7Block Labs
Blockchain Technology

ByAUJay

Summary: So, rollups and sidechains are basically ways to help scale things up by moving some of the heavy lifting away from a crowded Layer 1 (L1) or by bringing in a whole new group of validators. But the way they're performing and the costs they're facing seem to have hit a bit of a limit. This is mainly because of how data availability is managed and the guidelines for settlement. On the other hand, we've got those real-time settlement platforms--like some of the specialized Layer 1 blockchains and regulated networks--that are really shaking things up. They offer super fast transaction finality, often in less than a second, and can handle a ton of transactions at once. Plus, they usually only charge a few cents per transaction, sometimes even matching the efficiency of central bank settlements. Pretty cool, right? This change is seriously shaking things up, especially when it comes to costs, speed, and the options you'll have to deal with in your operations.

Can You Explain How Rollups or Sidechains Differ from Purpose-Built Real-Time Settlement Platforms in Terms of Throughput and Cost?

These days, decision-makers have three great choices when it comes to high-throughput, low-cost blockchain settlement:

So, here’s the scoop on rollups on Ethereum, whether we’re talking about optimistic ones or ZK rollups. They’ve got this awesome feature where they actually tap into the security of Layer 1 (L1). Plus, they take care of the costs related to L1 data availability (DA). How neat is that? So, when we talk about sidechains or L1 appchains, they have their own validators. A good example of this would be BNB Smart Chain or Polygon PoS, which is actually making some cool changes and moving towards validium. Next up, we have the purpose-built real-time settlement platforms. These are some seriously high-performing Layer 1s that focus on nailing that sub-second finality (think Solana, Aptos, Sui, Sei). Plus, we can't overlook the regulated institutional setups, like Fnality or JPMorgan’s Onyx/Kinexys, which are also stepping up the game.

Let’s take a closer look at throughput, finality, and fees, and see how they all stack up against each other. Plus, we’ll chat about the exciting upgrades coming in 2025, particularly Ethereum’s Pectra, and how they’re really shaking things up for rollup economics. It's definitely a topic worth exploring!


TL;DR: What changes your throughput and cost most

  • So, let’s talk about rollups for a sec. These days, data availability is really becoming the main focus when it comes to costs. So, with Ethereum's Dencun (EIP‑4844) and Pectra (EIP‑7691) on the horizon, we're talking about these awesome blobs that provide super affordable, short-term data storage. It's a pretty exciting development! This is a total game changer! It basically doubles the blob space you get per block and brings those L2 user fees down to nearly zero. Crazy, right? Your actual throughput really depends on the blob space you have and how well your batch/prover pipeline is working together. (eips.ethereum.org).

Sidechains can really fly when it comes to transaction speeds and keeping fees low, which is pretty awesome. But there's a bit of a downside to consider. They rely on their own validators for security, and those bridges between chains? Well, they’ve had their share of problems in the past. So, while the speed and cost benefits are great, just keep in mind that there are some risks involved! Finality happens pretty fast--usually just a matter of seconds. But the trade-off is that you don’t get those solid settlement guarantees that Ethereum offers. (docs.bnbchain.org).

On the flip side, there are some specialized real-time platforms out there that really stand out. They can process transactions in under a second and handle a massive amount of data at the same time. It's pretty impressive! Let's talk about Solana, for example. You can actually get your transactions done for less than a buck! You get to enjoy slots that run between 400 to 600 milliseconds. Layer 1s like Aptos and Sui are really stepping up their game! They’re achieving finality in less than a second thanks to their parallel execution capabilities. Pretty impressive, right? If you're running a bank or any kind of financial institution, you might want to check out Fnality or JPM's Onyx. They’re great options for dependable cash settlements that align with what central banks expect. These platforms are really changing the game when it comes to latency and how things operate overall. (solana.com).


1) Definitions that matter for cost and throughput

  • Rollups (Optimistic/ZK): These cool guys take care of transactions off Layer 1 (L1) while still keeping Ethereum in the mix. They do this by posting state data using calldata or blobs. Optimistic rollups take a pretty chill approach, like “trust but verify.” They operate under the assumption that everything is on the up and up, unless someone spots a problem and calls it out. Typically, you’d be looking at a withdrawal window of around 6... 4 to 7 days). On the other hand, ZK rollups really help speed up finality on Layer 1 by using proofs. You'll get user confirmations pretty fast, but just keep in mind that the "official" L1 settlement still depends on those challenge and proof windows. Check out more here.
  • Sidechains/appchains: Think of these as their own mini-universes! Each one has its own set of rules, consensus methods, and a team of validators keeping things in check. So, these options are usually quick and don’t break the bank, which is great! Just keep in mind that they don’t come with the same level of security that Ethereum offers. If you want to dive deeper into it, just check it out here.
  • Custom-built real-time settlement platforms: We’ve got some seriously impressive public L1s out there that are built for lightning-fast performance and super low latency. Take Solana, for example; it’s rocking those 400-600 ms slots. Then there's Sei, which is shooting for about 400 ms when it comes to deterministic finality. And let’s not forget Aptos and Sui--they’ve got their pipelines down to under a second! Pretty cool stuff! If you're interested in exploring that topic further, you can check it out here. Hey, just a heads up--there are some pretty cool settlement options out there specifically designed for institutions. Take Fnality, for example; they’ve got this impressive settlement finality designation. Then there’s Onyx/Kinexys, which is doing some serious business, moving around $2 billion every single day! How wild is that? You can check out all the details over here.

2) Throughput in practice

Rollups: bounded by blobspace and your batching/prover

Alright, so here’s the deal: EIP-4844 introduced this new concept of "blob-carrying" transactions. Each one is 128 KB and they stick around for about 18 days before they get pruned. It's pretty cool how this all works! They’ve got their own fee market, so they’re priced separately. Then on May 7, 2025, Pectra rolled in and really shook things up by increasing the target and maximum blobs per block from 3/6 to 6/9. Pretty cool, right? This change really boosts how much data is available and helps make those fluctuating posting costs a lot more stable. What’s really interesting is that it ramps up the effective Layer 2 transactions per second (TPS). Plus, it helps keep user fees down, especially when there isn't a ton of demand for blobs. Pretty neat, right? (eips.ethereum.org).

Operators are really enjoying those "almost-free" blob times since Pectra rolled out. It looks like, on average, their usage is staying below the new target, which is great news! The total daily capacity has soared into the multi-GB per day range. So, what this really means is that more Layer 2 operations can take place in every batch. And that’s great news because it ultimately leads to lower costs for each operation. (ethmrc.com).

Hey, just to give you a quick snapshot, OP Mainnet has been buzzing lately! They recently reached a daily operation high in the low millions, handling around 17 to 28 user operations per second (UOPS). Pretty impressive, right? Sure, while raw throughput numbers can look impressive, the actual rates we see in real life really depend on a few key factors. It all boils down to things like how much demand there is, how frequently batches get processed, and the limits on data availability. (l2beat.com).

Sidechains: higher TPS, fast blocks

The BNB Smart Chain (PoSA) is all about speed! You can look forward to block times that are around 3 seconds, and when it comes to getting that final confirmation, you're looking at just about 6 seconds. Pretty quick, right? Typically, you can expect transactions per second (TPS) to be in the hundreds. Plus, you'll love the low and predictable fees! With larger gas limits and fewer bandwidth issues than Ethereum L1, it’s definitely a smoother experience. (docs.bnbchain.org).

Purpose-built real-time L1s: parallelism and pipeline

  • Solana: So, the slot times are roughly between 400 to 600 milliseconds, which is pretty quick! Plus, when it comes to fees, you’re typically only paying a tiny fraction of a cent. It's super affordable! Recent tests with Firedancer and Frankendancer have shown some pretty impressive results: Solana can actually handle over 1 million transactions per second when it needs to! Of course, on a regular day, the mainnet usually manages a few thousand transactions per second, depending on the workload. It's amazing to see just how much potential this platform has! The whole setup really emphasizes parallel execution, all thanks to Sealevel, and it also boasts some high-bandwidth networking. Want to learn more? Head over to solana.com and dive in!
  • Aptos: They've really been shaking things up with their cool features like Block‑STM for parallel execution. Plus, their latest research, Zaptos and Raptr, is looking pretty promising. All this is helping them achieve sub-second latencies without compromising on high transactions per second (TPS) during testing. They're always working to shrink the end-to-end latency, especially in setups spread across different locations. If you want to know more, feel free to check out theblock.co. It’s a great resource!
  • Sui: They’re taking a unique approach with a DAG-based setup, powered by Narwhal/Bullshark and the Mysticeti protocol. The goal? To hit those commit times around 250-500 ms during their testing phase. Pretty ambitious, right? The mainnet has been pretty impressive lately, showing off some serious performance with thousands of transactions per second and confirmations happening in under a second! It’s pretty cool to see how it’s leveraging that object-centric parallelism to make it all work. Check out all the juicy details at blog.sui.io. It’s definitely worth a look!
  • Sei: Sei is putting a lot of effort into perfecting Tendermint with their Twin Turbo setup. They're shooting for around 400 milliseconds for that rock-solid finality, which is awesome news for EVM developers and anyone working with CLOB-style applications. If you want to learn more, just hop over to docs.sei.io. There’s plenty of info waiting for you!

Institutional settlement platforms are all about making sure there’s a straightforward and legally binding process for settling transactions. They prioritize that over just showcasing their transaction per second (TPS) rates. Here's why that’s important:.

Fnality, also known as Sterling FnPS, is really shaking things up! It's actually the UK’s very first regulated DLT wholesale payment system. Pretty cool, right? It really comes down to making sure that settlements are final and secure, and this ties in with RTGS and CHAPS systems. This setup is tailored just for intraday, real-time wholesale settlements. Take a look at this: fnality.com. You’ll find some interesting info there!

So, at JPMorgan, their Onyx/Kinexys division is doing some pretty amazing stuff. They're settling around $2 billion daily using their own specialized systems. That's quite the feat! Right now, they're diving into tokenized flows and looking into deposit tokens across various chains--like Base, for instance. If you want to dive deeper into this, check out this article: (coindesk.com). It’s got all the juicy details!


3) Cost today (Dec 2025): what you’ll actually pay

Rollups: post‑Dencun/Pectra fees are down materially

Just a heads up, here are some live snapshots for you! But remember, fees can fluctuate depending on how much blob supply and demand there is, as well as the current price of ETH. So, keep an eye on those factors!

If you want to stay in the loop, head over to L2Fees.info for the latest updates. Right now, transferring ETH on the main Layer 2s is costing around $0! 07 and $0. So, it's 19 right now, and those swaps are hanging out around zero. 18 to $0. It really comes down to the stack and proofs you're working with, so it could be around 57. When we talk about user costs, we're actually considering a couple of things: the L2 execution and how we spread out the costs of L1 data availability over time. (l2fees.info).

So, what's the scoop? Well, blobs are pretty cool because they help keep rollup data availability separate from the gas costs you deal with on the EVM. After about 18 days, they get pruned, which is a fancy way of saying they get cleaned up. This means that posting costs can be anywhere from 10 to 100 times cheaper than calldata at certain points. Pretty wild, right? On top of that, Pectra's aim for blobs (6/9) has definitely cranked up the supply! (blocknative.com).

ZK vs. Optimistic Cost Nuances:

When you're getting into layer-2 solutions, it's super important to wrap your head around how Zero-Knowledge (ZK) rollups stack up against Optimistic rollups in terms of cost. Let me break it down for you a bit:

1. Transaction Fees

  • ZK Rollups: So, these guys tend to keep transaction costs down since they group a bunch of transactions together into a single proof. Pretty neat, right? This really makes everything so much more efficient!
  • Optimistic Rollups: These can sometimes come with higher fees, especially when the network gets crowded. That’s because they use a fraud-proof system, which can add a bit of extra processing on top.

2. Proving Mechanisms

  • ZK Rollups: These bad boys use some fancy cryptographic proofs known as zk-SNARKs or zk-STARKs to get the job done. Sure, setting up these proofs comes with some upfront costs, but the great thing is that once they’re in place, they really help keep transaction fees down in the long run.
  • Optimistic Rollups: So, here’s the deal with these - they work on the idea that transactions are good to go right off the bat. When someone challenges a transaction, it sets off a whole validation process that can be a bit of a hassle to manage. This extra work often leads to higher fees.

3. Finality Times

  • ZK Rollups: These guys really speed things up! Once a transaction is confirmed, it’s a wrap--no waiting around! This means you get quicker confirmations, making the whole experience a lot smoother and more enjoyable for everyone involved.
  • Optimistic Rollups: So, here's the deal--finality can take a bit longer with these because they have this challenge period that usually lasts about a week. It's all about making sure there’s no funny business going on, so they take their time! In certain situations, this might slow them down a bit.

4. Security Considerations

  • ZK Rollups: These guys are really impressive when it comes to security. The proofs they use are mathematically verifiable, so you can count on the system without having to second-guess anything. It’s like having that warm, fuzzy feeling of trust!
  • Optimistic Rollups: These rely a bit on the good intentions of the participants involved. If someone decides to cheat, the system has a way of catching them, but it doesn’t happen instantly. It takes a little time to work through the details.

5. Development Complexity

  • ZK Rollups: So, the technology behind ZK proofs can get a bit tricky. It might take some deeper knowledge and skills to pull it off right. This might lead to increased costs for development teams.
  • Optimistic Rollups: So, these are usually pretty straightforward to work with. When things are simpler, it usually means that we can get things done quicker and save some cash along the way.

Conclusion

When it comes to the ZK versus Optimistic debate, each side has its strengths and weaknesses, especially regarding costs and user-friendliness. It really comes down to what you need and how you plan to use it. No matter if you’re into security, speed, or keeping your costs down, getting a grasp on these details will definitely help you make a smarter choice. So, which option do you think fits your project the best?

Sure, ZK rollups do have their share of proof costs, but the bright side is that they steer clear of those frustrating week-long wait times for withdrawals. When it comes to Ethereum, getting that hard finality can take a bit of time--typically, you're looking at a wait of about 3 to 6 hours or even longer for some setups. It really depends on how frequently those proofs get grouped together. So, when it comes to Linea, it’s currently showing a gentle finality of around 2 seconds, which is pretty cool. The hard finality, on the other hand, can take anywhere from 6 to 32 hours right now. However, the good news is that this is expected to speed up as time goes on! If you want to dive deeper, feel free to check out more details here. Happy exploring!

Hey there! So, with Optimistic rollups, you can expect super quick soft confirmations, almost right away. Pretty cool, right? So, if you're planning to make a withdrawal to L1, just keep in mind that you might have to wait about 6. It usually takes about 4 to 7 days, but we can definitely tweak that if needed! To simplify things, there are some liquidity networks--often called “fast withdrawals”--that can help you move things along a bit quicker. Just a heads up, though: they might bring along some trust concerns or extra fees, so it’s good to keep that in mind! If you're looking for more details, just check out this link. Happy exploring!

Sidechains: low fees, different security model

When you’re dealing with BNB Smart Chain, you’ll find that the fees are super reasonable--usually just a few cents! Plus, you can count on block times being around 3 seconds, which is pretty quick. Costs tend to stay pretty steady, thanks to the higher gas limits and a tighter group of validators. Just a heads up: the security and governance aspects here are a bit different from what you’d find with Ethereum L2s. (docs.bnbchain.org).

Purpose-built real-time L1s: fractions of a cent at web scale

  • Solana: Right now, the base fee is around 5,000 lamports, which is roughly 0. 000005 SOL). Even when those priority fees come into play during busy times, you're typically just looking at basic transaction costs of around $0. 0005 to $0. So, when the price of SOL hits $100, it's labeled as 001. Even though those program hotspots can drive up priority fees, the median costs are still pretty low when you compare them to EVM L1. Feel free to dive into more details by clicking here! You'll find everything you need to know!

Regulated settlement rails: fee schedule depends on membership and structure

Fnality and other wholesale systems like it are all about using top-notch, central-bank-level settlement assets. They also have some unique fee structures for participants, steering clear of the typical public-gas auction approach. The real economic perks actually come from saving on intraday liquidity and boosting operational and netting efficiencies, rather than from those minuscule transaction fees. (fnality.com).


4) Settlement and finality: what “real-time” means in each model

  • Optimistic rollups: When it comes to user experience, it’s almost instant! So, if you want to withdraw or send messages back to L1, just keep in mind that there's a challenge period that typically lasts about six days. It's something to think about! It usually takes about 4 to 7 days, especially when you’re working with Arbitrum-type stacks. Many apps deal with this by teaming up with liquidity providers to make those quick exits happen. " (docs.arbitrum.io).
  • ZK rollups: Get ready for a super speedy user experience! Just keep in mind that when we talk about L1 hard finality, it really hinges on proof posting and verification. At the moment, it takes a few hours, but some vendors are really trying to speed things up and get it down to just a few minutes. So, for instance, Linea is currently hitting a hard finality time of around 6 to 32 hours. But guess what? They’re working on speeding things up even further! Pretty exciting, right? (docs.linea.build).
  • Ethereum Bridges (Operational Practice): Take a look at Circle’s CCTP! It's pretty neat because it relies on real block confirmations to get things done smoothly. So, here’s the scoop: when it comes to attestation times, Base, OP, and Arbitrum each clock in at roughly 8 seconds. Solana is hanging out in that same ballpark at about 8 seconds too. Meanwhile, Ethereum takes a bit longer, coming in at around 20 seconds on the quicker route. Just a heads up, this is really focused on "operational finality" rather than full epoch finality. So, as you’re mapping out your fund flows, keep that in mind! (developers.circle.com).
  • High-performance L1s: So, these platforms are really pushing the envelope with their super fast finality times--think sub-second. Sei, in particular, is all about nailing that down! Solana usually wraps things up pretty quickly, finalizing within just a few slots and taking only a few hundred milliseconds for each one. At the same time, Aptos and Sui are really shaking things up with their super-fast commit paths that take advantage of parallel pipelines--talk about cutting-edge! (docs.sei.io).
  • Regulated RTS: So, Fnality is all set up for settlement finality in sterling. This means it can do some really interesting stuff with intraday wholesale DvP (Delivery versus Payment) and PvP (Payment versus Payment) designs. Pretty exciting, right? It's not just about probabilistic finality here; it’s really about achieving “legal finality” on systems that are as solid as those used by banks. (fnality.com).

5) Security and risk surface that influence your cost of capital

Rollups make use of Ethereum’s security to ensure that data is available and that the state is validated. Thanks to the updates in EIP-4844/7691, we've got a lot more blobs available now! This is great news because it helps bring those fees down while still maintaining a strong security framework. So, it’s a win-win all around! There’s definitely some risk involved with centralized sequencers, but the good news is that people are actively working to make things more decentralized. (eips.ethereum.org).

So, here’s the deal with sidechains: they have their own validators and bridges, which is cool, but let's not ignore the fact that there have been some major hiccups in the past. I mean, just look at the Ronin hack or that whole mess with the BSC Token Hub--those were some huge losses! It's a reminder that while sidechains can be awesome, they definitely come with their risks. That's why it's really crucial to have insurance, solid custody controls, and circuit breakers in place. They're essential for keeping everything in check! (coindesk.com).

Alright, so here’s the deal: shared sequencers are stepping up to address the whole centralization issue that's been a bit of a headache for rollups. But honestly, the outcomes so far have been pretty hit or miss. So, here’s an example: Astria kicked off its mainnet in 2024, but by December 2025, they decided to pull the plug on it. Crazy how quickly things can change, right? If you're counting on these, just be sure to watch out for any vendor risks and check those neutrality service level agreements (SLAs). They can really make a difference! (astria.org).

So, when we talk about institutional settlement networks like Fnality and Onyx, we're diving into the permissioned side of things. This means they've got some pretty strict rules about who gets to join in. Plus, they've got cash on their ledgers that’s backed by central banks or bank liabilities. It's a secure setup, for sure! This setup does a great job of reducing counterparty and bridge risks, which is definitely a plus. However, it does mean we lose a bit of that open composability we often like to have. (fnality.com).


6) Concrete examples and what they cost/throughput

  1. Using stablecoins for everyday consumer payments on a larger scale. Visa is really upping its game by broadening its support for stablecoin settlements! They're now diving into Ethereum and Solana, and they're also bringing Stellar and Avalanche on board! This shift is really focused on making payment options quicker and more affordable, especially for USD retail. Thanks to Solana, users get to experience super low fees--often just a fraction of a cent--and lightning-fast transaction speeds, all on a massive scale. It's pretty impressive! For more info, just take a look here.

PYUSD is really making a splash! You can find it on Ethereum and Solana, and it's gearing up for a launch on Arbitrum in 2025. Exciting times ahead! This allows teams to really tailor their user experience and security measures according to what they need for every single transaction. If you want super-fast user experiences that can reach people all over the globe, you should definitely check out the integrations on Solana and Arbitrum. They really lower transaction costs and reduce wait times, especially when you compare them to just using Layer 1. It makes a big difference! If you're curious to dive deeper, feel free to check it out here. There's some great info waiting for you!

2) High-frequency trading or CLOB settlement

So, when you're talking about rollups, you really have to consider the limitations that come from L1 data availability and the frequency of L2 blocks being generated. If you're looking to keep your order placements or matches under 500 milliseconds in terms of determinism, it could be a good idea to take a look at a specialized Layer 1 solution. Imagine Sei, which operates in about 400 milliseconds, or take Solana, where you’re looking at speeds under a second. Pretty impressive, right? On the flip side, you might want to check out an appchain that's tailored for a Central Limit Order Book (CLOB), like dYdX v4's Cosmos appchain. It's pretty cool because it has these super quick 1-second blocks and handles order propagation off-chain, which can really make a difference! If you're curious to learn more about that, just check it out here.

On Solana, the super fast parallel runtime and crazy low fees mean that we can actually run fully on-chain order books, like the ones from Phoenix or Serum. How cool is that? With this setup, you can count on getting quick, reliable fills even when managing larger volumes. And honestly, the transaction fees are just small change compared to the profits you’re likely to rake in from your strategy. Curious to learn more? Just click here and explore!

3) Intraday Repo and DvP for Institutions

  • Fnality: So, this is a live sterling payment system that comes with some pretty good safeguards to ensure that once a settlement is made, it really sticks. The pilot programs for intraday repo and cross-chain DvP are really looking good! We've seen the cycles go from taking hours to just minutes, which is a fantastic improvement. On top of that, they're cutting down on balance sheet expenses when you compare it to the existing systems. For banks, those savings can really pile up, often outweighing the transaction fees they incur for things like gas. If you’re looking for more info, you can dive into the details here.
  • JPM Onyx/Kinexys: So, I’ve heard they’re settling around $2 billion every single day! Pretty impressive, right? Plus, they’re diving into some really cool pilots for tokenized assets. Can't wait to see how that all unfolds! Just a quick note: everything here is managed by the bank, so it doesn’t depend on public mempool settlement. If you want more detailed info, check it out here. It’s a pretty interesting read!

7) Emerging best practices (late 2025)

For Teams Considering Rollups:

  • Embrace Blob-First: Let's kick things off by thinking that affordable blob data availability (DA) is going to be the norm once we move past Pectra. When you're designing your batchers, make sure they fill up those blobs whenever possible. It’s also a good idea to compress any state differences to save some space. And hey, don't forget to tweak your posting intervals based on what’s happening in the market--keeping an eye on those trends can really pay off! Make sure to watch the blob base fee closely and aim for utilization to be as close to the target as possible. If you're curious to learn more about it, just click here. There's some interesting info waiting for you!
  • Model End-to-End Finality: It's important to clearly show the difference between "soft" and "hard" timelines in how users experience things. So, let’s break it down a bit. When it comes to Base stages, you’re looking at roughly 200 milliseconds for a flashblock. Then, it takes around 2 seconds to get that layer 2 inclusion sorted out. After that, we’re talking about nearly 2 minutes to batch everything up and send it to layer 1. Finally, for that layer 1 batch to be fully finalized, you’ll need to wait around 20 minutes. Oh, and if you're feeling positive about things, be sure to tell users about those 7-day exit options! Having on/off-ramp partners is a smart move--they can really help with liquidity! If you want to dig deeper into the details, you can check it out here.
  • DA Flexibility: If your cost model is really sensitive, you might want to check out modular DA options like Celestia for certain chains. It could be a game changer! So, it looks like the pricing is pretty affordable right now--around a few cents for every megabyte. In fact, for each rollup transaction, it can even dip below a cent. Not too shabby! Before you dive in, it's really important to think about a few key things like trust, how mature the market is, and whether different systems can work together smoothly. Just take a moment to weigh those factors! If you're curious and want to dive deeper into the topic, you can check it out here. Happy reading!
  • Be Smart About Shared Sequencing: When you're counting on shared sequencers to keep everything stable and unbiased, make sure to hash out those SLAs and have some backup plans ready just in case. It’s really important to pay attention to operator health, and just take a look at Astria’s shutdown in 2025 as a big red flag. If you’re looking for more details, just take a look at this link here. It's got all the info you need!

For sidechains/appchains:

  • Strengthening bridges and governance: You know, some of the biggest losses in the game have really stemmed from problems with bridges. You know, it’s definitely smart to think about safety measures like circuit breakers, rate limits, and multi-party controls. They can really help keep things running smoothly and safely! When you think about budgeting for security reviews, try to see it as a regular expense (OPEX) instead of just a one-off payment (CAPEX). It’s like adding it to your monthly bills; it should be part of your routine, not just something you tackle once and forget about. (cointelegraph.com).
  • Don’t let things drag on: It's important to tweak your consensus timeouts and gossip settings depending on where your operations are happening.
    Sure, you can totally go for those BSC-style 3-second blocks with quick finality! Just keep in mind that you'll need to take charge of the validator operations and keep an eye on everything. It’s all on you! (docs.bnbchain.org).

For Purpose-Built Real-Time L1s:

  • Engineer for Hotspots: So, if you’re diving into Solana, just keep in mind that those priority fees are going to spike when things get super busy. It's super important to set up your system with local fee markets and a good request-queue arrangement. So, while the fees are generally low, just remember that latency can sometimes jump around depending on the specific apps you're using. For more details, swing by solanafloor.com. It’s definitely worth a look!
  • Parallelize Everything: If you check out how Move, Sui, and Aptos operate, you'll notice they really excel at object-centric design and conflict-free batching. Have you heard about Sei’s pipelined consensus? It’s a total game changer! It really speeds things up by reducing the wall-clock time for each state transition. Make sure you spend some time tweaking your workflows to really take advantage of that parallel execution. It's worth it! If you want to learn more about this, check out docs.sei.io. It’s a great resource!

For institutions:
If you’re in need of some legal certainty and have to handle claims involving central bank money, you should definitely take a look at Fnality rails. If you're into bank money settlements on a larger scale and want a bit of programmability to boot, you should definitely check out Onyx/Kinexys and the networks tied to Canton. They're doing some pretty cool stuff! Hey, you might want to consider trying out a pilot program for intraday repo or DvP. It could be really helpful to see how much you save on your balance sheet compared to the gas fees. (fnality.com).


8) Decision guide: which path fits your KPI?

If you're trying to find the best deal on fees for end-users while still ensuring speedy web performance and a smooth user experience, you’ve got a couple of solid choices. You can’t go wrong with a high-performance Layer 1 like Solana, or you might want to check out ZK rollups. They help keep those data availability costs down, so you can offer users experiences for just a fraction of a cent. Hey, just a quick reminder to take a moment to double-check those live fee trackers and the latency Service Level Objectives (SLOs). It’s always good to stay on top of things! (solana.com).

If you’re looking to tap into the Ethereum ecosystem for extra liquidity and flexibility, but don’t want to break the bank or compromise on Layer 1 security, you might want to think about launching on platforms like OP, Arbitrum, or zkEVM. They can really help you strike that balance! Just a heads-up to remember to optimize blob usage after Pectra. It’s important to keep things running smoothly! Oh, and make sure to account for the withdrawal user experience! It really depends on whether you're going for an optimistic approach or if you're leaning towards ZK proof latency. Just something to keep in mind! (blog.ethereum.org).

If you need super quick trades that go off in less than a second, definitely look into platforms like Sei or Solana. They’re built for speed. You might also want to consider a custom appchain, like the one dYdX offers. It could be just what you're looking for! In this situation, you might notice that a specialized solution actually does a better job at keeping latency low compared to a general-purpose rollup. Make sure you check the 95th and 99th percentiles thoroughly from start to finish, and try not to get too caught up in just the average. It’s important to see the full picture! (docs.sei.io).

If you're diving into regulated cash settlement options like intraday repo, DvP, or PvP, it might be a good idea to check out what Fnality has to offer, along with the current bank systems. They could have some pretty useful solutions for you! You might want to consider connecting to public chains to reach a wider audience. The important takeaway is that your biggest cost savings will come from how well you manage liquidity and capital efficiency, not just from gas fees. (fnality.com).


9) Quick math: framing cost per million actions

  • Rollup with blobs: If you want to figure out the per-operation data availability cost, here's a simple way to do it: take the number of bytes per transaction and divide it by 128 KB (that's the size of each blob). Then, multiply that by the base fee for the blob. After that, divide everything by the number of transactions in a batch, and don’t forget to add the L2 execution gas. Easy peasy! Once the Pectra update rolls out, you'll notice that the blob base fee tends to stick pretty close to the minimum. This happens because of the 6/9 target/max setup, which gives your amortization curve a real lift when you start packing those batches. Just double-check this against the most recent L2Fees and your batcher stats, okay? Want to dive deeper into this? Head over to the Ethereum Blog for all the juicy details!
  • Solana: So, if you're thinking about doing a million simple transfers, the base fee is pretty much zero. So, when you break it down, 000005 SOL is around 5,000 SOL lamports, which translates to about 5 SOL in base fees. Honestly, even if you add in a few small priority fees, it’s still way more affordable compared to L1 EVM. Hey, don’t forget to keep an eye on those program-specific priority fee increases, especially for contracts that could get pretty busy. It’s worth considering! If you’re looking for more details, check out Solana’s website. They've got a bunch of info that should help you out!
  • Celestia DA (for modular rollups): From what I’ve been seeing in the forums, it looks like the costs for DA are currently hanging around the $0 mark. It looks like you’re talking about the 08/MB based on the current setup. So, if you're working with a 256-byte state difference, you might notice that the DA costs start to pop up around $0.00001 for each operation. Since we're still figuring out the pricing, think of this as a fluctuating OPEX and just keep an eye on it. If you want to learn more about this, check out the conversation happening on the Celestia Forum. It's a great spot to get into the nitty-gritty details!

10) What changed in 2025 that you should bake into roadmaps

So, Ethereum’s Pectra has really upped its game with blob throughput - we're talking about hitting those 6/9 target or max levels. Plus, it’s set a limit on the worst-case calldata, which is super helpful for tackling those rollup data availability bottlenecks. It's exciting to see these improvements! They're also cooking up some more blob scaling stuff, and PeerDAS is on the way next! Take a look at this link: (blog.ethereum.org). It's definitely worth checking out!

So, over in the Solana world, we’re seeing some high-performance Layer 1 clients really stepping up their game. For example, Firedancer is starting to make its mark in the validator sets. It’s exciting to see how this is all developing! This is definitely going to enhance Solana's reliability and increase its capacity. Also, big players in the stablecoin scene, like Visa and PayPal, are expanding their multi-chain settlement efforts. This shows that those lightning-fast, budget-friendly payment systems are actually being used in real-world transactions. If you're looking for more details, check out this link: solana.com. You'll find everything you need there!

Shared-sequencer experiments have definitely come a long way. However, they’ve also shown us some areas that need a little work, especially with things like the shutdown of Astria. It's a good idea to think of these as nice little extras rather than things you absolutely need, at least until the networks show they can really deliver. Check out all the details right here: (unchainedcrypto.com).


Bottom line for decision-makers

If you're on the hunt for Ethereum options with some nice, predictable, and low fees, you should definitely take a look at ZK rollups and optimistic rollups. They're pretty cool! And hey, once Pectra rolls out, make sure to optimize for blobs too. It'll make a difference! This setup is going to give you great performance for pretty much any web app you throw at it. Plus, as the blob supply increases, you’ll notice that your unit costs are going to improve quite a bit. (blog.ethereum.org).

If you're looking for lightning-fast exchange-level latency that's nearly undetectable to the human eye, you should definitely check out purpose-built L1s like Solana, Sei, Aptos, or Sui. They’re going to totally outperform regular L2s when it comes to tail latency and consistency. Hey, just a quick reminder to keep an eye on those p95 and p99 latencies! Also, don’t forget to check how much Transactions Per Second (TPS) you can handle when things get busy. It's super important to stay on top of that! (solana.com).

Looking for a solid option that gives you reliable cash settlement and ensures everything is legally binding? You might want to check out regulated DLT platforms like Fnality or take a look at bank rails like Onyx and Kinexys. They’ve got what you need! These options basically change the way we think about costs, shifting the focus from gas fees to liquidity and how efficiently we use our capital. Even though it's a bit of a different angle, you can think of it like "real-time settlement" when it comes to legal stuff. (fnality.com).

7Block Labs is ready to help you navigate those tricky trade-offs with a model that’s all set to go! When you're diving into rollups, it’s all about keeping an eye on batch sizes and how blob usage changes over time. You’ll want to think about sizing for validators and the infrastructure you’re working with. Plus, getting a grasp on how the fee market works on platforms like Solana, Sei, Aptos, and Sui is super important. It can really help you make informed decisions! We've got you covered with operational and functional testing for institutional networks, like Fnality and those linked to Canton. This way, you can find that sweet spot between throughput, latency, and cost that your roadmap really needs. Check it out: (canton.network).


Notes and sources

Hey, if you're curious about EIP-4844 blobs and the recent increase from Pectra (happening on June 9th), you should really check out the official EIPs and the Ethereum Foundation's blog. They’ve got all the details laid out! (eips.ethereum.org). If you're curious about live L2 fees, just check out L2Fees.info! It's a great resource. (l2fees.info). If you're wondering about Solana's fees and slot times, definitely check out the official docs! They’ve got all the info you need. (solana.com). Hey there! If you're into high-performance L1 research, you should definitely take a look at what's going on with Sei, Aptos, and Sui. They've got some pretty interesting insights to offer! (docs.sei.io). Hey there! If you’re curious about institutional frameworks, you should definitely check out the Fnality designation and the Onyx volumes. They’ve got some interesting stuff going on! (fnality.com).

  • If you want to get a better understanding of bridge risks, be sure to check out this article! (coindesk.com).

Absolutely! If you’d like, we can dive into your current workload together. Let’s look at things like transactions per user, state diff size, DA policy, and your target p95 latency. We can then organize all that info into a handy one-page “architecture bill of materials.” Sound good? "This will really help us take a closer look at a few different things: a blob-optimized rollup, how Solana handles deployments, and what it’s like to integrate with an institutional RTS. Can’t wait to see how they stack up against each other!" Once we have that info, we can work out the best way to move forward, plus we’ll be able to estimate OPEX and latency for the next year.

Like what you're reading? Let's build together.

Get a free 30-minute consultation with our engineering team.

7BlockLabs

Full-stack blockchain product studio: DeFi, dApps, audits, integrations.

7Block Labs is a trading name of JAYANTH TECHNOLOGIES LIMITED.

Registered in England and Wales (Company No. 16589283).

Registered Office address: Office 13536, 182-184 High Street North, East Ham, London, E6 2JA.

© 2026 7BlockLabs. All rights reserved.