io.net and the Phala network continue to expand their decentralized networks, increasing the network effect of each. Future plans for the partnership involve joint research, testing, and benchmarking Nvidia’s latest GPU models. The partnership is currently at the forefront of scaling decentralized applications through coprocessing. It offers a novel alternative to the cost-prohibitive centralized cloud providers and is improving decentralized AI applications’ overall efficiency, security, and scalability.

‍