In the big data era, the digital commons (and associated data and control) are increasingly controlled by a small number of platforms that intermediate digital interactions. Blockchains are an emerging paradigm for permissionless distributed computing, that can remove the intermediary while at the same time providing trusted computing and storage guarantees. While applications of blockchains beyond cryptocurrencies into generalized digital platforms, are still in their beginning stages, blockchain technology is expected to create several trillion dollars of value in the next few years, potentially aiding the democratization of trust, data and control in digital platforms. In our research, we take a first-principles, full-stack approach to blockchains: we view blockchains as a whole integrated system involving networking, distributed consensus, data storage, incentives, and applications. We explore the foundations of how each of these layers are designed (these foundational design principles as well as specific algorithms are the subject of cutting-edge research). A central feature of these blockchain protocols is to tolerate adversarial users as well as rational users. While much pioneering work has happened in distributed computing on tolerating adversaries, it has focused on the permissioned setting, where there is a permission authority that decides which nodes participate in the protocol. We focus on the permissionless setting without a central permissioning authority. Proving security results in this setting involves understanding convergence of certain random processes on trees (including branching random walks) under adversarial actions. Random processes plays a central role in the understanding of blockchains. Our research entails ideas from a variety of theoretical areas including, but not limited to, distributed computing, information theory, queuing theory, networking, coding theory and game theory in designing these systems successfully.

News

 July, 2022 Sreeram joined Kevin Sekniqi and Logan Jastremski to discuss on Security models & Ethereum PoS. Check the podcast here. Order-Fair Consensus in the Permissionless Setting won the Best-Paper Award at ACM APKC 2022. Sreeram joined Anna Rose and Tarun Chitra at Zero Knowledge Podcast to discuss on his research on blockchains and its interaction with information theory. Check the podcast here. Two talks on Themis: Fast, Strong Order-Fairness in Byzantine Consensus and BFT Protocol Forensics accepted at The Science of Blockchain Conference 2022 (SBC'22) Check out our recent work on Minotaur: Multi-Resource Blockchain Consensus. Check out our recent work on Babylon: Reusing Bitcoin Mining to Enhance Proof-of-Stake Security. Our work on Free2Shard: Adaptive-adversary-resistant sharding via Dynamic Self Allocation has been accepted for presentation at ACM Sigmetrics 2022. BFT Protocol Forensics won the Best-Paper Award Runner Up at ACM CCS 2021. Soubhik gave a talk on PoSAT: Proof-of-Work Availability and Unpredictability, without the Work at ConsensusDays 21 organized by Protocol Labs. Sreeram gave a tutorial on Blockchain protocols made efficient and scalable at Twenty Seventh National Conference on Communications (NCC-2021). Sreeram spoke at the National Academy of Engineering US-Japan meet on Blockchain: introduction and challenges''. Sreeram gave a talk on fair ordering for permissionless blockchains at Sixth London Symposium on Information Theory. Soubhik gave a talk on PoSAT: Proof-of-Work Availability and Unpredictability, without the Work at Cornell . Our work on BFT Protocol Forensics has been accepted for presentation at ACM CCS 2021. Our work on Securing Parallel-chain Protocols under Variable Mining Power , which was co-lead by Viswa, has been accepted for presentation at ACM CCS 2021. Our work on Blockchain CAP Theorem Allows User-Dependent Adaptivity and Finality was presented at FC 2021. Our work on ACeD: Scalable Data Availability Oracle, which was co-lead by Bowen, was presented at FC 2021. Our work on PoSAT: Proof-of-Work Availability and Unpredictability, without the Work was presented at FC 2021. Our work on Everything is a Race and Nakamoto Always Wins was presented at ACM CCS 2020. Our work on Perigee: Efficient Peer-to-Peer Network Design for Blockchains was presented at ACM PODC 2020. Our work on PolyShard: Coded Sharding Achieves Linearly Scaling Efficiency and Security Simultaneously has been accepted for publication at IEEE Transactions on Information Forensics and Security. Sreeram gave a talk on (insert topic) at Stanford Blockchain Conference 2020. Our work on Coded Merkle Tree: Solving Data Availability Attacks in Blockchains was presented at FC 2020. Our work on Prism: Deconstructing the Blockchain to Approach Physical Limits was presented at ACM CCS 2019. Excited to have Bowen, Robert, Soubhik and Viswa join UW Blockchain Lab for their graduate research. Welcome all of you!