Session 02 Mysticeti v1

Sui Network
18 Jun 202416:32

Summary

TLDRThe speaker discusses the engineering team's breakthrough in transaction finality, introducing 'Misti', the world's fastest consensus engine aiming to achieve instant finality for shared objects. They highlight the importance of reducing latency for a better user experience and outline the innovative multi-leader DAG-based consensus protocol, which significantly reduces latency and promises unlimited throughput, setting a new standard for web 3 and global decentralized networks.

Takeaways

  • 🚀 The engineering team has over 70 PhDs working on cutting-edge technologies like consensus, cryptography, and storage for web 3 and the broader web.
  • 🔍 They are developing a solution called 'Misti' to achieve instant transaction finality for shared objects, similar to the speed of single-writer objects.
  • 🏁 Misti aims to reduce transaction finality time significantly, with current industry-leading times being 2 seconds, and the goal being under half a second.
  • 📉 Latency in user experience negatively impacts satisfaction, with each second of delay causing a 19% worse experience, highlighting the importance of reducing wait times.
  • 🔄 The company's blockchain has unlimited throughput, emphasizing the need to focus on reducing latency beyond just increasing throughput.
  • 🛠️ Different transaction types have varying finality times; single owner objects are instant, while shared object transactions take longer and are the focus of optimization.
  • 🔧 The 'fast path' allows single object transactions to bypass consensus, significantly reducing their finality time to 500 milliseconds.
  • 🌐 The company uses a Directed Acyclic Graph (DAG) based system for consensus, which offers advantages over linear-based systems in terms of resilience and performance.
  • 📈 The introduction of 'MRTI' (presumably a next-generation consensus mechanism) promises an 80% reduction in latency, with block times as low as 80 milliseconds.
  • 🎯 MRTI's design allows for a more efficient consensus mechanism, which does not require verifying the DAG for each block progression, streamlining the process.
  • 📚 The company's research and development are peer-reviewed and scientifically robust, with educational institutions set to teach MRTI as a course.

Q & A

  • What is the main focus of the engineering team's work as described in the script?

    -The engineering team is focused on developing advanced technologies related to consensus, cryptography, and storage, aiming to revolutionize the world of web 3 and the web as a whole.

  • What is the current challenge the team is trying to address in transaction finality?

    -The team is working on making transaction finality for shared objects as fast as single writer objects, aiming to achieve instant finality even when transactions require coordination.

  • What does 'Misti' stand for and what is its significance?

    -Misti is the name given to the world's fastest consensus engine developed by the team. It signifies the breakthrough in achieving faster transaction finality and reducing latency in blockchain transactions.

  • Why is latency a critical issue in blockchain technology according to the script?

    -Latency is a critical issue because it directly impacts user experience. Research has shown that every second of latency can worsen the user experience by 19%, which is why reducing latency is a priority.

  • What is the current transaction finality time for shared object transactions on the blockchain mentioned in the script?

    -The current transaction finality time for shared object transactions is between one and a half to two seconds.

  • How does the script differentiate between single owner objects and shared object transactions?

    -Single owner objects, such as asset transfers, happen instantly within 500 milliseconds. Shared object transactions, like marketplace transactions and collaborative games, currently take longer to finalize, between one and a half to two seconds.

  • What are the key components of transaction finality that the script mentions?

    -The key components of transaction finality mentioned in the script are block time, transaction finality, and settlement finality, which all contribute to the overall speed and efficiency of the blockchain.

  • What is the significance of the 'fast path' in the context of the script?

    -The 'fast path' allows single object transactions to bypass the consensus process, leading to a faster transaction finality of 500 milliseconds, showcasing the potential for high-speed transactions.

  • What are the two types of consensus algorithms discussed in the script?

    -The two types of consensus algorithms discussed are linear-based consensus algorithms and DAG (Directed Acyclic Graph) based systems, each with their own advantages and disadvantages in terms of latency and throughput.

  • What is the goal of the 'MRTI' consensus protocol as described in the script?

    -The goal of the 'MRTI' consensus protocol is to combine the benefits of a DAG-based system with the low latency of a linear-based system, aiming to achieve an 80% reduction in latency and an 80-millisecond block time.

  • What are the expected improvements with the implementation of MRTT in the mainnet by July 2024?

    -The implementation of MRTT is expected to reduce consensus latency from 1.9 seconds to 390 milliseconds and settlement end-to-end latency from 2.2 seconds to 690 milliseconds, making it the fastest chain for latency in the world.

Outlines

00:00

🚀 Introduction to the Engineering Team's Innovations

The speaker introduces the engineering team's efforts to revolutionize transaction finality in shared objects, aiming to match the speed of single-writer transactions. The team's expertise in consensus, cryptography, and storage is highlighted, with a focus on Misti, a new consensus engine designed for instant finality. The current state of blockchain transaction times is discussed, with a goal to reduce latency and improve user experience, as every second of delay negatively impacts satisfaction.

05:02

🔄 Understanding Transaction Lifecycle and Consensus Mechanisms

This paragraph delves into the lifecycle of a transaction on the blockchain, from creation to finalization, and the complexities involved in shared object transactions. The concept of 'fast path' for single object transactions is introduced, which bypasses consensus for speed. The engineering team's challenge to improve performance for consensus objects is presented, along with an explanation of linear-based and DAG-based consensus algorithms, and the advantages of the latter in terms of avoiding bottlenecks and providing higher throughput.

10:03

🌐 Introducing Misti: The Future of Consensus Protocols

The speaker reveals Misti as a groundbreaking consensus protocol that significantly reduces latency, with an 80% improvement over traditional methods. Misti operates on a DAG-based system, offering a lower block time of 80 milliseconds and faster transaction finality. The benefits of Misti include its multi-leader approach, which ensures robustness against poor performance or downtime, and its potential to finalize transactions without full consensus in some cases. The paragraph concludes with a preview of the dramatic reduction in latency that Misti will bring to the blockchain.

15:04

🏆 The Impact of Misti on User Experience and Future Developments

The final paragraph emphasizes the transformative impact of Misti on user experience, promising an unmatched level of satisfaction in a decentralized network. The speaker outlines the expected latency times for single writer and shared object transactions, which are set to be the fastest in the industry. The potential for Misti to be taught in universities and its peer-reviewed status underline the protocol's significance. The paragraph concludes with a look forward to further integration and improvements in the second half of 2024, solidifying the protocol's position as a leader in user experience and throughput.

Mindmap

Keywords

💡Engineering Team

The engineering team mentioned in the script refers to a group of highly skilled professionals within the company who are responsible for developing and implementing technological solutions. They are described as 'some of the best in the industry,' indicating their high level of expertise. In the context of the video, their work is crucial for the advancement of blockchain technology, particularly in areas like consensus, cryptography, and storage solutions.

💡PhDs

PhDs, or Doctor of Philosophy degrees, represent the highest academic degree awarded by universities in many fields. In the script, the company boasts 'over 70 PhDs working on consensus, cryptography,' which underscores the high level of specialized knowledge and research capabilities within their team. This highlights the company's commitment to innovation and research in cutting-edge technologies.

💡Consensus

Consensus in the context of the video refers to the process by which a blockchain network agrees on the validity of transactions and the current state of the ledger. It is a fundamental aspect of blockchain technology, ensuring security and trust among participants. The script discusses the development of a faster consensus engine, 'Misti,' aiming to achieve 'instant finality' for transactions.

💡Cryptography

Cryptography is the practice and study of secure communication techniques that ensure the integrity and confidentiality of information. In the script, the company's focus on cryptography suggests that they are developing secure methods for transactions and data storage within their blockchain platform, which is essential for protecting user information and ensuring the reliability of the system.

💡Web 3.0

Web 3.0, often referred to as the 'decentralized web,' is a concept that envisions a new iteration of the internet where content and services are decentralized, user-centric, and built on blockchain technology. The script mentions that the company's work on blockchain is aimed at 'changing the world for Web 3 and just the web as a whole,' indicating their goal to innovate and improve upon the current internet infrastructure.

💡Transaction Finality

Transaction finality refers to the point at which a transaction is considered complete and irreversible. In the script, the speaker discusses the challenge of achieving 'instant finality' for shared objects within the blockchain, contrasting it with the current speed of single-writer transactions. The goal is to make the finality of transactions as fast as possible to improve user experience.

💡Misti

Misti, as introduced in the script, is the name given to the company's new consensus engine, described as 'the world's fastest.' It represents an innovation in blockchain technology aimed at reducing transaction finality times. The script emphasizes the importance of latency and the goal of making transaction finality 'as fast as we have for single writer objects.'

💡Latency

Latency in the context of the video refers to the delay or time it takes for a transaction to be confirmed on the blockchain. The script discusses the importance of reducing latency to improve user experience, with the speaker stating that 'latency is the hardest problem to solve beyond throughput.' The company aims to reduce latency to achieve faster transaction finality.

💡DAG (Directed Acyclic Graph)

A Directed Acyclic Graph (DAG) is a data structure used in some blockchain technologies to manage transactions and blocks in a way that allows for higher throughput and potentially lower latency. In the script, the speaker mentions that their blockchain, 'onui,' is a 'DAG based system,' which is part of their strategy to improve performance and achieve faster consensus.

💡Throughput

Throughput in the context of the video refers to the rate at which transactions can be processed on the blockchain. The script mentions that the company has 'solved' the problem of throughput by having a blockchain with 'unlimited throughput,' indicating their focus on scalability and the ability to handle a large volume of transactions efficiently.

💡Settlement Finality

Settlement finality is the point at which the transfer of assets from one party to another is complete and cannot be reversed. In the script, the speaker discusses the importance of achieving fast 'settlement finality' alongside transaction finality, emphasizing the need for efficiency in the entire process from transaction submission to confirmation.

Highlights

Our engineering team includes over 70 PhDs working on consensus, cryptography, and storage innovations.

Misti is the world's fastest consensus engine, aimed at achieving instant transaction finality.

Current blockchain performance allows single writer transactions to finalize in half a second.

Shared object transactions currently take about 1.5 to 2 seconds to finalize on average.

The focus is on reducing latency and improving user experience by minimizing transaction finality time.

Execution of transactions can happen within 10 to 20 milliseconds, but consensus currently takes longer.

Misti aims to reduce latency from 2 seconds to milliseconds for faster transaction finality.

Misti is expected to have block times of 80 milliseconds, making it the fastest blockchain in history.

Misti uses a DAG-based consensus protocol to improve performance and reduce latency.

DAG-based systems offer more robust performance without being affected by single leader failures.

Misti's consensus mechanism allows for finality in 0.5 round trips, significantly reducing latency.

The introduction of Misti is expected to cut latency by 80%, with end-to-end transaction finality in 690 milliseconds.

Misti will be integrated into mainnet by July 2024, promising the best user experience for decentralized networks.

Misti's improvements will lead to unmatched user experience and performance in DeFi protocols.

Misti's innovations are recognized and taught in universities, highlighting its academic and practical significance.

Transcripts

play00:03

[Music]

play00:22

[Music]

play00:27

TR

play00:29

[Music]

play00:37

[Music]

play00:41

so I'm going to try my best to do as

play00:42

much Justice to this talk as possible um

play00:45

our engineering team are truly some of

play00:48

the best in the industry um we have in a

play00:52

company over 70 phds working on

play00:54

consensus

play00:56

cryptography um all sorts of really

play00:58

interesting um bleed bleed N Stuff all

play01:01

the way to storage that we think is

play01:02

going to change the world for web 3 and

play01:04

just the web as a whole um one of the

play01:07

things we've been working on is how do

play01:09

we

play01:10

make um transaction finality for shared

play01:14

objects as fast as single writer objects

play01:17

right now I was explaining before that

play01:18

to transact onui to do you know sending

play01:22

money to someone sending a token to

play01:24

someone sending an nft asset to somebody

play01:26

all that happens instantly right now in

play01:29

half a second what we want is when you

play01:31

have a transaction that's touching a de

play01:33

or any uh anything that requires

play01:34

coordination how do we get that to

play01:37

instant finality as well as as fast as

play01:39

we have for um single writer objects

play01:43

Misti is what we've called it Misti is

play01:46

the world's fastest consensus engine um

play01:49

people talk about throughput all the

play01:50

time throughput is interesting we've

play01:52

already solved that problem by having a

play01:54

blockchain that has unlimited throughput

play01:56

but what matters how do we go to instant

play01:59

finality latency is the hardest problem

play02:01

to solve Beyond

play02:03

throughput in this example we're just

play02:05

showing you what's happening Inu right

play02:07

now from when you approve a transaction

play02:10

to when it's final it's taking about 2

play02:12

seconds that's actually really good it's

play02:15

faster than all the other blockchains

play02:16

today it's actually really good but I

play02:19

think that's actually from a user

play02:21

perspective can we go better than that

play02:23

right it'll be amazing if that click is

play02:26

effectively instant as I click the

play02:28

button the finality is instant as a user

play02:31

would be much more pleased there's some

play02:32

research that's been undergone that

play02:34

every second that you have in latency

play02:37

for user experience makes it a a 19%

play02:41

worse experience for the user so users

play02:43

are going to be more and more upset the

play02:45

worse it gets that's why many people are

play02:47

un un um are not happy about blockchain

play02:50

performance today but we're talking

play02:51

about two seconds being the industry

play02:53

leading today we're not happy we want to

play02:55

go way below that we think half a second

play02:57

should ultimately be the goal

play03:00

now there are different types of

play03:02

transaction types onu as I mentioned

play03:04

there is um the single owner objects

play03:07

that's just an asset transfer um imagine

play03:10

a game making moves in a video game you

play03:12

do that singly and that's that's

play03:14

happening instantly um NF transfers all

play03:17

these things happen onui within 500

play03:20

milliseconds so when you click a button

play03:22

it's finalized within half a second but

play03:25

what about the other transaction types

play03:28

we're talking about shared object trans

play03:29

action types you're talking about

play03:31

Marketplace transactions you're talking

play03:33

about um collaborative games if you have

play03:35

multiplayer games I have two or three

play03:37

people playing a game or more these

play03:39

things right now are taking between one

play03:40

and a half to two seconds to finalize

play03:42

onu um even you know public accessible

play03:45

counters if you and I need to execute a

play03:47

transaction that requires an increment

play03:49

increment of a counter at any given time

play03:52

right now that requires coordination it

play03:54

goes through consensus so a bit of

play03:57

discussion about sweet speed the most

play03:59

important thing to think about is there

play04:00

are multiple components of this right

play04:03

there is block time transaction finality

play04:07

and settlement finality these are there

play04:09

multitude of things that's involved how

play04:11

fast can you propagate blocks how fast

play04:14

can you get transaction finality and how

play04:16

fast can you get settlement finality

play04:19

most people trying to give you

play04:20

optimistic numbers but those are great

play04:22

because when the world goes well but

play04:24

most times the world doesn't go so well

play04:25

so what are the what are the real

play04:27

numbers for finality from a user

play04:29

perspective when I've clicked the button

play04:30

and I know it can never be reversed so

play04:33

if you think about what matters it's the

play04:35

protocol efficiency which is the number

play04:38

of round trips across across the

play04:40

internet to the performance of a node

play04:43

consensus is the most expensive part

play04:45

execution is very low cost execution 10

play04:48

milliseconds is possible you can get

play04:51

maybe 20 30 but realistically onu

play04:53

execution is happening within 10 20

play04:55

milliseconds very very fast in parallel

play04:57

consensus as we say right now is taken

play04:59

above a second and a half to to do which

play05:02

is how do we get more efficient

play05:05

consensus mechanisms

play05:08

onu this is howu works today this is a

play05:11

life cycle of a transaction onu you

play05:14

would create a transaction once you

play05:16

create a transaction it would get

play05:18

effectively you collect a set of

play05:19

certificates once you have a set of

play05:21

certificates it would go for checks and

play05:24

verification once it's been verified it

play05:27

would go to ordering once it's been

play05:28

ordered it will go effectively to

play05:31

execution once it's been executed it's

play05:34

final so multiple steps required which

play05:36

is what's taking up to 2 seconds today

play05:39

so there are multiple complicated

play05:40

processes that have to happen for a

play05:42

transaction to be final if you see

play05:44

underneath there's something called fast

play05:45

path in a fast path which is a single

play05:48

object um transactions I talked about

play05:50

before you bypass consensus it's not

play05:53

needed so from when you verify a

play05:55

transaction you can go straight to

play05:57

executing it and that's happening in 500

play05:59

mill seconds can we get performance

play06:02

almost as fast if not faster than what

play06:04

we have today for single writer objects

play06:07

as we have for consensus objects that

play06:09

has being the task given to our

play06:11

Monumental engineering team today so now

play06:14

you know the anatomy of a three

play06:15

transaction let's talk about how it

play06:17

works with shared objects first

play06:20

transaction is submitted it's processed

play06:23

I get a certificate once I have a

play06:25

certificate it goes straight to the

play06:27

nodes for validation once it's been

play06:30

validated it goes to consensus for

play06:33

ordering once it's been ordered it goes

play06:36

to execution once it's executed we

play06:39

basically there's four R round trips

play06:41

required for that and ultimately it's

play06:43

finalized so you have something about

play06:46

you have up to about um four and a half

play06:48

rtts involved in that whole process butu

play06:51

is a very complicated system what can we

play06:54

do to simplify that and what can we do

play06:56

to get even more performance out of the

play06:58

system

play07:01

today there if I want to draw your

play07:03

attention of two types of consensus

play07:05

algorithms there are linear based

play07:07

consensus algorithms where you submit

play07:10

you create a block once you've create a

play07:12

block you for it you send it you you

play07:14

refer to the previous transactions in a

play07:15

block you you basically have another

play07:18

leader propose a um um the block you

play07:21

refer to that and you keep going over

play07:23

and over and over again the bottleneck

play07:25

is by the you can get bottleneck by a

play07:28

leader that's got poor band with you

play07:30

basically lead if a leader fails you

play07:32

have low throughput as a result which is

play07:34

a problem with most consensus based

play07:36

systems today and low latency it's about

play07:38

two to three round trips very very low

play07:40

very very um low latency but very

play07:43

susceptible to

play07:45

failure dag systems which is what we

play07:47

have today not bull shark onui today

play07:49

onui is a dag based system much more

play07:52

complicated but you can see that

play07:54

ultimately what we're doing is every

play07:56

single round we have multiple gossips

play07:58

happening where blocks have been

play07:59

proposed in parallel over and over and

play08:01

over again ultimately the benefit is you

play08:04

have it you have more latency but what

play08:07

you have is you're not susceptible to

play08:09

bad um performance of a single leader

play08:11

and over time you also have um you know

play08:14

it's it's a more lightweight algorithm

play08:15

from a perspective of implementation

play08:17

from our perspective but at the same

play08:19

time you have stronger guarantees as

play08:20

well so we think a dag based system is

play08:23

the way to go it's we use as a dag BAS

play08:25

system today but the the the downside of

play08:28

a dag BAS system we have is actually an

play08:30

increase in

play08:33

latency so again traditional leader

play08:36

driven um it's linear communication low

play08:39

latency two to three round trips and

play08:41

it's of course leader the con is a

play08:44

leader is a Bott neck in a in a in a

play08:46

protocol our dag based consensus is

play08:49

effectively way faster like it's you can

play08:52

get more throughput out of it as a

play08:53

result but again as I touched on we

play08:55

think you know we want to get better in

play08:57

terms of latency and finality so what is

play09:00

mrti Miss setti is is our ambition to

play09:04

effectively make a dag based consensus

play09:07

protocol have as low as uh um um as low

play09:11

latency as you get with a linear base

play09:13

system and we're able to achieve that

play09:16

our consensus 1o today is final at 1.5

play09:19

round trips effectively for a single

play09:22

round it takes about one to five round

play09:25

trips sorry with Mr setti consenso it's

play09:30

half a roundtrip to create a single

play09:33

block in fact the block time of mrti is

play09:36

80

play09:37

milliseconds it is the lowest block

play09:40

block time of any blockchain in history

play09:42

80 millisecond block times as a result

play09:45

of

play09:46

0.5% round trips to create a single

play09:49

round in missii it is a game Cher it is

play09:51

a industry leading mechanism and the

play09:54

reason we can do that is we do not have

play09:56

to verify the dag as part of progressing

play09:58

the the block

play10:01

so the difference is as I mentioned

play10:03

consensus 1.0 require the verification

play10:06

of a dag so every time you receive a dag

play10:09

it has to go through a certification

play10:11

with consensus 2.0 it it requires an

play10:14

unverified dag so after about um two to

play10:16

three rounds you've got full

play10:18

verification that a transaction is final

play10:20

ultimately it's it the committer done in

play10:23

pipeline in parallel and multileaders

play10:26

you have multiple leaders if one leader

play10:27

is out it doesn't matter the chain

play10:28

doesn't know notice it keeps going on it

play10:30

it can take care of um of downtime um it

play10:33

can take care of bad leaders it take

play10:34

care of people who have very low

play10:36

performance it's not affected by the

play10:38

worst leader like most Protocols are but

play10:40

ultimately what it gives us is what

play10:42

we're seeing an 80% reduction in latency

play10:46

today this is what we have in testet you

play10:49

could see this is naral and bull Shar

play10:51

that we're running right now that is

play10:52

showing close to 2 seconds and as soon

play10:54

as Miss got turned on it cuts down to 20

play10:57

397 milliseconds

play11:00

by the time this hits production this

play11:02

will be the fastest chain for latency in

play11:05

the world s we already believe it's

play11:07

there but we think Mr setti just sets a

play11:09

whole new bar for what users should

play11:11

experience in a Global decentralized

play11:14

Network another benefit as well is p50

play11:18

shared object latency what does that

play11:20

mean 50% of users which is going to be

play11:23

the average of users that engage withu

play11:25

will see finality end to endend finality

play11:27

from when you click a button to when

play11:29

it's Final in 690 milliseconds that is

play11:33

insane before the website is loaded your

play11:35

transaction is final that means the

play11:38

amount of use Cas that you can engage

play11:40

with is broader we just gave you a card

play11:43

this card that you scanned today when

play11:45

you put it on your phone and you clicked

play11:47

on approve it took about 2 seconds to

play11:48

finalize that is to do with now war

play11:51

bullshark in a few months once it's live

play11:53

in mainnet when you do that same scan

play11:55

when you click on approve it will be

play11:56

instantly Final in your account so that

play11:59

is what it means about building the best

play12:01

user experience for users over and over

play12:03

again is thinking about all the way from

play12:05

when a user submits a transaction to

play12:07

when the chain gives you feedback that

play12:09

it's final this is endtoend latency

play12:11

optimistic latency is a lot lower way

play12:14

lower block times of 80 milliseconds

play12:16

through entend finality of 690

play12:18

milliseconds But ultimately the user

play12:20

will have the best experience that they

play12:21

can ever have of any chain not having a

play12:25

not not going to see failed transactions

play12:27

not going to see latency Spike up in

play12:29

times when a network is busy not going

play12:31

to see a maximum throughput ultimately

play12:33

the SC ch um they change skilles

play12:35

infinitely will be releasing we released

play12:38

a paper already called pilotfish and

play12:40

pilot fish was able to show that when

play12:42

you added eight times more machines to S

play12:45

you got eight times a throughput so

play12:47

there is no maximum throughput

play12:48

separately the latency is constant no

play12:52

matter how many transactions you throw

play12:54

atui and the amount of um U machines you

play12:56

throw at the network the latency always

play12:58

stays same and the throughput goes

play13:00

higher and higher this is how internet

play13:02

skill infrastructure should be

play13:05

built and one thing we're going to be

play13:07

working on is effectively cutting down

play13:10

the latency again for single writer

play13:13

objects right now I explain that there

play13:15

are 500 milliseconds by the time we

play13:17

merge the code base between um the

play13:20

ingestion phase and also the

play13:23

verification phase we think we'll be

play13:25

able to get latency for single writer

play13:26

objects end to end down to 300 M Ms so

play13:30

you have 300 milliseconds for single

play13:32

writer objects Final End to End finality

play13:34

and 690 milliseconds end to end finality

play13:37

for shared objects in fact we've seen as

play13:39

low as 500 milliseconds for shared

play13:41

objects in testut as well so we think

play13:43

we'll have low subsequent finality for

play13:46

all transactions inui over over over and

play13:49

over again um another thing of pilot

play13:51

fish is we think we can actually get a

play13:53

scenario where some transactions that

play13:55

need consensus can be finalized without

play13:57

consensus that's going into too much

play13:59

detail but there are people who have way

play14:01

more phds on me on this topic that can

play14:03

talk about that in detail but we think

play14:05

that's going to be another variation of

play14:06

what we can do

play14:08

here so to summarize Miss setti 80%

play14:12

latency reduction consensus has gone

play14:15

down from 1.9 seconds to 390

play14:17

milliseconds settlement end end to end

play14:20

from 2.2 seconds to 690 millisecs um it

play14:24

will be in main net by July 2024 in fact

play14:27

it's running in test net now the

play14:28

numberers is already published it's

play14:30

running the same codebase the same

play14:32

validator setup across the globe and

play14:34

we're seeing remarkable results and we

play14:36

we want to turn on in main net in July

play14:39

and we're going to be integrating um um

play14:42

congestion um phase with mrti in uh

play14:45

basically H2 2024 so there'll be a lot

play14:48

of more upgrades coming to S as a whole

play14:49

But ultimately we think D5 Protocols are

play14:52

going to see an amazing Improvement in

play14:54

user experience it's already Indy

play14:56

leading but I think it's going to take

play14:57

it to a whole new level but overall we

play15:00

think more most importantly the user

play15:02

experience across wi would be just

play15:04

unmatched unrivaled to what any other

play15:06

ecosystem as made possible this is all

play15:09

possible by having world renowned

play15:11

researchers in a team everything we have

play15:13

is peer-reviewed and has some of the

play15:15

most amazing scientific reviews um in

play15:18

the world that looked at it it's all

play15:20

verified all it's it's not just code

play15:23

that goes out with with uh all sorts of

play15:25

caveat but these are things that you

play15:26

should read up on yourself we actually

play15:28

have some University they're going to be

play15:29

teaching Mr setti as a course in the in

play15:32

universities as a result because for the

play15:34

first time they can actually understand

play15:35

a proper division team for tolerant

play15:37

protocol with a dag based system in a

play15:39

way that it's easy to actually

play15:41

internalize so we believe we're making a

play15:43

lot of great changes for the world but

play15:45

more importantly we think we're building

play15:46

the best developer platform for web 3 as

play15:49

a

play15:50

whole um if you want more information on

play15:52

mrti do scan this QR code I'm happy to

play15:55

answer any questions you may have but

play15:56

ultimately this is a summary of the

play15:59

elements of missii it's a multi-leader

play16:01

dag based consensus protocol it has

play16:03

industry-leading latency um lowest

play16:05

latency that's ever been found in any

play16:07

protocol it orders are maned faster than

play16:09

what exists today and it's supported by

play16:11

it's going to be supported by pish

play16:13

effectively unlimited throughput base

play16:15

system thank you very much thank you

play16:20

[Applause]

Rate This

5.0 / 5 (0 votes)

Related Tags
Blockchain TechConsensus EngineTransaction FinalityWeb 3.0CryptographyDAG SystemDecentralized NetworkHigh ThroughputLow LatencyUser ExperiencePh.D. Research