🔍

ROLLUPS - The Ultimate Ethereum Scaling Strategy? Arbitrum & Optimism Explained - YouTube

Channel: Finematics

[0]

so what are roll-ups all about what's

[2]

the difference between optimistic

[3]

and zk roll ups how is our bedroom

[6]

different from optimism

[7]

and why are rollups considered to be the

[9]

holy grail when it comes to scaling

[11]

ethereum

[12]

you'll find answers to these questions

[13]

in this video

[15]

before we begin if you want to learn

[16]

more about decentralized finance and the

[18]

technology behind it

[20]

make sure you subscribe to my channel

[21]

hit the bell icon and enable all

[23]

notifications

[24]

you can also consider supporting us on

[26]

patreon and joining our defy community

[29]

on discord

[30]

ethereum scaling has been one of the

[32]

most discussed topics in crypto

[34]

the scaling debate usually heats up

[36]

during periods of high network activity

[38]

such as the cryptokitties craze in 2017

[41]

defy summer of 2020 or the crypto bull

[44]

market at the beginning of 2021.

[46]

during these periods the unparalleled

[48]

demand for the ethereum network

[50]

resulted in extremely high gas fees

[52]

making it expensive for everyday users

[54]

to pay for the transactions to tackle

[57]

this problem

[58]

the search for the ultimate skilling

[59]

solution has been one of the top

[61]

priorities for multiple teams

[63]

and the ethereum community as a whole in

[65]

general

[66]

there are three main ways to scale

[67]

ethereum or in fact

[69]

most other blockchains scaling the

[71]

blockchain itself

[72]

layer 1 scaling building on top of layer

[75]

[76]

layer 2 scaling and building on the side

[78]

of layer 1

[79]

side chains when it comes to layer 1

[82]

eth2

[83]

is the chosen solution for scaling the

[84]

ethereum blockchain

[86]

eth2 refers to a set of interconnected

[89]

changes

[89]

such as the migration to proof-of-stake

[92]

merging the state of the proof-of-work

[94]

blockchain into the new proof-of-stake

[96]

chain

[96]

and sharding sharding in particular can

[99]

dramatically increase the throughput of

[101]

the ethereum network

[102]

especially when combined with roll-ups

[105]

if you'd like to learn more about the

[106]

eth2

[107]

you can check out this video here when

[110]

it comes to scaling outside of layer 1

[112]

multiple different scaling solutions

[113]

have been tried with some mixed results

[116]

on one hand we have layer 2 solutions

[118]

such as channels that are fully secured

[120]

by ethereum

[121]

but work well only for a specific set of

[124]

applications

[125]

side chains on the other hand are

[127]

usually evm compatible

[128]

and can scale general purpose

[130]

applications the main drawback

[132]

they are less secure than layer 2

[134]

solutions by not relying on the security

[136]

[137]

ethereum and instead having their own

[139]

consensus models

[141]

most roll-ups aim at achieving the best

[143]

of these two worlds

[144]

by creating a general-purpose scaling

[146]

solution while still fully relying on

[148]

the security of

[149]

ethereum this is the holy grail of

[151]

scaling as it allows for deploying

[153]

all of the existing smart contracts

[156]

present on ethereum

[157]

to a rollup with little or no changes

[160]

while not sacrificing security

[162]

no wonder roll ups are probably the most

[164]

anticipated scaling solutions of them

[166]

all

[166]

but what are roll-ups in the first place

[169]

a rollup is a type of scaling solution

[171]

that works by executing transactions

[173]

outside of layer 1

[175]

but posting transaction data on layer 1.

[178]

this allows the rollup to scale the

[180]

network and still derive its security

[183]

from the ethereum consensus moving

[185]

computation

[186]

off chain allows for essentially

[188]

processing more transactions in total

[190]

as only some of the data of the roll-up

[192]

transactions has to fit into the

[194]

ethereum blocks

[195]

to achieve this roll-up transactions are

[198]

executed on a separate chain

[200]

that can even run a rollup specific

[202]

version of the evm

[204]

the next step after executing

[205]

transactions on a rollup is to batch

[207]

them together

[208]

and post them to the main ethereum chain

[211]

the whole process

[212]

essentially executes transactions takes

[215]

the data

[216]

compresses it and rolls it up to the

[218]

main chain in a single batch

[220]

hence the name a roll-up although this

[222]

looks like a potentially good solution

[224]

there is a natural question that comes

[226]

next how does ethereum know

[229]

that the posted data is valid and wasn't

[231]

submitted by a bad actor

[233]

trying to benefit themselves the exact

[236]

answer depends on a specific role up

[238]

implementation

[239]

but in general each rollup deploys a set

[241]

of smart contracts on layer 1

[243]

that are responsible for processing

[245]

deposits and withdrawals

[247]

and verifying proofs proofs are also

[250]

where the main distinction between

[251]

different types of rollups comes into

[253]

play

[254]

optimistic roll-ups use fraud proofs in

[257]

contrast

[258]

zika roll-ups use validity proofs let's

[261]

explore these two types of roll-ups

[262]

further

[263]

optimistic roll-ups post data to layer 1

[266]

and assume it's correct hence the name

[268]

optimistic if the posted data is valid

[272]

we are on the happy path

[273]

and nothing else has to be done the

[275]

optimistic rollup benefits from not

[277]

having to do any additional work

[279]

in the optimistic scenario in case of an

[282]

invalid

[282]

transaction the system has to be able to

[285]

identify it

[286]

recover the correct state and penalize

[288]

the party that submits such a

[290]

transaction

[291]

to achieve this optimistic roll-ups

[293]

implement a dispute resolution system

[295]

that is able to verify fraud proofs

[298]

detect fraudulent transactions and

[300]

disincentivize bad actors from

[302]

submitting other invalid transactions

[304]

or incorrect fraud proofs in most of the

[307]

optimistic rollup implementations

[309]

the party that is able to submit batches

[311]

of transactions to layer 1

[313]

has to provide a bond usually in the

[315]

form of

[316]

eth any other network participant can

[319]

submit a fraud proof

[320]

if they spot an incorrect transaction

[323]

after a fraud proof is submitted

[325]

the system enters the dispute resolution

[327]

mode in this mode

[329]

the suspicious transaction is executed

[331]

again this time

[332]

on the main ethereum chain if the

[334]

execution proves that the transaction

[336]

was indeed fraudulent

[338]

the party that submitted this

[339]

transaction is punished usually by

[342]

having their bonded eath slashed

[344]

to prevent the bad actors from spamming

[346]

the network with incorrect thought

[348]

proofs

[349]

the parties wishing to submit thought

[350]

proofs usually also have to provide a

[353]

bond that can be subject to slashing

[356]

in order to be able to execute a rollup

[358]

transaction on layer 1

[359]

optimistic rollups have to implement a

[362]

system that is able to replay a

[364]

transaction with the exact

[366]

state that was present when the

[367]

transaction was originally executed on

[370]

the rollup

[371]

this is one of the complicated parts of

[373]

optimistic rollups

[374]

and is usually achieved by creating a

[376]

separate manager contract

[378]

that replaces certain function calls

[380]

with a state from the rollup

[382]

it's worth noting that the system can

[384]

work as expected

[386]

and detect fraud even if there is only

[388]

one honest party

[390]

that monitors the state of the rollup

[392]

and submits fraud proofs if needed

[394]

it's also worth mentioning that because

[396]

of the correct incentives within the

[398]

rollup system

[399]

entering the dispute resolution process

[401]

should be an exceptional situation

[403]

and not something that happens all the

[405]

time when it comes to zk roll ups there

[408]

is no dispute resolution at

[410]

all this is possible by leveraging a

[412]

clever piece of cryptography

[414]

called zero knowledge proofs hence the

[416]

name zk rollaps

[418]

in this model every batch posted on

[420]

layer 1 includes a cryptographic

[422]

proof called sdk snark the proof can be

[425]

quickly verified by the layer 1 contract

[428]

when the transaction batch is submitted

[430]

and invalid batches can be rejected

[432]

straight away

[433]

sounds simple right maybe on the surface

[436]

in practice

[437]

to make it work multiple researchers

[439]

spend countless hours

[441]

iterating on these clever pieces of

[443]

cryptography

[444]

and math there are a few other

[446]

differences between optimistic and zk

[448]

rollups

[449]

so let's go through them one by one due

[451]

to the nature of the dispute resolution

[453]

process

[454]

optimistic rollups have to give enough

[456]

time to all the network participants to

[459]

submit the fraud proofs before

[460]

finalizing a transaction on layer 1.

[463]

this period is usually quite long to

[465]

make sure that even in the worst case

[467]

scenario

[468]

fraudulent transactions can still be

[470]

disputed

[471]

this causes the withdrawals from

[473]

optimistic roll-ups

[474]

to be quite long as the users have to

[476]

wait even as much

[478]

as a week or two to be able to withdraw

[480]

their funds back to layer 1.

[482]

fortunately there are a few projects

[484]

that are working on improving the

[485]

situation

[486]

by providing fast liquidity exits these

[489]

projects offer almost

[490]

instant withdrawals back to layer 1

[492]

another layer 2

[493]

or even a side chain and charge a small

[496]

fee for the convenience

[497]

the hop protocol and context are the

[500]

products to look at

[501]

zk roll ups don't have the problem of

[503]

long withdrawals as the fans are

[505]

available for withdrawals

[507]

as soon as the rollup badge together

[509]

with a validity proof

[510]

is submitted to layer 1. so far it looks

[513]

like zk rollups are just a better

[515]

version of optimistic roll ups

[517]

but they also come with a few drawbacks

[520]

due to the complexity of the technology

[522]

it's much harder to create an evm

[525]

compatible zika rollup

[526]

which makes it more difficult to scale

[528]

general purpose applications

[530]

without having to rewrite the

[532]

application logic

[533]

saying this zk sync is making

[535]

significant progress in this area and

[537]

they might be able to launch an

[539]

evm compatible zika rollup quite soon

[542]

optimistic rollups have a little bit of

[544]

an easier time with the evm

[546]

compatibility

[547]

they still have to run their own version

[549]

of the evm with a few modifications

[551]

but 99 of contracts can be ported

[554]

without making any changes

[556]

zika roll-ups are also way more

[558]

computational heavy than optimistic

[560]

roll-ups

[561]

this means that nodes that compute zk

[563]

proofs have to be high-spec machines

[565]

making it hard for other users to run

[567]

them when it comes to scanning

[569]

improvements

[570]

both types of roll-ups should be able to

[572]

scale ethereum

[573]

from around 15-45 transactions per

[576]

second

[576]

depending on the transaction type up to

[578]

as many

[579]

as 1 000 to 4 000 transactions per

[582]

second

[583]

it's worth noting that it's possible to

[585]

process even more transactions per

[586]

second

[587]

by offering more space for the roll up

[589]

batches on layer 1.

[591]

this is also why eth2 can create a

[593]

massive synergy with rollups as it

[595]

increases the possible data availability

[597]

space

[598]

by creating multiple shards each one of

[600]

them able to store a significant amount

[602]

of data

[603]

the combination of eth2 and rollups

[606]

could bring ethereum transaction speed

[608]

to as many as 100

[609]

000 transactions per second now let's

[612]

talk about all the different projects

[614]

working on both

[615]

optimistic and zk roll ups but before we

[618]

do this

[618]

if you've enjoyed this video so far hit

[620]

the like button so that this kind of

[622]

content can reach a wider audience

[625]

back to the video optimism and

[627]

arbitrarium are currently the most

[628]

popular options when it comes to

[630]

optimistic roll-ups

[632]

optimism has been partially rolled out

[634]

to the ether maintenance with a limited

[636]

set of partners

[637]

such as synthetics or unisub to ensure

[640]

that the technology works as expected

[642]

before the full launch arbitrarum

[644]

already deployed its version to the

[646]

mainness

[647]

and started onboarding different

[648]

projects into its ecosystem

[650]

instead of allowing only limited

[652]

partners to be able to deploy the

[654]

protocols first

[655]

they decided to give a window of time

[657]

for all protocols that want to launch on

[659]

the roll apps

[660]

when this period of time is finished

[662]

they will open the flat gates

[663]

to all the users in one go some of the

[666]

most notable projects launching on our

[668]

beach room are

[669]

uni swap sushi bancor auger

[672]

channeling ave and many more arbitroom

[676]

has also recently announced its

[678]

partnership with reddit

[679]

they'll be focusing on launching a

[680]

separate roll-up chain that will allow

[682]

reddit to scale their reward system

[685]

optimism is partnering with maker dao to

[688]

create the optimism die bridge

[690]

and enable fast withdrawals of dai and

[692]

other tokens

[693]

back to layer 1. although both

[696]

arbitrarium and optimism

[697]

try to achieve the same goal building an

[699]

evm compatible optimistic roll-ups

[702]

solution

[702]

there are a few differences in their

[704]

design arbitrarium has a different

[706]

dispute resolution model

[708]

instead of re-running a whole

[709]

transaction on layer 1 to verify

[712]

if the fraud proof is valid they have

[714]

come up with an

[715]

interactive multi-round model which

[717]

allows narrowing down the scope of the

[719]

dispute

[720]

and potentially executes only a few

[722]

instructions on layer 1

[723]

to check if a suspicious transaction is

[726]

valid

[726]

this also resulted in a nice side effect

[729]

where smart contracts deployed on

[731]

our beach room can be larger than the

[733]

maximum allowed contract size on

[735]

ethereum

[736]

another major difference is the approach

[738]

to handling transaction ordering

[740]

and mev arbitrary will be initially

[743]

running

[744]

a sequencer that is responsible for

[746]

ordering transactions but they want to

[748]

decentralize it in the long run

[750]

optimism prefers another approach where

[752]

the ordering of transactions

[754]

enhanced the mev can be auctioned off to

[757]

other parties

[758]

for a certain period of time it's also

[760]

worth mentioning a few other projects

[762]

working on optimistic roll-ups

[764]

fuel the omg team with the omgx

[767]

and cartesi to name a few most of them

[770]

try to also work on an evm compatible

[772]

version of the overall apps

[774]

although it looks like the ethereum

[776]

community is mostly focusing on

[778]

optimistic roll-ups

[779]

at least in the short run let's not

[781]

forget that the projects working on zika

[783]

roll-ups are also progressing extremely

[785]

quickly

[786]

with zika roll-ups we have a few options

[788]

available

[789]

loopring uses zk roll-up technology to

[792]

scale its exchange and payment protocol

[795]

hermes and zktube are working on scaling

[797]

payments using ck roll ups with hermes

[800]

also building

[801]

an evm compatible zkroll app aztec is

[804]

focusing on bringing privacy features to

[806]

their zika rollup technology

[809]

starkware based rollups are already

[811]

extensively used by projects such as

[813]

diversify

[814]

immutable x and dydx as we mentioned

[817]

earlier

[818]

zk singh is working on an evm compatible

[820]

virtual machine

[821]

that will be able to fully support any

[824]

arbitrary smart contracts

[825]

written in solidity as we can see there

[828]

are a lot of things going on in both the

[830]

optimistic

[831]

and the zk roll up camps and the

[833]

competition between different roll ups

[835]

will be interesting to watch

[837]

roll ups should also have a big impact

[839]

on defy users who were previously not

[841]

able to transact on ethereum

[843]

due to high transaction fees will be

[845]

able to stay in the ecosystem

[847]

the next time the network activity is

[849]

high they will also enable a new breed

[852]

of applications that require cheaper

[854]

transactions

[855]

and faster confirmation time all of this

[857]

while being fully secured by the

[859]

ethereum consensus

[860]

it looks like roll-ups may trigger

[862]

another high growth period for d-phi

[865]

there are however a few challenges when

[867]

it comes to roll-ups

[868]

composability is one of them in order to

[870]

compose a transaction that uses multiple

[873]

protocols

[873]

all of them would have to be deployed on

[875]

the same rollup

[877]

another challenge is fractured liquidity

[879]

for example without the new money coming

[881]

into the ethereum ecosystem as a whole

[884]

the existing liquidity present on layer

[886]

1 in protocols such as uni swap or

[889]

other will be shared between layer 1 and

[891]

multiple rollup implementations

[893]

lower liquidity usually means higher

[895]

slippage and worse trade execution

[898]

this also means that naturally there

[900]

will be winners and losers

[902]

at the moment the existing ethereum

[904]

ecosystem is not big enough to make use

[906]

all scaling solutions this main probably

[909]

will change in the long run but in the

[911]

short run

[912]

we may see some of the roll ups and

[913]

ROLLUPS - The Ultimate Ethereum Scaling Strategy? Arbitrum &amp; Optimism Explained - YouTube

ROLLUPS - The Ultimate Ethereum Scaling Strategy? Arbitrum & Optimism Explained - YouTube