High Performance Memory For The World Computer

on YouTube ↗18m

" summary

Optimum ships a memory stack — Optimum P2P, DROM and DAM — built on Random Linear Network Coding (RLNC) that composes equations at intermediate nodes instead of forwarding shards. Testnet benchmarks show Optimum P2P cuts average latency and its standard deviation versus GossipSub, with the largest gains under congestion where gossip latency spikes nonlinearly. The system runs live with validators on board, improves bandwidth utilization and reduces transaction-delays and gas-fee exposure, and the encoded, decentralized propagation proves functional in production tests.

AI-written summary · about

" chapters (17)

1. 0:00Introduction
2. 0:31Speaker Background & Motivation
3. 1:22Tech Transfer from MIT & Optimum’s Mission
4. 2:02Defining Memory in Decentralized Systems
5. 3:11The Role of Memory in Computation & Web3 Stack
6. 5:28Challenges: Bandwidth, Storage, and Scaling
7. 6:09Optimum’s Product Suite Overview
8. 6:32Random Linear Network Coding (RLNC) Introduction
9. 7:13Performance Benchmarks & Testnet Results
10. 8:45Nonlinear Network Behavior & Congestion
11. 10:10Benefits Under Network Stress
12. 10:35How RLNC Works
13. 12:15Unique Features: Equation Composition & Decentralization
14. 13:22Impact: Bandwidth, Latency, and Tradeoffs
15. 15:22Who Benefits: Validators, dApps, End Users
16. 16:51Collaboration & Specialization in Blockchain Infrastructure
17. 17:50Conclusion

" related talks

• 13mBuilding the Fastest Decentralized Internet Protocol for Web3Muriel Medard · Cosmoverse 2025
• 25mDisaggregating Persistent Memory and Controlling Them RemotelyYizhou Shan, Yiying Zhang, Shin-Yeh Tsai · USENIX ATC '20
• 45mBeyond Machine Boundaries: Experience with Memory DisaggregationFelix Eberhardt, Andreas Grapentin · OpenPOWER Summit NA 2021
• 17mTopoOpt: Co-optimizing Network Topology and Parallelization Strategy for Distributed...Ying Zhang, Zhihao Jia, Manya Ghobadi, Dheevatsa Mudigere, Weiyang Wang, Moein Khazraee, Zhizhen Zhong, Anthony Kewitsch · NSDI 2023
• 12mCharacterizing and Optimizing Remote Persistent Memory with RDMA and NVMRong Chen, Haibo Chen, Xingda Wei, Xiating Xie, Binyu Zang · USENIX ATC '21
• 16mOneSidedMW: Managing Disaggregated Memory Efficiently, Flexibly, and Securely with RNICYubin Xia, Xingda Wei, Zixuan Wang, Jinyu Gu · NSDI '26
• 15mMore Than Capacity: Performance-oriented Evolution of Pangu in AlibabaChao Han, Qiang Li, Qiao Xiang, Jiesheng Wu, Jiwu Shu, Jiaji Zhu, Yuanyuan Dong, Zhaosheng Zhu, Yaohui Wu, Shaozong Liu, Zhongjie Wu, Junping Wu, Zheng Cao, Jinbo Wu, Lulu Chen, Shuo Huang, Wenhui Yao, Shanyang Liu, Huayong Wang, Haonan Qiu, Derui Liu, Zhiwu Wu, Zicheng Luo, Yuchao Shao, Gexiao Tian, Yuxin Wang, Haohao Song, Ridi Wen, Shuqi Zhao · FAST '23: 21st USENIX Conference on File and Storage Technologies
• 28mPersistent Memory plus RDMA, new age remote deviceXiao Yang · linux.conf.au 2022