Privacy takes center stage in this week’s episode as Jorrin Bruns (Support Engineer, Parity Technologies) talks with Marvin Tong, co-founder and CEO of the Phala Network, a trustless, privacy-preserving cloud computing network based on Substrate. They dive deep into Phala’s tech, exploring how Phala tackles the issue of trust in the computational cloud by shielding user data from centralized organizations like Google, with the same level of computational power as existing cloud services.
Tong details several topics, including node security, Phala’s Substrate-based runtime and bespoke pallets, integrating alternative trusted execution environment (TEE) hardware, and how it tackles consensus. They delve into on-chain data privacy, deploying private smart contract computation for decentralized applications (DApps) and decentralized finance (DeFi) and services on Phala Network including Web3 Analytics and cross-chain interoperability.
01:10 What is Phala Network?
04:40 Privacy in decentralized cloud computation
06:54 Scalability through TEE
09:53 Using multiple TEEs to mitigate trust
15:10 Integrating alternative hardware
17:55 Phala’s Substrate runtime
25:49 Consensus within Phala Network
30:54 Computation on TEE
35:44 Use cases for projects running on Phala
44:41 Bridging Diem (formerly known as Libra) to Polkadot
55:36 Targeting the second parachain slot
“Separating computation from consensus is the key to how Phala can bring the benefits of blockchain while delivering computational power on the scale of a cloud server.”
“We think there are two major types of smart contracts, for now, the first is EVM, and the other is written by ink! or Rust language running in Wasm, WebAssembly. WebAssembly is the type we chose. We want to put WebAssembly inside of TEE so that we can support this kind of security technique grade. We think it will be a major choice for not only Polkadot but the whole industry for Web3.”
“Using Web3 Analytics, all of the analysis code is running as a confidential smart contract. This means that the data is not deployed by a centralized server like Google or any other big platform. Using Google Analytics means that you trust to put your user’s data node to Google so that they can analyze it for you. From this perspective, on Web3 Analytics, the data is not trusted by any centralized server or centralized gatekeeper, it is just encrypted by Phala’s system, and how to encrypt it is in the visitor or users hands.”
“We believe that in the next three years, or five years, the major technology of all computation cloud will be confidential computing cloud. That’s what we can see from what Google, Amazon, and Facebook are doing.”