Roman Oliynykov - Bitcoin and Blockchain: Technologies, Scaling and Forks
1 like84 views
The document discusses Bitcoin and blockchain technologies. It describes the key components of Bitcoin including transactions, blocks, mining, and the blockchain. It explains how consensus is reached in Bitcoin to prevent double spending. It also discusses some of Bitcoin's advantages as well as open problems regarding scalability and centralization. Finally, it provides overviews of some approaches to scaling Bitcoin like SegWit and forks like Bitcoin Cash and Bitcoin Gold.
Roman Oliynykov - Bitcoin and Blockchain: Technologies, Scaling and Forks
- 1. Bitcoin and Blockchain: Technologies, Scaling and Forks Roman Oliynykov Ph.D., Dr.Habil. roman.oliynykov@iohk.io Input Output HK V. N. Karazin Kharkiv National University November 25th, 2017 Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 1 / 29
- 2. Outline Distinguishing characteristics of Bitcoin and other blockchain-based altcoins Transactions, Merkle trees, blocks and blockchain Consensus and mining, its complexity Bitcoin advantages & open problems Scaling Bitcoin Bitcoin forks Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 2 / 29
- 3. Distinguishing characteristics of Bitcoin and other blockchain-based altcoins a virtual decentralized currency with no trusted parties and with no pre-assumed identities among the participants; network operation support by moderately hard (solvable by participants) proof-of-work puzzles with rewards; a public distributed ledger (to detect and prevent money double-spending). Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 3 / 29
- 4. Three main technical components transactions (including scripts) united into blocks; the consensus protocol; the communication network. Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 4 / 29
- 5. Bitcoin transactions One bitcoin is divided into 108 satoshi Conservation of value principle: the sum of the values of all transaction is less than or equal to the sum of the values of all inputs (with an exception: coinbase transactions used to create new units of currency). There is no identities or accounts that ”own” bitcoins. Ownership is a knowledge of a private key which is able to make a signature that redeems certain outputs. Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 5 / 29
- 6. Merkle tree: uniting transactions into blocks Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 6 / 29
- 7. Transaction sequence Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 7 / 29
- 8. Blockchain: a hash-linked list of data blocks NB: tree data structure for blockchains: GHOST Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 8 / 29
- 9. Double-spending attack Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 9 / 29
- 10. Double-spending prevention Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 10 / 29
- 11. Bitcoin addresses Hashes of public keys in "pay-to-pub-key-hash" transactions function represented at base58 encoding as pseudonymous identities within the system and referred as addresses. No real-world names or identifying information are required. 1G9dguLRGKd16YBmM53iCuFVTTAUVog7s9 Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 11 / 29
- 12. Bitcoin consensus and mining Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 12 / 29
- 13. Bitcoins core innovation: Nakamoto consensus The process for choosing a new block (to be added to the public distributed ledger, or the Bitcoin blockchain): the first announced valid block containing a solution to the computational puzzle is considered correct; on hearing of the new announced valid block, other participants are meant to begin working to find a followup block; if an announced block contains invalid transactions or is otherwise malformed, all other participants are meant to reject it and continue working until they have found a solution for a valid block; at any given time, the consensus blockchain is the ”longest” version (with the biggest cumulative difficulty) of already known variants. Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 13 / 29
- 14. The consensus on the different branches (fork handling) The consensus blockchain is the longest version (with the greatest expected difficulty to produce). It is also possible for two valid solutions to be found at approximately the same time (depending on network latency): there are two equal-length chains (a temporary fork); miners can choose any fork in this scenario; one blockchain will eventually be extended further than the other at which point all miners should adopt it due to the random nature of the computational puzzle; the other fork will be discarded being ”shorter” comparing to another. Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 14 / 29
- 15. Confidence that payment has been made (will not be discarded in the future): theory Sequential nature of this consensus mechanism implies that users must wait for blocks to be found in order to gain high confidence that a transaction is permanently included in the blockchain. In theory, users can never be completely sure that a transaction wont eventually be removed by a very deep fork (there is always negligible probability that the currently available blockchain is a fork and will be discarded in the future). Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 15 / 29
- 16. Confidence that payment has been made (will not be discarded in the future): practice In practice, most Bitcoin clients require six confirmation blocks before accepting that a transaction is published (the probability of discarding such a block becomes smaller than 10−3 ). Deep forks are also prevented in an ad-hoc manner by including hard-coded blockchain prefixes (checkpoints) with the default Bitcoin client which clients require in any valid blockchain (it is considered by some researches that Bitcoin does not use a true decentralized consensus protocol, as checkpoints are chosen in a centralized manner). Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 16 / 29
- 17. Bitcoin miners Miners are participants who are working (racing in practice) to solve the computational puzzle in exchange for monetary rewards. A participant who finds a block can insert a coinbase transaction minting a specified amount of currency (a block reward) and transferring it to an address of their choosing. Miners receive all new currency initially and there is no other allowed mechanism for money creation. Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 17 / 29
- 18. The computational puzzle Requires finding a pre-image for SHA-256 hash function (to find a block whose hash is less than a target value starts with d consecutive zero bits; for November, 2017 this value is approximately 72 zero bits; cf.: the DES cipher has a 56-bit key). The randomized nature of this puzzle is important, each miner has a probability of finding the next block proportional to his share of the competing computational power (otherwise, in non-randomized case, the most powerful individual miner could be expected to find every block first). Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 18 / 29
- 19. The puzzle complexity The difficulty of the puzzle is calibrated so that a new block is found, on average, once every 10 minutes. To maintain this, the difficulty is adjusted once every 2016 blocks, or approximately every two weeks, by analysis of timestamps included in the previous blocks (with sanity checks to prevent manipulations by network rejection of the implausible timestamps). The block reward is determined by a fixed schedule. Initially, each block created B50. This has been halved to B25 and then to B12.5, and is scheduled to halve roughly every four years until approximately 2140 at which point no new bitcoins will be created. Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 19 / 29
- 20. A transaction fee Additional income: miners are also allowed to claim the net difference in value between all input and all output transactions in this block (a transaction fee). Fee values have primarily been determined by defaults in the reference client, with an option to pay higher fees to have their transactions published more quickly. Transaction fees have primarily been used to discourage overuse of the network with many small transactions (called penny flooding) and have never provided more than 12% of mining revenue. After 2140 transaction fees are planned to be main incentive for miners. Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 20 / 29
- 21. Mining pools: the majority of mining power since 2013 Miners often collaborate in mining pools to lower the variance of their revenue by sharing rewards with a group of other miners. Mining pools are typically administered by a manager, who, for a small fee, collects mining rewards found by all participating miners and allocates the funds to members in proportion to the amount of work they have performed on behalf of the pool (there is also small amount of ad hoc p2p mining pools). Participation is proved (probabilistically) by sending ”near-blocks” whose hash starts with a large number of zeros (d = 40 or more) but are not valid Bitcoin blocks yet. There is no reason to miner to hide the valid block, because block reward address is always pool manager’s address (and no other ”near-blocks” are accepted for the pool reward). Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 21 / 29
- 22. Bitcoin advantages the first wide spread virtual decentralized currency with no trusted parties and with no pre-assumed identities among the participants; easy-to-join both for users and miners; list-of-transactions only: no users, no balances or identities; a public distributed ledger with detection and prevention money double-spending; robust to cheaters: the network works normally when it just has an honest majority among all participants. Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 22 / 29
- 23. Open problems for Bitcoin a scalability problem: much smaller throughput comparing to VISA, MasterCard due to network propagation delay for consensus and complexity of heavy block processing (attempts to solve via SegWit, etc.); network latency: long time for transaction approval (not suitable for high frequency trading, etc.); centralization: if someone (a big mining pool or governmental agency) can control significant part of the network, it can effectively to work against default mining rules; mining pools unfair competition: one big pool can potentially infiltrate another big pool and decrease its mining efficiency; huge electricity consumption. Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 23 / 29
- 24. Scaling Bitcoin Segregated Witness (BIP141) Increasing block size SegWit2x Lightning Network · · · Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 24 / 29
- 25. Bitcoin forks Bitcoin Cash Bitcoin Gold Bitcoin Diamond · · · Bitcoin XT Bitcoin Classic Bitcoin Unlimited · · · Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 25 / 29
- 26. Bitcoin Cash Increased block size (up to 8 MB) No Segwit Emergency Difficulty Adjustment (EDA) for keeping stability of block issuance Replay protection and difficulty adjustment each block Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 26 / 29
- 27. Bitcoin Gold Changing proof-of-work: Equihash instead of SHA-256 (moving from ASICs to GPUs to eliminate centralization) A premined coins for developers (up to 1%) Segwit for scaling Replay protection and difficulty adjustment each block Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 27 / 29
- 28. Bitcoin Diamond Changing proof-of-work: X13 instead of SHA-256 (moving from ASICs to GPUs to eliminate centralization) Changed coin supply (210 millions) Segwit & Increased block size Replay protection Claims for more confidentiality (encrypted amount) Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 28 / 29
- 29. Roman Oliynykov, Ph.D., Dr.Habil. — Input Output HK 29 / 29