---
title: "Parallelizing EVM w/Sei | Twitter-Space-Aufnahme | AlphaGrowth"
description: "Twitter-Space-Aufnahme vom 2024-01-25 mit OKX Blockdream Ventures, Sei 🔴💨, Kiwibig.eth"
url: "https://alphagrowth.io/de/spaces/parallelizing-evm-wsei"
---

# Parallelizing EVM w/Sei

Aufgezeichnet: Jan 25, 2024 Dauer: 0:38:09

## Space-Aufnahme

[Original-Space öffnen](https://twitter.com/i/spaces/1MYxNoNQYZpKw)

0:00 38:09

### Vollständiges Transkript

good morning good morning mike test one two three

hi grover

hey how are things

yes I'm waiting for more listeners to join shall we wait one minute sure yeah

let's give it a couple minutes sounds good

oh can we start so hi everyone

a great to have you all here and welcome to our Twitter space today we will

discuss paralyzing EVM this to the space is hosted by OKX winners we are the

winter arm of the world leading crypto exchange OKX we hold the vision to

support and fund most innovative web 3 projects including all the directions

like infrastructures DeFi and the gaming networks and other applications etc so in

short OKX winters is a long-term value investment investor and try to inject

capital as well as resources to high-quality teams to spot sustainable

development for the whole blockchain industry and Jenny from OKX Ventures

player to be the host for today's session and pretty and pretty and very honored to

invite a grower head of marketing from say team and kiwi researcher research

researcher of OKX Ventures to join this Twitter space and share the insights

hello how are you doing hey thanks thanks for the intro yeah great to be here

maybe I'll do a quick intro so yeah my name is Grover I'm head of marketing here

at say I've been with the team since nearly start not quite but nearly and

yeah I'm really excited to share some more info about say and our plans to

parallelize the EVM today. Thanks Grover yeah I'm Kiwi from OKX Ventures and I'm

responsible for the research team of OKX Ventures and we are very

glad to have you invested and very glad to see the achievements of the network

has achieved yeah that's all. Thank you for the introduction the first question is

it's for Grover can you give us a brief overview of say and the core vision

we had it sure yeah so I'll go back to the very start so back at the beginning

back in 2021 originally co-founder of say labs Jay was at Robin Hood and at

that time he went through I think some of you might be familiar with like the

GameStop saga where sort of the centralized offering switched off the buy

button and he was somewhat disillusioned by that and decided he wanted to build

an on-chain exchange that could rival centralized exchanges himself to go out

and sort of change change how things were being played out but he realized

back in 2021 that the infrastructure just wasn't there so this was the

original inspiration for say to build a layer one chain that facilitates

crypto apps that have a web to like user experience so you get all the

benefits of crypto like self custody decentralization and all of the sort of

the web to like user experience this was the vision so a lot of the team are from

the Bay Area in San Francisco and have experience in other high-tech high-paced

tech companies so that's a little bit of context we think that's the best way

to approach building a layer one blockchain is to basically crank things

and build the most performance fastest time to finality blockchain

infrastructure possible which is accessible and open to as many developers

as possible to truly move the needle in the crypto space so if say is successful

then applications will be able to offer nearly the same UX as any web to

competitor and that we think is really really powerful so yeah we'll get into

sort of like the next steps and how we're parallelizing the EVM comes into

this as well so yeah that's a brief intro into into what say is thank you

Grover yeah we got a lot of reports about the network we got very research

many research about about say network and kiwi can you give us a brief

overview of okx ventures and yell and you'll see thesis and call vision yeah

yeah of course getting I think everyone has no one about okx venture and okx

exchange and okx venture is a venture department or okx exchange and okx

venture has been attended for on the firm stance during the very difficult time

in the past few years in in the last year we have invested over 50 million

in the market we all know it was a very hard last year yeah and okx ventures

investment covers many core areas as such as infrastructure, gamify, DeFi,

web story, Bitcoin, AI, dipping and etc yeah we are very interested in the

inter the projects that can bring innovation to the whole industry in

terms of quality the quantity I think 50% of deals are infrastructure and the

DeFi projects and I think maybe 20% are gamify projects and in the past few

years the market is bad but I think it's a good time for infrastructure to

build so we put a lot of effort in infra but recently we have been

improved more attention to the applications and we have invested in many

high quality projects in computing c-net work and we are looking forward to do

more and help the innovative projects to grow yeah that's that's my answer

Janet thank you kiwi for the overview of okx ventures and the next question is

for say team what inspired the development of say version 2 and what

were the motivations to propose this upgrade to say so yeah I think one thing

is really really clear just to set the context or the scene here which is that

the EVM is here to stay so the ethereum virtual machine is the software that

runs on top of aetherium and all of the layer twos around aetherium we think

that it has somewhat terminal velocity and it's extremely unlikely for the EVM

to lose its majority market share in the next at least 12 to 18 months you

know in the next like the foreseeable future that's a long time in crypto like

kiwi mentioned in the last year probably felt like about 10 years for many of us

so other execution environments like move VM and fuel and things exist but

they're gonna have a tough time we think overthrowing EVM for now so the the

current problem that the EVM has is that it doesn't scale throughput very well so

on east L1 and east roll ups you can't get more than 30 TPS something

30 to 50 TPS which really really limits the design space on aetherium because many

applications that you want to be building will want to submit more than that in

30 TPS so by building a parallelized EVM and we'll talk about what

parallelization means in a bit it lets you get the best of aetherium and also

some of the best parts of Solana as well so you get all the EVM sort of

mindshare and tooling and teams and all the work that's been done and you

unblock a lot of developers there we think in the Ethereum ecosystem plus you

get the fast execution environment that Solana kind of offers so yeah that some

of the the the reasoning for that is we spoke to a ton of developers and just

came to the conclusion that this was really really significant and this is a

way that that's a can sort of move the needle for the crypto space we're

implementing this in a somewhat specific way we can get into some of the

details we're adding Geth which is a client that run EVM client that's the

most battle-tested and most well-understood and well-used version of

the EVM and running that on say and so they will be completely bytecode

compatible with existing applications that use the EVM so anybody who's

building apps on aetherium right now or you sell twos could just come over with

no changes and deployed on say and then they'll get all the benefits of

running it's you know 500x higher TPS than usual and faster time to

finality as well so we think that that's gonna be a big game-changer yeah I

mean like overall like just zooming out for a sec as well paralyzing the EVM we

think is is a really useful and high potential scaling approach for the

broader aetherium ecosystem so we really hope that the aetherium community

can leverage some of the open source code base of say to achieve new

possibilities for scaling I think a lot of people view the industry is zero-sum

however the crypto industry is still really really early really nascent in

the grand scheme of things so we take a positive some viewpoints where the

theorems a layer two is going to help to scale the industry in future there's a

bunch of other stuff we can get into but that's sort of some of the high

level thinking here thank you for the reply we know say is the word the world

first parallel you may be opening new opportunities for global EVM developers

making a significant step in sales for utilization of Saluna Salana and

it's a rain that's too highly optimized execution layers so the next question is

for kiwi can you share some insights into the key technological solutions you are

exact you are excited about in the crypto space right now how do you think

about the high performance integrated chains like say and Salana yeah of course

absolutely the crypto area is consistently evolving and there is a

worry couple of tech advancement that is exciting first of all this is

based about a c-net worker and I wanted to talk about the development of

paralyzed the UEM like we have what we have seen in CV2 CV2 is a worry

intriguing the approach can potentially be a game changer for stability in the

theorem ecosystem because it promised to automatically had those some smart

contracts pyranization take a huge load it will take off a huge load for the

developer of and another thing is the integration with diverse smart contract

ecosystem with CV2 in labeling UEM contract to interact with other types of

smart contracts like cost system we are looking at a much more collaborative and

integrated future as opposed the high performance we all know it's the

one teacher OC network and also for you it used to for Solana I think Solana is

also faster but I think a bit less the faster that as I say and now I'm thrilled

that the way c-net work is pushing boundaries says boosting the affinity

under 400 ms and the promise the promising high transaction per second is

nothing short of impressive this puts us closer to achieve a web tool like

user experience in crypto it will introduce more and more applications

that needs a computation massive must work like a cause a lot of computing I

think this kind of application can be introduced by c-net work and we are

really looking forward to see what the what we all see network ecosystem will

grow this was I'm being excited about yeah Jenny that's that's all yeah thank

you from the ancient data we it shows there is a lot number of current users

and transaction one of our say midnight main night congratulations to say team the

next question is for Gruber to explain in simple terms what a paralyzed EVM means

and the key advantages to our listeners yes so I think this is one of the most

important things to touch on because if there's a lot of buzzwords and I think

it's super important that people have some kind of a mental model for for what

what we're talking about so we mentioned before what what is the EVM so ethereum

virtual machine is just the software that's running on top of ethereum and

that's where you deploy your applications is where smart contracts live

and that's kind of like the like Windows running on top of your your computer

it's the way to think about it so what is parallelization so I'll use an analogy

here for parallelization and this is the best one I have so far if anyone has a

better one please do share it let's say you have a vending machine and it's a

really really hot day so maybe you're you're on the beach and you've got it

you want to get a drink so the vending machine is serving let's say five

different types of drinks so coke Sprite Fanta 7-up whatever you want to

drink however you have to all queue so everybody you know if someone wants a

coke or a Sprite or a Fanta you all have to queue at the same time in the same

queue so instead if you parallelize that process now let's say you have five

different vending machines and they all serve one different type of drink so if

you want a Sprite or a coke or a Fanta you don't have to queue with the other

guys who want a different type of drink so with that in mind the way this

works is transactions that touch different parts of say state can take

place at the same time so let's imagine you have an NFT mint going on but you

just want to send some gas like you want to send them some tokens to somebody

else you don't have to queue with the people doing the NFT mint you can just

go ahead and you can send your your transaction if it's completely unrelated

and it won't have to queue with with those guys so that by itself leads to

a reasonable performance boost a reasonable one not game-changing on savee

to this process will be optimistic meaning that developers don't have to do

anything all of the transactions are assumed to be parallelizable so that

means that they don't have to make any changes to their smart contracts and

they'll just it would just be assumed that all these things can happen at

the same time there's more details about this in the save blog if you

would like to get into some of the details so by operating the EVM on top of

say not only does it benefit from this optimistic parallelization but it also

means that you get the fastest time to finality of any blockchain so I think

Kiwi you mentioned this before the the very unique thing about say is by

leveraging the work done by the Tendermint team and some of the cosmos

teams with Tendermint consensus you have single slot finality that means

that once a block is submitted so like once once you've had a transaction

included in a block and that block is built and it's accepted yep that's it

it can't be re-orgued it can't be changed that's it forever so you

achieve finality like the chain is it cannot change in that 390 milliseconds

which is very very different so on Solana I think you need multiple you

need multiple blocks so you get for about 10x faster time to finality than

Solana and then aetherium it's like 20 minutes you know so like this is like

fundamentally very very different so that's what's one thing to mention as

well we think overall this opens up a huge new design space for the team is

building on and around aetherium so with that with that very very short time

to finality around the same time as the humans reaction times and also with

the vastly higher throughput that you can get on say we're looking at

something like twelve and a half thousand TPS theoretical you'll be able

to just build completely different types of applications in the first place so all

you guys will be able to come and use metamask and other aetherium wallets to

sign transactions and say metamask will actually have a snap so you can sign

both EVM and cosmwasm transactions and say so that this is yeah we think it's

going to be really fun and I'm really looking forward to it launching

Thank you Grosvenor. Yes because optimizing EVM is a continuous

in the crypto world and it said it is said the year of 2020 for will be the

year of parallel with EVM will be a very important year for say network the

next question is for is also for you apart from optimistic parallelization

what are what other improvements does say v2 bring yeah so yeah parallelization is

sort of the buzzword but there the savee to proposal touches pretty much every

single part of the stack in some way so first of all yeah we'll have both

cosmwasm which is like the cosmos smart contract language and execution running

at the same time as the aetherium virtual machine on top of say together both part

of say is core binary so not as like smart contracts or a layer to this is

this is like part of the blockchain itself the second thing is the

optimistic parallelization we talked about the third thing is something

called say DB so this is an effort that's inspired by a lot of work done

over the past year in collaboration with other teams around cosmos and some

other some other blockchain teams as well and that ensures that say can

squeeze out the maximum performance from paralyzed EVM and addresses state

bloat as a problem as well so one other concept that to introduce here is the

idea state bloat so if you can imagine that like so say has already done over

2x more blocks than aetherium has in its entire life already so that's a

lot of data like the rate of information and data being being stored is

very very high so you need to consider what changes you need to make here so

that it's sustainable in the long term so that people don't need to have like 20

terabytes of hard drive storage SSD storage so yeah say DB approaches this

and solves a lot of these problems around state bloat it solves a lot of

problems around just how state is accessed again there's a ton of detail

there we are releasing a blog post I believe either today or tomorrow which

which goes through all of say DB's improvements in detail but that's quite

an important part of the upgrade as well and lastly I think twin turbo

consensus so the same consensus mechanism I spoke about before which

achieves you know 390 milliseconds time to finality that will remain part of

say as well what's really important is all of these components are either fully

battle-tested or rigorously tested in production since say has been live since

it's August and then all of these components are like very well understood

pieces of software that have been running for a long time and yeah like I

said well understood which which you think is quite important iterative

improvements one of the one of the sort of the theses of the team is you

launch you gather feedback iterates and then you relaunch that's that's going to

lead to vastly higher and better outcomes than building for a long long

product protracted length of time and doing something completely brand new

from scratch like this is going to be a little better outcomes so that that's

kind of this this idea of ruthless prioritization that the say labs team

puts into place you know start with what people need build the software

based on that and then iterate over time so yeah hopefully that answers some of

your questions there yeah yes Grover in summary I think seed network V2 is a new

generation of network technology and it has highly several highlights is is of

higher performance stronger scalability better mode expression ability and higher

security in my in my personal opinion so the next question is apart from what is

the roadmap to to man night for say V2 and how can teams get involved and who

might be interested in taking apart yeah so I say V2 is code complete right now

has been live on deafnet since December the the next step here is to move to

testnet and and that is scheduled soup very very soon Q1 is is the official

answer there I think we'll over promise under promise and over delivered there

mainnet will be subject to governance and audits but that's scheduled for the

first half of 2024 you can check out the state ecosystem right now which is

obviously live and like you mentioned before is kind of there's a lot of

activity right now which is awesome to see this there's a bunch of NFT projects

and some mean coins and defi is starting to pick up so definitely come

and check out the existing say ecosystem we're seeing a lot of like I'd call it

grassroots kind of development happening now which is just really really awesome

to see and I think the whole team here at say labs is really really happy to

see it happening yeah so one thing to bear in mind like hey you know testnet

will come out and there'll be some bugs to fix and there'll be some teams to

deploy so that's just like the the next milestone there is then people can you

know teams will be able to start deploying their applications so if you're

an ethereum developer if you're a team who's been building around ethereum for

a while you might want to come and check it out you know you might find a

couple a couple things here and there to help us fix and then you can deploy

your your applications as well there are a bunch of I guess you could say

household names that are ready to deploy on and save you to out of respect for

those teams I would drop drop alpha here but yeah there's there's some really

awesome partnerships that are forming now I think a lot of people are quite

excited to see see how this plays out

yeah yes as the one of the investors okax ventures would like would like to

introduce you more partners as well the next question is for kiwi from out from

your perspective what potential to say we to hold what are you looking forward to

what are you looking forward to in 2024 oh yeah yeah yeah yeah I think for for

same network we wish version 2 I think there are many thing to expect first is

the you know vision step CV2 is ambitious upgrades it mean to turning seed

blockchain into the first paralyzed uvm uvm mode blockchain and this could make a

sean leader in this full field and we can say many competitor our foremost see

network and I think see we all need to see in this area and the second is I think

do I protection is very important I think as I mentioned in last question I

think if you see network contract more developer and there will be more

application and see network and under say we to facilitate easy deployment from

other uvm compatible blockchain so this will attract a very broad range of

developer and the community and the third part is that I think tooling and the

user experience developer can get access to a very familiar tool I would do

developing tools like a metamask simply the transaction and forcing of

developer-friendly ecosystem so well users can get to getting to see network

worry very easy without getting new new new one list and this is very important

and the fourth part is that the polarizing and the perform performance we

think obtain optimistic polarization in CV2 allows more efficient transaction

hi Jenny hi TV yeah yeah you you you're I can hear you you're typing yeah okay you

can close your mic yeah that's the the keyboard of legend I can hear being

typed on there it's just very loud yeah I think the first part of it that is the

test and the launch timeline I think with the major upgrade is nearly could

the company is CV2 maybe maybe in q1 I think it's a good community to test and

engage in the plant a new feature so this is just a four points is what I'm

expecting from CV2 and for the next question what I'm looking forward in

2024 I think the most important thing I'm looking for that is a the state

community I think if there is a more more medication and more users this is

this is more important and I think if they can connect with some new narrative

like AI like dipping like something like Bitcoin I think this will a broader make

make this networks vision more broader I think it's the most of helps the

network but I don't invest in more detail so it's just for what I'm I just

to think about yeah so that's all yeah thank you Jenny thank you kiwi for the

answer the next then next we'll select three questions from Twitter comment the

first one is how does the proposed save save it to upgrade enable etherim and

ether if they're in there to depth to leverage say blockchain the last speed

and throughput and what will be the user experience when interacting with say

using apps like Metamask yeah good question so the the upgrade means that

say will be fully bytecode compatible backwards compatible with any

applications have been built on Ethereum and Ethereum layer twos or most of them

at least and this means that they can just deploy with hardly any changes

whatsoever just directly on to say and they'll immediately get all the benefits

of the faster throughput and faster time to finality that you would you would

get if you were deploying your app on just like natively on say in the first

place but also yeah it's gonna lead to a vastly different user experience so if

you go along and you connect your Metamask to save you two and you start

using the apps you should find that the slowest part of that process will

actually be Metamask refreshing the transactions will be finalized almost

instantly on say and then as soon as you refresh it's it's just done so you

should find it's much much snappier than aetherium over time I would like

to see the refresh rate on Metamask sped up so you can get the the true

experience that I want to see which is like a web to like experience but you'll

find that yeah it will just be done faster than you would have expected I

think that's the user experience that you can kind of look forward to when

you're using apps on on say using Metamask well thank you and the next

question is also for Grover so compared to other layer one blockchains how does

say custom consensus mechanism contribute to speed and transaction

penalty yeah so basically say started out with Tendermint which was consensus

mechanism used by a lot of cosmos chains in the past you know some some big ones

some well-tested ones ones that went through a lot of volatility and a lot of

crazy activity and basically took that and that originally it has six seconds

time to finality and and cranked it to to the max I suppose is the best way of

putting it so adding things like optimistic block propagation and things

like so basically changing the way that blocks are built to make it faster got

that down from six seconds to something like 250 milliseconds and then added a

little bit back in because that was just a little bit too fast so that's

because it's single slot finality so once one block is finalized that's it

it means you have this really really fast time to finality which is important

for anybody who wants to do things like high-frequency trading you know they

don't want to get reorgs later on there's like a risk if they're like

hedging position or they're trying to execute some kind of a strategy that

they don't want it to turn out that ten blocks later that things get reorged and

we're using a different chain now and actually your position is different so

like that that kind of stuff is quite important to people who want to start

executing these strategies on things like order books so yeah that's just a

little piece of what's going on there with say consensus mechanism which we

call twin turbo consensus to make it much more exciting I want you to think of a

fast car with a gigantic engine that's my job in marketing so I want you to

think about that when you think about say yeah thank you for that say network

is I think it's a high-speed layer one population it's born for transactions

and it's called decentralized the not stuck so the last question is what are

sales plans after the net launch what are the interesting new things that we

will see from say yeah so the mainnet launch is again I would always say

we're very early it's just the start there are a bunch of things on stage

roadmap beyond safety we'll definitely be listening to a lot of feedback say

labs we're listening to a bunch of feedback from the community and what

people need and what people are looking for I know one thing that is quite

important to us is increasing the validator set and also ensuring that

people can verify transactions and say easily so right now you need to run a

validator node you need to have 64 gigs of RAM and 16 cores which is not

absurd I think people tend to think that you need a quantum computer to

validate say but that's not the case it's pretty reasonable one terabyte SSD

but in future we want it to be as accessible as possible for people to

verify transactions and say and this is all part of the decentralization push

so working on things like zk like clients is is definitely there in the

roadmap and yeah we'll be listening to more feedback there'll be I think many

more teams that come and deploy on say over time and start to build out

applications that specifically leverage the new design space so at the

beginning you know my personal best guess is that we'll see some applications

that you're familiar with that are basically it's very similar to what you

see on Ethereum and Ethereum layer 2s but just running a bit faster but over

the over time I expect that you'll see new types of apps that just couldn't be

built in that design space being deployed over say but that takes time

so yeah I think that's a little bit about what we'll see on say in the

future well yeah really looking forward to the new things so great if there are no

more questions let's wrap up both the day really appreciate your attendance and

insight for contributions discussion was very enlightening for me and I hope our

audience feels the same way thank you again for your time our TV and Grover

thank you a lot to our community as well see you next time

thank you Jenny bye thank you Robe thanks everyone

take care bye bye I think I run

### Host

- [OKX Blockdream Ventures](https://alphagrowth.io/de/spaces/participant/okx_ventures.md)
  [X-Profil](https://twitter.com/OKX_Ventures)

### Sprecher

- [Sei 🔴💨](https://alphagrowth.io/de/spaces/participant/sei.md) von [Sei Network](https://alphagrowth.io/de/sei-network.md)
  [X-Profil](https://twitter.com/SeiNetwork)
- [Kiwibig.eth](https://alphagrowth.io/de/spaces/participant/kiwicryptobig.md)
  [X-Profil](https://twitter.com/KiwiCryptoBig)

## Mehr TVL für dein Protokoll oder deine Blockchain?

AlphaGrowth steigert den TVL von Protokollen und Blockchains durch Asset-Akquise und -Integration, Grants, Incentives und LP-Deals. Wir haben diese Daten aufgebaut und halten sie kostenlos.

[So steigern wir TVL](https://alphagrowth.io/de/defi/index.md) [Sprich mit AlphaGrowth](https://alphagrowth.io/de/contact-us/index.md)
