TREEBLOCK.IO. Version Author: Anh Le, Advisor: Dr. Rex Yeap. Creating a high performance IoT Blockchain Network

Size: px

Start display at page:

Download "TREEBLOCK.IO. Version Author: Anh Le, Advisor: Dr. Rex Yeap. Creating a high performance IoT Blockchain Network"

Duane Sanders
5 years ago
Views:

TREEBLOCK.IO Version 1.3.2 Author: Anh Le, Advisor: Dr.

1 TREEBLOCK.IO Version Author: Anh Le, Advisor: Dr. Rex Yeap Creating a high performance IoT Blockchain Network Using Time Sharding Tree Architecture

2 Prominent Value-Propositions 1. Multi-Millions TPS with InterTree Architecture 2. Unlimited Scalability with Time Sharding Trees 3. Small Storage and Hardware Requirement 4. Divide Benefit Equally 5. ZERO FEE Tokenomy with New Tokenomic Relationships 6. Secure for Cross-Shard Transactions 7. Can be Used to Develop Any Type of Dapps 8. Durability by mdpos Consensus and Backup Nodes 9. Self-Defend from Attacks with The AnyShard as The AnyCast Network 10. Radically Decentralized with Billions IoT Devices

3 Table of Content 1. Hierarchy Sharding 9. Digital Signature Algorithm 2. Time Sharding 10. Virtual Machine & Smart Contract 3. Hierarchy + Time Sharding 11. InterTree + InterChain 4. TreeBlock's Nodes - Voting Tokenomy 12. Topologies AnyCast Coordinator Layer- Competitors 13. Token Circulation Flow & Community Fund 5. Block Producers Mgmt App 6. Consensus Algorithm 14. Applications 7. Producing Block Mechanism 15. Token Price & Allocation, Fund Distribution 8. Transaction Model 16. Roadmap

4 1. Hierarchy Sharding Concept Ex: 1 Shard with 4 Sub Shards Sub 1 Sub 2 Sub 3 Sub 4 1 Sub Shard with 4 Sub-Sub Shards Sub-Sub 1.1 Sub-Sub 1.2 Sub-Sub 1.3 Sub-Sub 1.4

5 1.1 Hierarchy Sharding's Advantage Fast Synchronization Cross Check Faster Search Faster Less Data on per Shard Can Be Dividing More Shards

6 1.2 Cross Check & State Synchronization amongst Shards Ex: with 4 Shards, we need to check = 4! : (2!2!) = 6 pairs k=2, n=4

7 1.3 Cross Check & State Synchronization with Horizontal Sharding Ex: with 100 Shards, we need to check 100! : (2!98!) = 4,950 pairs What about 1,000 Shards or more like other existing Blockchains? 1,000! : (2!998!) = 499,500 pairs

8 1.4 Cross Check & State Synchronization with Hierarchy Sharding Ex: If we divide 100 Shards into 10 Main Shards and each Main Shard has 10 Sub Shards, so we have [(10!:(2!8!))*10] + [10!:(2!8!)] = 495 pairs Can we divide more levels? Yes!

9 1.5 Cross Check & State Synchronization With Hierarchy Sharding Ex: And, If we divide 100 Shards into 5 Main Shards and each Main Shard has 5 Sub Shards, each Sub Shard has 4 Sub-Sub Shards [(4!:(2!2!))*5*5] + [5!:(2!3!)] = 170 pairs Much more effective! How do we divide shards with Hierarchy Architecture?

10 2. Time Sharding Concept We use Tx's timestamp

11 2.1 Time Sharding Concept Ex: Case 1 We divide 1 Second into 4 Segments 1 Segment = 250,000 mili-sec 1st /4 2nd /4 3rd /4 4th /4

12 2.2 Time Sharding Concept Ex: Case 2 We divide 1 Second into 100 Segments 1 Segment = 10,000 mili-sec 1st 2nd th

13 2.3 Time Sharding Concept Ex: Use case 1 with 52 Tx in 1 Sec 1 Time Segment = ¼ Sec 11 Tx 16 Tx 15 Tx 10 Tx 1s 1s 1s 1s

14 2.4 Processing Tx in 1 Second with Time Sharding T1 T2 T4 T3 11 Tx 16 Tx 15 Tx 10 Tx Pool 1 Pool 2 Pool 3 Pool 4 1 Pool = Groups of Nodes

15 2.5 Processing Tx in 1 Second Without Time Sharding 52 Tx Group 1 Group 2 Group 3 Group 4 Current Sharding Solutions

16 2.6 Normal Sharding vs Time Sharding Time Sharding Normal 52 Tx Group 1 Group 2Group 3 Group 4 vs 10 Tx 15 Tx 16 Tx 11 Tx

17 3. Hierarchy Sharding needs Time Sharding FOR Rules to Divide Shards Rules for Input Rules for Output Rules for Search Rules for Cross-Check Rules for Synchronization

18 3.1 Hierarchy Sharding needs Time Sharding Ex: 1 Main Shard has 4 Sub Shards for 4 Time Segments 1st /4 2nd /4 3rd /4 4th /4 Each Sub Shard has 4 other Sub-Sub Shards for other 4 Time Segments 1st /16 2nd /16 3rd /16 4th /16

19 3.2 Time Sharding Tree Hierarchy + Time Sharding = Time Sharding Tree 1 Time Sharding Tree = Multiple Levels of Time Segment Shards = 1 Second. In theory, 1 Time Sharding Tree could have 1 Million Shards in hierarchy architecture or more depending on the time unit. A Time Sharding Tree may be able to produce thousands of blocks to millions of blocks per "block time". We can scale to have a Green Planet with multiple billions of Time Sharding Trees to serve the entire world.

20 3.3 Time Sharding Tree Concept Note: 1/64 = 1st/64, 2/64 = 2nd/64, 1/16 = 1st/16, 2/16 = 2nd/16

21 4.1 TreeBlock's Nodes

22 4.2 Voting Select Top Voted Candidates

23 4.3 Tokenomy for ZERO FEE

24 5. Block Producers Mgmt App Easy Set Up a Block Producer Easy to manage Fund & operate a Block Producer Transparent amongst Block Producer and Voters

25 6. Preferred Consensus Algorithm DPoS Definition: One Delegated Node produces Block at a time slot Ex: DPoS = 11 Node 6/11 Nodes Consensus Block

26 6. Introduce & Suggest m-dpos = multiple DPoS Definition: Many Delegated Nodes compete to produce Block at a same time slot Ex: m-dpos = 3-11 Nodes 6/11 Nodes Consensus H/A with 3/11 Nodes Block Faster, Higher Consensus, Bigger Stake Win Byzantine

27 7. Producing Block Paradigm With Time Sharding 2 Ways to Produce Block with This Architecture PARALELLY vs SEQUENTIALLY

28 7.1 Parallelly Producing Block With Time Sharding Example: BlockTime is 1 Second

29 7.2 Parallelly Abstract Time-Chains 1/4s T1 T2 B11 B21 T3 T4 B31 B41 B32 B42 1s B12 B22

30 7.3 Synchronize States with m-dpos cross Time-Chains m Delegated Nodes of Time-Chains will cross check to synchronize States of previous Blocks

31 TreeBlock uses both "DAG" (to link Blocks cross Time-Chains) and "States" to secure the network! Figure: Visualization in the normal status of the branch 1/4 second of the small Time Sharding Tree on Slide 3.3

32 7.4 Sequentially Producing Block With Time Sharding Example: Every 1 Second per Time-chain Or Every Time Step based on Time Segment overall

33 7.5 Sequential Abstract Time-Chains 1/4s T1 T2 T4 T3 Main Chain B12 B22 B32 B42 1s 1s B11 B21 B31 B41

34 7.6 Block Attack on Sequential Time-Chains 1/4s T1 T2 T4 T3 B12 B22 B32 B42 1s 1s B11 B21 B31 B41

35 7.7 Case: Overload & Load Balance Ex: Maximum 50 Tps per Time-Chain +50 Tx +15 Tx 11 Tx 16 Tx 15 Tx 100 Tx 1s 1s 1s 1s Shift and Process overloaded Tx on other Time-Chains after Cross-Check

36 7.8 Parallelly - Heavy traffic when parallelly cross-check between TimeChains per cycle - Better if extend block producing time, 2 s or more - Better with high internet speed and Node power vs Sequentially - Sacrifice some times to check cross Time-Chains per cycle - Able to work with shorter cycle for block producing - Better with high internet speed and Node power - Scale more with smaller Time - Scale more with smaller Time Segments Segments

37 8. Transaction Model COMBINING UTXO Good for processing Tx parallely cross-shards ACCOUNT / BALANCE Good for checking & recovering the State of network

38 A visualized diagram to describe how we process transactions parallel with UTXO and asynchronous Account Balance, we simplifed processes on this diagram, like automatically moving UTXO cross-shards, updating Account Balance process, or lock-time status of transactions

39 9. Digital Signature Algorithm EC-Schnorr Fast & Support Multiple Signatures & Save Storage

40 EC-Schnorr

41 10. Virtual Machine & Smart Contract TreeBlock's VM is based on Web Assembly. With WebAssembly, we can use C/C++ and in the future with other popular programming languages to develop smart contracts. Millions of existing software developers can become the blockchain app developers without learning new programming languages.

42 11. InterTree + InterChain TreeBlock can be flexibly applied for many blockchain solutions. Figure: Using TreeBlock as an InterChain Solution via Smart Contracts, "InterTree" with 4 Time Sharding Trees serves 10 other blockchains to secure their transactions at a high-speed rate.

43 12.1 A Time Segment Shard Topology

44 12.2 A Time Sharding Tree Topology

45 12.3 Security with AnyShard as AnyCast

46 12.4 Coordinator Layer for Load Balancing & Aggregating Response Data

47 12.5 Competitors

48 13.1 Token Circulation Flow From Token Sale

49 13.2 Community Fund

50 14. Applications Can Be Built on TreeBlock

51 15.1 Token Price During the ICO, the 2,500,000,000 TREX will be allocated in exchange for USD at a kick-start rate 1 TREX = $0.02 USD following market rate once get listed on exchange platforms, and exchange for BTC, ETH at the market rate with ceiling rate of 1 ETH = 20,000 TREX 1 BTC = 275,000 TREX

52 15.2 Token Allocation 10% 25% 5% 3% 2% 10% 5% 40% For Sale Community Fund Founding Team (Lock 2 years) Reserve Development Fund Bounty (Lock 6 months) Advisors (Lock 6 months) Partners (Lock 6 months) Marketing

53 15.3 Fund Distribution 2% 3% 10% 10% 50% Research & Development Operation Outreach, Partnerships & Integration Marketing Legal Costs Other 25% *Depending on the amount of funds raised, the distribution of costs between R&D and marketing may vary by 5%.

54 16.1 Technical Roadmap H1 2019: Complete Architecture Details H Testnet 20k tps Q Testnet 100k tps Q Mainnet 10k tps, Smart Contract Support Q Mainnet 20k tps Q Mainnet 50k tps Q Mainnet 100k tps Q R&D, Continuing to Scale

55 16.2 Business Roadmap 2018 ICO 2019 Growing Community 2020 Growing Community & Develop dapp Store From 2021 Growing Community, Manage dapp Store, Develop & Invest into dapp Startups

56 Open Discussion FOUNDER & CTO Anh Le

TOKEN PAPER.

TOKEN PAPER. TOKEN PAPER www.velix.id Table of Contents PAGE 2 Introduction: Identity Token Model... 3 VXD: A Utility Token... Tokenomics... Token Transactions... Business-to-Business (B2B)... Business-to-Customer