BlockSTM: Scaling Blockchain Execution by Turning Ordering Curse to a Performance Blessing – Alex S

Alexander Spiegelman obtained his PhD from the Technion in 2018, where he worked with Idit Keidar on Distributed Sytems, focusing on reliable storage, reconfiguration, consensus, and concurrency. He has received best paper awards at PLDI 2016 and OPODIS 2019.
Alexander worked as a research scientist at VMware Research Group and the at Novi Research where he worked on Asynchronous BFT and Blockchain scalability.

BlockSTM is a parallel execution engine for smart contracts, built around the principles of Software Transactional Memory. Transactions are grouped in blocks, and every execution of the block must yield the same deterministic outcome. BlockSTM further enforces that the outcome is consistent with executing transactions according to a preset order, leveraging this order to dynamically detect dependencies and avoid conflicts during speculative transaction execution. At the core of BlockSTM is a novel, low-overhead collaborative scheduler of execution and validation tasks.

BlockSTM is implemented on the main branch of the Diem Blockchain code-base. The evaluation demonstrates that \BSTM is adaptive to workloads with different conflict rates and utilizes the inherent parallelism therein. BlockSTM achieves up to 130k tps in the benchmarks, which is a 17.5x improvement over the sequential baseline with 32 threads. The throughput on a contended workload is up to 50k tps.

Related paper: https://arxiv.org/abs/2203.06871

Make sure to give our Twitter account a follow! @ProtoResearch https://twitter.com/ProtoResearch

Interested in hearing about future talks, funding opportunities, and more? Sign up for our mailing list:
https://mailchi.mp/protocol/research-newsletter-signup

Update

Steve Rich's Exciting New Book: A Journey into the World of Forex Trading!

Interview

To suggest a talk topic or speaker, or if you are interested in giving a talk yourself, please email us or open an issue at github.com/protocol/research/issues/new/choose

Want to learn more about Protocol Labs Research? Visit our website:
https://research.protocol.ai/

Interested in working at Protocol Labs? Check out our job board:
https://boards.greenhouse.io/protocollabs