Blockchain and Holochain: Difference between revisions

From P2P Foundation
Jump to navigation Jump to search
No edit summary
No edit summary
Line 6: Line 6:
|-
|-
! Hash-chain approach
! Hash-chain approach
| Data-centric, ​​a ​​single​​ global ​​data​​set​​-​​one shared ​​reality ​​across ​​all ​​nodes.|| Agent-centric,​​ allows ​​nodes​​ to​​ act independently, ​​or ​​in​ ​tight ​​coordination ​​only with ​​counter parties,​​ and​ ​then ​​share independently​ ​evolving ​​data​ ​realities ​​that come ​​to ​​agreement ​​over ​​time.
| Data-centric, ​​a ​​single​​ global ​​data​​set​​-​​one shared ​​reality ​​across ​​all ​​nodes.|| Agent-centric,​​ allows ​​nodes​​ to​​ act independently, ​​or ​​in​ ​tight ​​coordination ​​only with ​​counterparties,​​ and​ ​then ​​share independently​ ​evolving ​​data​ ​realities ​​that come ​​to ​​agreement ​​over ​​time.
|-
|-
! Energy Use
! Energy Use
Line 20: Line 20:
| With ​​a ​​sharded ​​DHT, ​​the ​​transaction ​​load per ​​node ​​gets ​​lighter ​​as ​​the ​​network​ ​grows.
| With ​​a ​​sharded ​​DHT, ​​the ​​transaction ​​load per ​​node ​​gets ​​lighter ​​as ​​the ​​network​ ​grows.
|-
|-
! Energy Use
! Platform
| Bitcoin​​ consumes ​​more ​​than ​​0.1%​​ of​​ the world’s​​ electricity ​​​​to​​ power​​less ​​than 0.0001%​​ of ​​the ​​world’s ​​money.
| Can ​​now ​​only ​​run ​​effectively ​​with ​​special mining​​ rigs ​​or ​​wasteful ​​staking ​​algorithms.
| Since ​​no​ ​mining ​​is ​​required, ​​no ​​specialized processors ​​are ​​needed, ​​making​ ​it ​​feasible to ​​run ​​full ​​nodes ​​on ​​low-power ​​computers or ​​cell​​phones.
| Can ​​run ​​on ​​a ​​Raspberry ​​Pi ​​or ​​a ​​mobile phone.
 
|-
! Computational efficiency ​​of architecture (not ​1 ​​machine)
| O(n*m) ​​for​​ validating​​ transactions ​​on blockchain​​ as ​​a​​ whole​​ distributed architecture.
| O(n/m*log​​m) ​​for ​​validating ​​transactions.
|-
! Consensus Effects
| Core​​ consensus ​​algorithms​​ centralize power​​ (make ​​the ​​rich ​​richer). Proof-of-Work ​​results​​ in​​ infinitely ​​growing computational ​​over head ​​for ​​finite ​​data​​set.
| No​​ mining​​ or ​​staking. ​​No ​​consensus. ​​Not vulnerable ​​to ​​majority ​​attacks.​ ​You ​​only have​ ​to ​​trust ​​the ​​code​ ​on ​​your ​​own​ ​node and ​​can ​​validate ​​counterparty’s​ ​history directly.
|}
|}

Revision as of 09:34, 12 December 2017

Comparing​​ Blockchain ​​and ​​Holochain
Blockchain ​​Holochain
Hash-chain approach Data-centric, ​​a ​​single​​ global ​​data​​set​​-​​one shared ​​reality ​​across ​​all ​​nodes. Agent-centric,​​ allows ​​nodes​​ to​​ act independently, ​​or ​​in​ ​tight ​​coordination ​​only with ​​counterparties,​​ and​ ​then ​​share independently​ ​evolving ​​data​ ​realities ​​that come ​​to ​​agreement ​​over ​​time.
Energy Use Bitcoin​​ consumes ​​more ​​than ​​0.1%​​ of​​ the world’s​​ electricity ​​​​to​​ power​​less ​​than 0.0001%​​ of ​​the ​​world’s ​​money. Since ​​no​ ​mining ​​is ​​required, ​​no ​​specialized processors ​​are ​​needed, ​​making​ ​it ​​feasible to ​​run ​​full ​​nodes ​​on ​​low-power ​​computers or ​​cell​​phones.
Transaction Volume Neo currently ​​processes ​​+1000 transactions ​​per ​​second.​​ Bitcoin​​ and Ethereum ​​considerably​​ less ​​at ​​a ​​handful per ​​second. Expected ​​to ​​surpass ​​financial ​​exchange backbones ​​like​ ​the ​​Visa ​​network, ​​with ​​a ​​max of​ ​56,000 ​​transactions ​​per ​​second.
Scalability Even ​​ignoring ​​proof-of-work, ​​there​ ​are serious ​​scalability ​​limits​​ on ​​synchronizing ​​a global​ ​ledger ​​across​ ​many ​​nodes. With ​​a ​​sharded ​​DHT, ​​the ​​transaction ​​load per ​​node ​​gets ​​lighter ​​as ​​the ​​network​ ​grows.
Platform Can ​​now ​​only ​​run ​​effectively ​​with ​​special mining​​ rigs ​​or ​​wasteful ​​staking ​​algorithms. Can ​​run ​​on ​​a ​​Raspberry ​​Pi ​​or ​​a ​​mobile phone.
Computational efficiency ​​of architecture (not ​1 ​​machine) O(n*m) ​​for​​ validating​​ transactions ​​on blockchain​​ as ​​a​​ whole​​ distributed architecture. O(n/m*log​​m) ​​for ​​validating ​​transactions.
Consensus Effects Core​​ consensus ​​algorithms​​ centralize power​​ (make ​​the ​​rich ​​richer). Proof-of-Work ​​results​​ in​​ infinitely ​​growing computational ​​over head ​​for ​​finite ​​data​​set. No​​ mining​​ or ​​staking. ​​No ​​consensus. ​​Not vulnerable ​​to ​​majority ​​attacks.​ ​You ​​only have​ ​to ​​trust ​​the ​​code​ ​on ​​your ​​own​ ​node and ​​can ​​validate ​​counterparty’s​ ​history directly.