Lessons from LAUNCHING 70+ CHAINS | Managing Appchain Infra at Scale. By Dr. Ravi Chamria, Zeeve - video Dailymotion

zeeve

🔴Presented by: Dr. Ravi Chamria, Co-founder & CEO, Zeeve   📍Watch this session if you're building on OP Stack, Arbitrum Orbit, zkSync, Avalanche, Cosmos SDK, or planning to launch an AppChain in 2025  Special thanks to WeAreCommunity.io. This session was proudly hosted by the WeAreCommunity Blockchain Community — a dynamic network of Web3 builders, founders, engineers, and enthusiasts driving the future of decentralized technologies. From hands-on workshops to expert AMAs, WeAreCommunity champions collaboration and practical learning in the blockchain space. Thank you for enabling this powerful session!  The AppChain movement is gaining momentum — and Zeeve has been leading the way, helping Web3 projects go from concept to deployment across 70+ AppChains and 380+ testnets. In this high-impact session, Dr. Ravi Chamria (CEO, Zeeve) shares exclusive insights, real-world hurdles, and proven DevOps strategies to efficiently scale AppChain infrastructure.  Whether you're building a rollup with OP Stack or zkSync, or launching a sovereign L1 using Avalanche Subnets or Cosmos SDK, this session offers practical guidance at every stage — from validator onboarding to observability and service level agreements.  What you’ll learn in this video:   ✅ Why AppChains are the future of Web3   ✅ How modular rollups are evolving across L1s, L2s, and L3s   ✅ 3 primary categories of infrastructure challenges in rollups & AppChains   ✅ Proven practices for sequencer, validator, and node infrastructure   ✅ Upgrade methodologies for rollups like OP Stack, Orbit, CDK, and zkSync   ✅ Best-in-class monitoring, observability, and incident response   ✅ Real-world case studies from 5 leading AppChains powered by Zeeve   ✅ End-to-end support Zeeve provides for AppChain deployments  Zeeve supports:   ⭐ Optimistic Rollups: OP Stack, Arbitrum Orbit   ⭐ ZK Rollups: zkSync, Polygon CDK   ⭐ L1 Frameworks: Avalanche Subnets, Cosmos SDK, Polkadot Parachains   ⭐ Full-stack infrastructure: RPCs, Indexers, Bridges, Monitoring, Alerts, 24x7 SLAs  Case studies highlighted:   ⭐ Memento ZK Chain (zkSync) – Utilized in Deutsche Bank’s Project DAMA 2   ⭐ The Binary Network (Avalanche L1) – Telecom infrastructure for 169M+ users   ⭐ CodeNekt – Automotive blockchain spanning 17 countries   ⭐ ROAR (OP Stack) – DeFi L2 with staking, NFTs, and DAO   ⭐ Skynet (Orbit) – L3 chain enabling AI agent communication  Perfect for:   AppChain founders, Blockchain DevOps teams, Web3 startups, Enterprise blockchain leaders, RaaS platforms, Protocol engineers, Infrastructure architects

Transcript

00:00hello hello everyone and welcome back to blockchain community day i'm thrilled to announce our first

00:12speaker someone who truly understand what it means to operate at web3 scale first things first i'm

00:20faruk i'm gonna be your moderator for the whole conference managing a single blockchain is already

00:27a challenge but what about 60 70 plus chains that's exactly what we're about to explore please welcome

00:36dr ravi chamriya co-founder and ceo of sd india's leading no code blockchain infrastructure as a

00:44service platform empowering over 30 000 users throughout the world correct me if i'm wrong mr

00:51ravi absolutely uh the floor is totally yours um let's dive in yeah uh can you see my screen

01:06properly like the full screen we can see it perfectly okay thank you yeah hi everyone

01:14my name is ravi i'm the co-founder ceo of zeef and first of all before i

01:19um get started i would like to uh thank we are community it's an amazing initiative that they

01:26have and it's a uh pretty amazing conference that they have set up with the incredible

01:34lineup of speakers and and builders and in web free space and uh we at zeef are very very uh

01:42excited to partner with vr community to be part of this conference so uh thanks once again to

01:48vr community now uh uh the uh the topic which i'm covering today is managing app chain infrastructure

01:57at scale so first uh let's understand what app chains are so we have seen uh the enterprise adoption

02:05which started happening in 2016-17 with some of the permission protocols like hyperledger fabric besu

02:13then r3 coda came in as a bank consortium and then quite a few other stacks um so all the enterprises

02:20typically they were launching a consortium network or a network of their own based on a certain

02:26application so for example somebody wants to launch a trade finance platform and along with that they

02:32want to have their own blockchain so we call this an app chain an application specific blockchain

02:37so on the enterprise side we had some of the permission stacks like fabric besu and on the

02:44other hand even in the public domain we were seeing a lot of l1s coming into picture and when

02:50we say l1s i mean layer one blockchains public chains uh that people were launching and some of them were

02:57general purpose chains so general purpose chain you could take an example of for example ethereum

03:03or polygon pos or binance smart chain these are general purpose public chains or l1s on which you

03:10know any developer can go they can deploy their smart contract they can deploy their dApps and start

03:14using it on the other hand there are quite a few app chains and app chains means you're launching a

03:21platform specific to an application now it could examples could be anything somebody launching a trade

03:27finance platform like they were uh marco polo or in any other domain so we see a massive amount of

03:34app chains so the whole concept of app chains is application specific chains where you have your own

03:41layer one chain and it is specific to an application or a set of applications within a domain like for

03:47example injective uh focused on financial applications so that is what what uh we mean by app chains

03:54now why app chains have suddenly uh become such a rage in last two three years and and this i think

04:05is a is a natural progression so we saw that there are lots and lots of dApps being deployed on public

04:10chains like ethereum but then what happens as your dApp grows and you start having more transactions

04:17more users as part of your community then in that case you start fighting or competing for block space

04:23on the public chain and it may lead to higher gas prices as we have seen in quite a few incidences

04:30where based on a new nft project launch or a gaming project launch suddenly the gas prices spiked up

04:39so uh but but when you are launching your own application specific chain what you want

04:44you want complete control over the user experience and for that to happen you need to have you know your

04:50own set of configurations or uh you want to have control over throughput you want to control over

04:56gas economics and so and so forth so that is why app chains started picking up and a lot and a lot of

05:03people started launching their own app chains but one of the major challenge at that time was there

05:08were no proper stacks to launch an app chain forking a code which was a normal norm at that time

05:15taking you know ethereum fork or list fork it it was a nightmare because these code bases were not

05:21written to be replicated and that's why it was very difficult to fork a chain and create your own l1

05:27or or or your app chain then you know came the stacks you know proper stacks that you can use to

05:32launch so first we saw the advent of some of the app chain stacks like cosmos sdk or avalanche subnets

05:39and then substrate para chains uh where you can use some of these stacks and launch your own uh sovereign

05:46chain and then two years back came the concept of rollups again using rollup stacks like op stack or

05:53arbitram orbit you can launch your own app chain uh specifically called roller but it falls under the

06:00general category of app chain so this is basically the reason why app chains have become uh more and

06:09more important and why we are seeing lots and lots of builders launching their own app chains and rollups

06:14according to root scan a recent report which was released about a few weeks back out of almost 2000

06:21active chains 70 percent are app chains or modular rollups across layer one layer two or layer three

06:26and and and this is significant so we are seeing that you know there's a massive advent of uh builders

06:33launching their own app chains and most of this has been powered by some of the stacks which i just

06:37mentioned so uh what app chains provides one it provides superior user experience and user experience

06:46encompasses all the desired characteristics you know whether you want to have your own custom

06:51configurations if you're launching an enterprise chain and you want to have compliances you want to

06:56have data compliances or adherence to regulations or you require high throughput or you require you

07:03know a different kind of consensus or governance model so all this is possible only when you have full

07:08control over the underlying chain and and that's why app chains are can be considered as the future of web3

07:17now in terms of uh evolving landscape uh these are some of the primary stacks that i mentioned about

07:24so if you want to launch an optimistic rollup you have op stack and arbitra morbid primary uh uh optimistic

07:31rollup stacks and then we have zk stacks like zk sync and then we have uh typical app chain stacks you

07:36know launching your own sovereign l1s where we have cosmos sdk a massive ecosystem of cosmos where cosmos is

07:43a hub and you can create your scope chains it can be a completely independent sovereign l1 using cosmos

07:49the cosmos sdk or you can still connect with uh other chains within the cosmos ecosystem

07:54through using uh various interoperability tools similarly we have avalanche subnets which has now been

08:01rebranded as avalanche l1s so you can launch your own sovereign l1s using avalanche l1 stack

08:08and the best part is you can launch from complete permissionless chain to a permission chain

08:14and then you know uh we have para chains uh uh you know you can use substrate framework to launch your

08:20own para chains so these are primary uh stacks uh which you know constitute the uh the app chain

08:27landscape today our journey uh uh we have been in launching actions and rollups for more than four

08:35years now and our journey has been we started with enterprise protocols like fabric besu helped quite a

08:41few enterprises launch their blockchain applications using a complete permission stack highly secured

08:47compliant etc and then we started seeing this trend of launching app chains in the public uh web free

08:53space as well and that's where you know we partnered with some of these stacks and started offering

08:58app chains to different customers and today we are running almost 70 plus production app chains for

09:04various customers across various stacks and almost 350 plus test nets and uh uh uh in our journey you

09:12know there has been massive amount of learning which which uh i'm here to share today and uh we realized

09:20that you know launching your own app chain especially getting the same kind of experience when you deploy

09:25your dab on a public chain it should be similar it was not four years back when you know uh people were

09:32forking um ethereum fork or bitcoin fork or uh different other protocols it used to take hundreds of

09:39thousands of dollars to build such a platform and at the same time uh uh there was a massive cost of

09:46infrastructure managing it and there were massive massive overheads and once these stacks came in you

09:52know we started investing into these stacks and uh created a whole set of infrastructure tools

09:59to ensure that launching an app chain is as easy as launching a smart contract

10:07so uh now coming to the main topic as to what we have learned what what have been our learning

10:14so what are the major challenges when you talk about your own app chain now just to differentiate

10:19further when you are launching a smart contract on a public chain the desired infrastructure is already

10:25there you simply need to deploy your dap you need to deploy your smart contracts and you may need to

10:32use some of the third party integrations which are already available on that public chain so if you

10:36want to use interop or you want to use let's say uh decentralized oracles from chain link or redstone

10:42you can simply plug in because their support is already on on that public chain but when you launch

10:48your own app chain then you need to take care of each and every aspect of it it gives you a lot of

10:54flexibility independence control but at the same time you require uh to manage the entire infrastructure

11:01yourself so that is the downside of it and and that is where you know uh we we come in so first is the

11:07launch phase where you need to understand what kind of custom configurations you require depending on your

11:12use case if it's a d5 use case you may require high throughput if it's an enterprise use case in the

11:18financial services space you may require strict data compliances or let's say in healthcare or it's a

11:25payment use case so very very high throughput and scalability because uh infrastructure demand can go

11:31up and down anytime so depending on your use case uh you do require a lot of custom conflicts so uh that

11:38becomes a challenge as to because once you set the configurations many of the configurations cannot be

11:45changed later uh as you can do in a in a case of a public chain that is not true when you talk about

11:51an app chain and then of course you need to have a very solid infrastructure high availability of nodes

11:57and the sequences and validators all different kind of node infrastructure and then uh validator

12:03onboarding especially if you are launching let's say a poa based chain uh you may have a consortium

12:08where you want to have a permissioned onboarding or let's say you have a pos based public chain

12:15you still want to make it easy for the validators to join your chain

12:19and then smart contract versioning etc so these are some of the pieces that you need to take care

12:23when you are launching your chain the second phase is operations and life cycle management

12:28now that becomes very important because now you need to take care of uh the chain upgrades and all the

12:34maintenance uh overheads so um newer uh updates because all these stacks are like like piece of

12:42software where continuous upgrades and updates will continue to happen so you need to ensure that your

12:47chain always uh updates and upgrades whenever a new version is there and then of course security and

12:54decentralization trade-offs you need to be very clear as to uh depending on the use case depending

12:59on your requirements how much security and compliances need to be there how much decentralization needs to be

13:05there and accordingly uh choose the right right stack and configure it properly and then a highly scalable

13:13infra where you know on-demand scaling can happen uh at the same time you also need to ensure that your

13:20infrastructure is is uh not just highly available but well distributed across the globe so that uh

13:25uh there are you know it leads to higher decentralization at the same time uh lesser risk and more security

13:34and the third phase is reliability and support once you have launched your chain now you want a very

13:40solid management dashboard and monitoring capabilities so that you can monitor the uh chain all your nodes etc

13:47um so that you can be ensured that everything is all the performance criteria are being met and the user

13:54experience is top-notch and then of course it needs to have a very strict slas uh you know different

14:01kind of observability tooling and then in addition to that you also may require different third parties

14:06uh integrations whether it be account abstraction or interoperability or oracles so you need to take

14:12care of all these different pieces as well and uh so based on these challenges what are the best practices

14:21of creating your own app chain with a highly scalable and reliable node infrastructure so first is uh

14:28automate the bootstrapping of nodes so as i mentioned you know depending on whether you're

14:32launching a poa pos or any other different kind of consensus chains depending on whether how

14:38permissionless or permissioned it is you may require uh bootstrapping of nodes so uh why we created one

14:47one click test net the only reason is not just to bring in high level of automation and reducing

14:52human errors but at the same time when you are building you're launching your test net at the same

14:57time you may require some iterations let's say you did some initial conflicts launch your test net

15:02and then you know tried your use case um tried your smart contracts and and then you did some kind of

15:08benchmarking based on performance matrix and other matrix and you realize that okay we need to fine-tune the

15:13conflicts again so that's why you require that test nets can be created again and again on the fly

15:19and it should be very easy to launch a test net then pre-integrate core component so when you launch

15:25your chain there are certain set of core components like your chain will come up with this set of

15:30different kind of nodes it will come up with a block explorer because every chain needs to have a

15:35block explorer where people can see the transaction confirmation different transactions and so on so

15:39forth then you also need to have faucet if you're launching a test net so that you know uh the

15:45builders or or your own developers can use the faucet to mint tokens and use them while testing out the

15:51apps and you may require some essential bridges like when you're launching a roll-up you may require an

15:57l2 to l1 bridge or if you are even launching an app chain using some of the l1 stacks you still may require

16:03interoperability of bridges to bridge some tokens maybe your use case demands you to bridge usdc or

16:09usct uh so you require all these core components and then uh geo redundancies becomes very important

16:19because you want your infra to be highly distributed and if it's permissionless you want more and more

16:24validators to come in from different areas uh and become part of your chain so that again becomes very

16:31very important your rpc uh public rpc should be uh highly performant at the same time reliable so that

16:39uh all the dabs which are running on your chain can perform very properly and then uh at the same

16:46time you need to have uh load balancing active uh failover high availability so all the tenets of

16:53high availability that becomes very very important and then uh one simple thing is use uh um snapshots

17:01so that you know whenever you are let's say if you are upgrading the chain or you need to create a hard

17:06flow because you want to make some major changes in your file in your genesis configuration then

17:12having a snapshot makes it very easy to resync the chain and so on so forth uh now if you talk about

17:20the best practices of validator and sequencer now this uh is is very important because any chain is

17:26decentralized to the extent how their validators are performing so you need to have good number of

17:31validators and at the same time those validators must be active they must be validating the chain

17:36properly with high uptime and so on so forth in the case of rollups uh there's no concept of validator

17:43per se instead of that you have sequences and their sequences again become very important because

17:49sequences are the ones which are you know sequencing the transactions that are happening on the chain

17:55and then you know passing on using those transaction it relays it to data availability layer

18:01or to create generate the proofs whether optimistic proofs or zk proofs so it's it's very important

18:09uh to adhere to all the different best practices for managing your validator and sequencer network

18:15so one is as i mentioned onboarding access so that anyone can easily become a validator on your chain

18:22now it can be a simple documentation how to onboard and people can launch their own validator node on your

18:27chain or it can be uh resorting to platform like z to create one click uh validator onboarding so that

18:35even non-technical users can simply uh make few clicks and launch a validator on your chain

18:41and similarly same goes for the sequencer um or let's say in the case of optimistic proofs verifier nodes

18:49so same as goes for validator nodes in the sovereign ones similarly having verifier nodes or sequencer

18:55making it very very easy and highly available and then version control for consensus upgrades

19:02uh that is again very very important uh so that whenever a change is happening the validator binary

19:09also needs to be upgraded so we need to ensure that all the validators have been updated with the

19:14newest code base

19:15and uh in addition to that you know having a very solid 24-7 proactive monitoring is very important

19:24so that you can see because when you are launching a chain at the start you may have a lesser number of

19:29validators so let's say you start a chain with eight to ten validators now for example in the case of avalanche

19:34l1 if uh your more than 20 percent of validator nodes are not active then your chain may go down

19:42so that's why it is very very important to uh ensure that the validators are active and they

19:48are performing their job then best practices for rollups and upgrades um again uh one is you know

19:56each roll-up stack most of the roll-up stacks are comparatively new we can still say that optimistic

20:02roll-up stacks like op stack and arbitram orbit are very very mature but zk roll-up stacks like uh

20:07zk sync etc are still in the process of maturing we are now seeing a lot of production chains coming

20:13out of zk stack but still they're a long way to go so there are continuous updates and upgrades that

20:18happen so to ensure that whenever there's a release uh the release is applied to your your chain which

20:24has been built using one of those stacks then having you know orchestrate uh safe version rollouts

20:32um use canary deployments before full rollout and then you know manage app layer versioning separately

20:39against super important and then uh to top it all you know have a sound monitoring in place to monitor

20:45not just one node type but you know rpc nodes your sequencer nodes your prover nodes your verifier nodes

20:52and then you know your public rpc l2 and l1 let's say you're creating a l2 roller then both l2 rpc as well as

20:57the l1 rpc then uh uh monitoring and observability uh uh is is very very important especially as your

21:08chain grows initially when you launch an app chain it may be your you may be end up running all your

21:14validators and rpc uh comparatively it is much easier to manage that kind of infra but as your chain

21:20grows if it's a permission chain with a consortium then different consortium members will start setting

21:25up the nodes if it's a permissionless chain then public nodes will start coming into picture and

21:31uh there would be public rpc usage as well so in that case uh having a high performance is very very

21:37important so uh in this case you know having 24 7 monitoring in place having a proper alerts and

21:45notification system so that automatic alerts can come in and at the same time some proactive alerts

21:52so depending on you can define some of the matrix like if the cpu usage goes beyond 60 percent if

21:57the memory usage or or storage is is um you know getting blocked up so all our bandwidth is getting

22:04clogged so all those kind of alerts and notifications into need to be in place so that proper monitoring can

22:10happen so both proactive monitoring as well as 24 7 monitoring uh becomes very very important

22:15um now in terms of slain incident management um so slas ensures that your chain is always active it is

22:25performing as per the uh criteria and at the same time you know all updates upgrades etc are being taken

22:32place and this becomes very very important because uh today user experience has been one of the most

22:38talked about challenge in the web3 space and if you're launching your own app chain then the entire

22:44responsibility of having an amazing user experience falls on you so that's why uh um having you know

22:51proper slas to manage the chain becomes very important and then you know automate the incident

22:56response having an l1 l2 support so that um all the uh incidents can be easily managed whether it be

23:04security or non-security incidents and then uh uh uh you know create learnings out of each incident that

23:10has happened so that uh the next iteration of chain is much stronger and much much much more reliable

23:20uh i would like to create some case studies here um so memento zk chain is one of the latest chain

23:28that we launched about few weeks back it is one of its kind it's an institutional grade

23:35consortium chain or a permission chain uh using zk sync previdium

23:40it's the first chain using previdium and it is part of uh project guardian with deutsche bank being

23:47the first uh part of it uh it is built using zk chain a zk rollup with a validium setup or a previdium

23:55setup uh so validium setup in a rollup is uh uh there you have control over your data and previdium is

24:02you know where you have even better compliances of your data then another example is a binary network

24:09which is a startup in the telecom space they already have more than 40 to 45 million wallets

24:16um and then there are a lot of different kind of apps are being launched from the network

24:20it is uh has been built using avalanche l1 it's a complete sovereign l1 chain permissionless and uh

24:29different uh users can join the chain and and uh already most of their wallet customers are joining it

24:36and as they partner with more and more telecom players we are going to see much much more traction

24:40the best part is using avalanche l1 scalability is not a challenge it's a very lean stack and you can

24:46launch a chain which can scale to any numbers then we have code neck which is again uh uh it's more of a

24:55uh between permissionless and permission it's a it's a digital product passport use case in the

25:03automotive space where they are tracking each and every automobile and their maintenance history and

25:09their logs etc uh using blockchain bringing transparency uh for not just the vehicle owners

25:16but also the insurance players and and other participants in the automotive industry it again

25:23has been built using avalanche l1 with some permission in place especially ensuring the privacy of data

25:29like the pia data and so on support then we have road which is uh um uh it's it's built on opstack

25:38it is a layer to optimistic rollup settling all their transactions on ethereum chain um and and they

25:45have a very strong ecosystem it's a d5 ecosystem that they have launched including nfts including uh dex

25:53and cross-chain interoperability of liquidity and then we have skynet which is uh built using arbitram

26:00orbit another optimistic rollup stack it's a layer three chain so when we say layer three that means

26:06your settlement layer is layer two so in this case arbitram nova is the settlement layer and uh skynet is a

26:14orbit layer three chain and it is purpose built for ai agent communication uh designed for high speed

26:20coordination and low latency task execution so uh key takeaways uh if you talk about so one you know

26:30all app chains need a production ready highly scalable uh uh infrastructure and and that kind of mindset

26:36is required so just imagine in web 2 when you are launching an enterprise grade application

26:42infrastructure becomes super super important similarly when you are launching an app chain your

26:46infrastructure requires um um you know a clear strategy clear planning and clear execution

26:54and then you know crosstack experience is critical so uh at times you know we provide we provide a

27:00massive consulting on which stack to choose depending on your use case depending on your ecosystem preference

27:06depending on uh what kind of smart contracts you want any language preferences etc you know choice of the

27:14right stack becomes very very important and that requires you know gaining expertise across different

27:19stacks and then bringing automation 24 7 monitoring uh again becomes very important so that you can spin up

27:27test nets on the fly test it out and ensure that your infrastructure is highly reliable at the same time for

27:32high performing and then you know having sound monitoring in place so to ensure the longevity

27:38performance and user experience and uh um so this is uh these are some of the key takeaways based on

27:46our experience of uh uh running so many app chains and rollups today um if you want to build your own

27:54app chain you know z provides a complete automated platform to launch your test nets in just a few clicks

28:01and at the same time a complete uh platform to launch your production ready main nets or production

28:07app chains and rollups it includes you know helping you onboard to monitoring to uh provide third-party

28:15integrations with 60 60 plus partners that we have and then you know ongoing monitoring and management of

28:20it to ensure that your chain is top-notch and up and running so this is uh all which i wanted to cover

28:28um if you have any questions uh you can cover on

28:41you

Lessons from LAUNCHING 70+ CHAINS | Managing Appchain Infra at Scale. By Dr. Ravi Chamria, Zeeve

Category

Transcript

Recommended