Blob Schedules and Capacity Planning: Forecasting DA Costs for L2 Teams

Layer 2 teams can now think of using blobs like handling a utility bill. This guide will help you figure out how to size, schedule, and budget for Ethereum blobspace after Pectra. Let's dive in! Plus, it'll help you prepare your capacity model for the PeerDAS and BPO increases that are rolling out this December. We’re going to dive into everything with some handy formulas, up-to-the-minute parameters, and tips for setting things up like a pro. Take a look at this: (blog.ethereum.org). It's definitely worth a peek!

TL;DR for decision‑makers

So, on May 7, 2025, Ethereum launched the Pectra upgrade, and it made quite an impact! It really ramped up its blob throughput, bumping the goal from 3 to 6 and pushing the max from 6 all the way up to 9. Pretty impressive, right? So, it turns out the daily target capacity shot up to roughly 43,200 blobs, which is around 5. You’ve got a max capacity of about 64,800 blobs now, which is pretty cool! The data flow is currently at a rate of 3 GiB per day. Just thought I’d share that tidbit! 9 GiB/day). Ever since the Pectra upgrade rolled out, blob fees have pretty much been sticking to the minimum level. If you want to dive deeper into the details, just click here.

Let’s jump ahead to December 3, 2025, when Fusaka rolled out their new creation, PeerDAS. Beginning the week of December 9, they rolled out the “Blob-Parameter-Only” (BPO) forks. They initially set the per-block target and max to 10/15 and then gradually increased it to 14/21. With these adjustments, we’re looking at a target capacity of around 8. You start off with about 8 GiB each day (at 10), and then it’ll gradually bump up to around 12. 3 GiB/day (at 14). If you’re looking for more info, you can find all the details here.

• Here’s a handy budgeting tip for you: right now, the blob base fee tends to be at its lowest point, which is just 1 wei for every blob gas. So, what this really boils down to is that most of your costs are going to come from the execution gas for those type-3 transactions that include the blobs. Usually, this ends up costing around $0. 50 to $3. Typically, it’s around $00 for each transaction when everything's running smoothly. Just a heads up--when there's a lot of excitement around events like airdrops or inscriptions, prices can really shoot up, sometimes reaching several dollars for each blob. It’s a good idea to set up schedules and throttles to help manage demand effectively. If you're interested in diving deeper into this topic, you can check it out here. It’ll give you some great insights!

---

1) What changed in 2025--and what changes again in December

Pectra, which launched on May 7, 2025, came with the introduction of EIP‑7691. This update bumped the blob target/max from 3/6 to 6/9. Alright, let's break it down a bit. If you do the calculations, we're aiming for a daily target of around 6 blobs for each block. Each block is 128 KiB, and we’re processing about 7,200 blocks in a day. When you put all that together, it comes to roughly 5. 3 GiB/day (in binary). The highest it gets? Almost 7. 9 GiB/day. Networks have largely stayed under this new target, which means that blobspace costs are essentially “almost free.” " (blog.ethereum.org).

• EIP-7623 definitely stirred things up by increasing the floor cost for calldata. This move really helps minimize the worst-case variations in block size and shifts those DA workloads over to blobs instead. Think of calldata as a handy tool to help ease congestion, not just your first choice for everything. (eips.ethereum.org).

So, mark your calendars for December 3, 2025--hello, Fusaka! That's when PeerDAS is set to roll out some pretty cool sampling-based availability checks, all thanks to EIP-7594. Also, Ethereum is planning to increase the "BPO" parameters. They'll kick things off by raising it to 10/15, and then later bumping it up to 14/21 blobs per block, starting on December 9, 2025. Exciting stuff! These changes are really focused on increasing our throughput once PeerDAS steps in. (blog.ethereum.org).

What This Means for Your Budget

If you take a peek into 2025, you’ll see that the cost per byte is likely to drop quite a bit, and we’re also expecting a big boost in capacity compared to 2024. Exciting times ahead! Oh, and don’t forget, there’s another significant increase on the horizon for December. Most of the time, the gas fees from batch transactions are going to be the main player, outpacing blob base fees. But every now and then, when demand really kicks up, that’s when things can flip for a bit. If you want to dive deeper into the details, just click here!

---

2) The blob fee mechanism you have to forecast against

Core Constants (EIP-4844):

So, EIP-4844 brings in some pretty crucial constants that are really important for how the Ethereum network runs its day-to-day operations. Alright, let's dive into what you really need to know:

Constants Overview

• MAX_CONTRACT_SIZE: This is basically the biggest size a smart contract can be. This way, contracts stay manageable and keep things running smoothly.
• GAS_LIMIT: Think of this as the maximum amount of gas you can use for a transaction. It sets a limit on how much you can spend on processing that particular transaction. It really helps to keep things running smoothly by preventing any unnecessary strain on resources.
• BASE_FEE: This is basically the lowest amount you need to pay to get your transactions processed. This fee can change based on how busy the network is, but it gives everyone a basic starting point to work from.
• GAS_PRICE: This is basically how much Ether you're cool with spending for each unit of gas. Getting this set up right can really help make sure your transactions go through on time.

Example Constants in Code

Here's a little sneak peek at how these constants could be set up in code:

uint256 constant MAX_CONTRACT_SIZE = 24576; // Example size limit
uint256 constant GAS_LIMIT = 12400000;      // Example transaction gas limit
uint256 constant BASE_FEE = 1000000000;     // Example base fee in Wei
uint256 constant GAS_PRICE = 20000000000;    // Example gas price in Wei

If developers keep these key points in mind, they can create apps that aren’t just super efficient but also really easy to use.

• Blob size: So, we’re talking about 4,096 field elements, and each one is 32 bytes. When you do the math, that comes out to a pretty impressive 131,072 bytes (or 128 KiB) for each blob.
• Gas per blob: We're looking at a solid 2^17, which comes out to 131,072 blob gas. Plus, the minimum base fee for that blob gas is only 1 wei. Pretty straightforward!
• Retention: So, the nodes have to hold onto those blob sidecars for around 4,096 epochs. That works out to about 18 days in total. Once that’s done, the blobs are trimmed down. (eips.ethereum.org).

Pricing Dynamics:

In the business world, pricing isn't just about slapping a number on something; it’s really a whole strategy in itself. How prices are set can totally affect everything, from how well products sell to where a company stands in the market. Let’s dive into some important factors you should keep in mind when you're checking out how pricing works!

1. Consumer Demand

Pricing usually depends on what consumers are ready to spend. When there's a lot of demand for something, prices can really shoot up. On the other hand, if demand starts to drop, companies might need to cut their prices a bit to draw in customers.

2. Competition

Make sure to keep tabs on what your competitors are up to. If your competitors start slashing their prices, you might want to consider adjusting yours too, so you can keep up and stay in the game. It's always a juggling act to keep your prices appealing while still maintaining quality and profits. You want to attract customers, but you also don’t want to cut corners or lose money. It's definitely a tricky balance!

3. Cost of Production

Hey, just a quick reminder to keep an eye on the costs that come with creating your product. It's super important to factor those in! If production costs go up, you might have to tweak your prices a bit to make sure you're still making a good profit. When you’re figuring out your prices, don’t forget to consider stuff like materials, labor, and overhead costs. It’s super important to keep those in mind!

4. Market Trends

Staying in the loop with market trends can really help you tweak your pricing strategies to fit what’s happening out there. Whether it’s the changing seasons or shifts in the economy, keeping yourself informed can really give you an edge.

5. Value Perception

Customers really want to feel like they’re getting their money’s worth when they make a purchase. It's all about seeing that value in what they're buying! It's not just about how much something costs; what really matters are the perks and the overall experience you'll get from using your product or service. Be sure to express that value in a way that really resonates.

6. Psychological Pricing

You know, how a price is shown can really affect whether or not people decide to buy something. One popular strategy is to price something at $19. It's a pretty clever move because it feels more budget-friendly than rounding it up to $20. You know how that little $1 difference can make a big impact on how people perceive a price? It's all about making it seem more accessible and enticing for buyers.
Switching from $20 to $19.99 can really change how customers view the price. It sounds a lot more appealing, doesn’t it?

7. Regulatory Factors

Make sure you pay attention to any regulations that could impact pricing. It's easy to miss them, but they can really make a difference!
Sometimes, there are laws or industry standards that can really influence your pricing strategies or even cap how much you can charge.

8. Geographical Factors

You know, prices can really change a lot depending on where you are. What works in one market might not necessarily work in another. Each market has its unique quirks and vibes! Adjusting your prices based on different regions can really help you reach a wider audience.

Keeping these factors in mind can really help you find your way through the tricky world of pricing. It’s really about striking that perfect balance--setting a price that draws in customers but also keeps your business healthy and growing.

So, the base fee that comes from EIP-1559 gets tweaked depending on how many blobs are aimed for in each block. So, with Pectra (6/9), when a block is completely full, the blob base fee for the next block tends to rise by roughly 8. 2%. On the flip side, if there aren't any blobs in a block, the fee actually goes down by about 14. 5%. You’ll notice that resets are quicker after quieter periods and that things ramp up more smoothly when it gets busy, especially compared to the old 3/6 system. (blockscholes.com).

Two Implications That Matter to Budgets:

1. Budget Busters: When costs suddenly spike, it can really throw a wrench in your budget plans. For example, if the price of raw materials spikes, it can really mess up your whole financial game plan. It's really important to keep an eye on these ups and downs so you don’t end up going over budget.

2. Revenue Shortfalls: You know how it goes--sometimes the money we thought was coming just doesn't show up like we expected. This might be happening because sales aren't hitting the numbers we expected or maybe clients are taking longer to make their payments. When you're facing these shortfalls, it can really tighten the budget. That’s why it's super important to tweak your spending to make things work.

By wrapping your head around these two points, you can get your budget in order and make sure you're heading in the right direction.

• Almost hitting rock bottom most of the time: When the usage doesn't quite hit the target, you’ll notice that the blob base fee tends to stay super low--usually just chilling at 1 wei per blob gas. Because of this, the blob part of your transaction isn’t really that big of a deal compared to the execution gas. (ethmrc.com).
• Spike mechanics: When things get hectic, like during major airdrops or inscriptions, you might notice the blob base fee suddenly skyrocketing into the multi-dollar range for a few hours. It's definitely something to keep an eye on! Back in March and October 2024, we noticed some pretty big jumps--prices shot up to around $4 to $300 per blob before they finally calmed down a bit. It's definitely smart to brace yourself for those unexpected ups and downs. Having a protective schedule and backup plans in place can really save you from a lot of stress. (coindesk.com).

Tip/Priority Notes and Mempool Behavior:

When it comes to how transactions work in the Bitcoin network, there are a few important things to keep in mind about tips, priority, and what goes on in the mempool.

Tips and Priorities

1. Transaction Fees: The fee you set for your transaction really impacts how fast it gets processed. When the network is congested, paying higher fees usually gets your transactions confirmed quicker.

2. Priority: So, if your transaction has been sitting in the queue for a bit or it’s dealing with a hefty amount of bitcoins, there’s a good chance it could get bumped up in priority, even if the fee isn’t the biggest out there. The reason for this is that older transactions, or those with bigger amounts, tend to have a bigger impact in the mempool.

Mempool Behavior

So, let’s talk about the mempool--it’s basically the hangout spot for all those unconfirmed transactions. And honestly, it’s pretty fascinating how it works!

• Transaction Replacement: So, let's say you sent a transaction with a pretty low fee, but then the network gets all busy and jammed up. No worries! You can actually increase that fee later on using something known as RBF (Replace-By-Fee). It's a handy little trick to help get your transaction moving again! By doing this, you can get miners to prioritize your transaction and process it faster.
• Eviction: So, if your transaction is hanging out in limbo for too long and it’s got a low fee, there's a chance it could be pushed aside to make space for ones that offer better incentives. Make sure to keep an eye on the mempool so you don't run into any issues!
• Size Matters: The size of your transaction, measured in bytes, can really impact how fast it gets noticed. Smaller transactions usually make things smoother for miners, so they often get bumped up the priority list.

If you pay attention to these factors, you'll have a better idea of how to speed up your Bitcoin transactions and get them confirmed faster!

For Type-3 (blob) transactions, just remember to set a max_fee_per_blob_gas. The good news? You don’t have to stress about a blob-specific tip!
Just a heads up, you'll need to cover the usual EVM base fees and the priority gas fees for that call that connects those blobs. So, if you're thinking about replacing a transaction in the mempool, just remember that you'll typically need to increase the fee by at least 10% or so. There's also that EIP-4844 which recommends aiming for a 1. Just a heads up, it's 1× for the blob base! Just a heads up: when you're making these replacements, keep an eye out for the chance that some client groups might start doubling their fees. It’s something to consider! For more info, check this out: eips.ethereum.org. You’ll find all the details you need there!

---

3) From theory to budgets: a simple forecasting model

Define:

So, a definition is basically just a way to say what a word or term really means. It's like giving you a little insight into how that word fits into the bigger picture. It really helps break down the meaning of a concept, so it’s way easier to wrap your head around it! You can find definitions in dictionaries or textbooks, but you can also come up with them based on how words are used in different situations.

Types of Definitions:

• What It Is: You’ll usually find these in dictionaries. They give you clear, specific descriptions, but they can feel a bit stiff and formal.
• Casual Explanation: These are basically how people talk about things in everyday life, and they can change a lot depending on someone’s own experiences or stories they’ve heard.
• Operational Definition: This is all about explaining something based on a specific process or a series of steps you go through. You’ll typically find this kind of definition in scientific or technical settings.

Why Definitions Matter:

• Clarity: They really make it easier to understand what a term means, clearing up any confusion.
• Communication: Having clear definitions helps make sure everyone understands what we're talking about. It keeps us all on the same wavelength, which is super important in any conversation!
• Learning: They play a crucial role in education, making it easier for students to understand new ideas.

If you want to dive deeper into definitions, feel free to check this out here.

• B = the amount of blobs in each batch (just a whole number).
• S represents the average size of each filled blob, which can be up to a maximum of 131,072 bytes.
• Pb = this is the estimated base fee for gas charged per blob, measured in wei.
• Gtx = the total number of execution gas used by the batch-posting transaction (measured in units).
• Peg is basically the estimated base fee for EVM, measured in wei per gas, while Tip refers to the estimated priority fee, also in wei per gas.
• ETHUSD = this is the exchange rate you’re anticipating, according to your assumptions.

Absolutely! Just share the text you want me to rework, and I'll get started on it.

So, when you're figuring out the cost of the blob component in each batch (in ETH), you can use this formula: it’s about B times 131,072 blob gas, multiplied by Pb, and then you divide that by 1e18. So, when it comes to the execution part of each batch for ETH, you can think of it like this: it’s Gtx times (Peg plus Tip), all divided by 1e18. Alright, here’s the scoop: to figure out the total cost for each batch, just throw all those numbers together and add them up. Then, if you want to know the cost per MiB, you’ll need to take that total and divide it by (B × S / 1,048,576). Easy peasy!

Planning Examples (use these in your spreadsheet):

Check out these awesome examples that’ll kickstart your planning!

Example 1: Weekly Meal Plan

Example 2: Project Timeline

Example 3: Budget Planner

Just pop these into your spreadsheet, and you'll have a great foundation to kick off your planning!

So, right now, we’re looking at the baseline for “quiet hours” sitting at Pb = 1 wei, which is the minimum you can get. Meanwhile, Peg is hovering around those low single-digit gwei levels. So, when we're talking about the blob component for each batch, it's pretty much around zero in USD. What really stands out here is execution, which takes the spotlight. On average, each blob transaction usually comes out to about $0. 50 and $3. 00. When you're calculating your blob count, try to make sure you're filling those blobs to maximize your per-transaction costs. It’s a smart way to keep everything balanced! Hey, if you’re interested, you can find all the details over at ethresear.ch. It's definitely worth a look!

Hey, just a quick note on the “spike risk” thing: it could be a good idea to set Pb in your model between 10 and 50 gwei when you're doing those stress tests. That should give you a solid range to work with! Give it a shot and run those simulations for about 1 to 3 hours at those higher price points. This will give you a hand in figuring out your backlog buffer and throttle policies. If you want more details, just check out Section 5! So, just to give you some background, on blobscription day, we actually saw gas fees shoot up to over 500 gwei! Crazy, right? And let’s not forget the Scroll airdrop--it was sitting around $4 at that time. Good times! Can you believe it? 52 per blob! It's wild out there! Check out this article from CoinDesk for all the details. (coindesk.com)

Seasonalities:

Seasonalities are basically those regular patterns or trends that pop up around the same time each year. You know, like holiday shopping spikes in December or summer vacation vibes in June.
You can spot these patterns in all sorts of places--like the weather, sales trends, or even how people behave socially. Getting a grip on seasonal trends can really help both businesses and individuals make smarter choices.

Key Points about Seasonalities:

• Nature of Patterns: You’ll often find that seasonal trends are closely linked to the time of year. For instance, you often notice that retail sales really take off during the holiday season.
• Impact on Business: Figuring out seasonal trends helps businesses get a leg up on planning their inventory and marketing strategies. When they know what to expect, they can make smarter decisions and snag more sales.
• Examples:
• Winter: When the chill sets in, people really start looking for cozy clothes and heating gadgets to stay warm.
• Summer: This is when we see a big boost in sales for air conditioners and swimwear.
• Data Analysis: If you want to dive into seasonal trends in what people are interested in, tools like Google Trends are super helpful. They really give you a good look at how consumer behavior shifts over time.

How to Identify Seasonalities:

1. Collect Data: Take some time to gather data from multiple years so you can spot any patterns that keep popping up. 2. Check Out Trends: Grab some graphs or charts to help you see patterns at a glance. It makes spotting trends way easier! 3. Statistical Tools: Don’t forget to make the most of tools like Excel or R. They’re super handy for diving deeper into your data analysis!

Conclusion

Getting a grip on seasonal trends can really turn the tide when it comes to making decisions. If you're a business owner or even just someone who likes to stay one step ahead, paying attention to these patterns can really make a difference!

Blob demand usually goes up during certain “event windows,” like when there are airdrops or upgrades happening. To stay on top of things, it’s a good idea to have an “events calendar” close by. Also, consider switching your batcher to a more cautious setup--like lowering the maximum number of blobs per transaction and shortening the channel duration--about 12 to 24 hours before any upcoming events. This way, you’ll be all set! (cointelegraph.com).

---

4) Designing blob schedules: cadence, fill, and inclusion

Four Knobs to Tweak (OP Stack in Focus)

Check out these four knobs you can tweak on the OP Stack (you’ll find similar ones on other stacks as well):

1. Batch Interval (Channel Duration): Okay, so here’s the deal with batch intervals: when we stretch them out a bit, it can really help us fill those blobs up and make better use of our resources. The downside, though, is that they might push back the “safe head.” Optimism recommends hanging on to around 5 hours of data, which works out to about 1,500 L1 blocks, to keep costs down. Just a heads-up: the safe head might take a while to get going, which isn’t ideal if you’re dealing with CEX or bridge service level objectives (SLOs). For most chains, it’s usually a good idea to target a steady state of about 10 to 60 minutes. You can definitely push it for a few hours, but that's really when things are pretty chill and your apps can handle a little delay without going haywire. If you want to dive deeper into the topic, go ahead and check it out here.
2. Multi-Blob Transactions. Alright, let's dive into multi-frame channels! Our goal here is to target around 5 to 6 blobs per transaction, especially on those medium to high-throughput chains. This way, we can help distribute the execution gas costs more evenly. How about we think about raising the minimum base fee and tip caps just a little--maybe by an extra 2 gwei? It could be a good move! Plus, we can look at extending the resubmission timeouts too. What do you think? This way, we can avoid those frustrating mempool replacement doubling problems. Feel free to take a look at the details right over here. You’ll find everything you need!

3) Compression Policy

For compression, I'd recommend using brotli at level 10 or 11. Those settings really do the trick! The default ratio that OP Stack usually goes with is roughly around zero. 6. Imagine compression as this super handy tool. When you take raw data and compress it--kind of like how you'd use a dictionary to make the language simpler--you can actually free up tons of space every day. It's pretty amazing when you think about it, especially when you're dealing with large amounts of data. Hey, if you want to dive into the details, just click here. You'll find everything you need!

4) Auto DA Route

Don’t forget to keep the “auto” option turned on! That way, you can easily switch to calldata whenever there's a spike in blobspace. Hey, just a quick heads up! EIP‑7623 has raised the calldata floor, which should really help in trimming down the size of those massive worst-case blocks. So, it looks like this route has turned into more of a backup plan instead of the main choice you’d rely on.

Make sure to keep a close watch on Peg and Pb. You’ll want to tweak your policy thresholds according to your cost service level objectives. If you want to dive deeper into the details, just head over to docs.optimism.io. It'll give you all the info you need!

Inclusion Realities to Plan For:

• Mix of Perspectives: When you're building a team or organizing an event, keep in mind that everyone has their own unique background and experiences. When we welcome different perspectives, it really spices up the conversation and often helps us come up with more creative solutions.
• Accessibility Matters: It's super important to create physical spaces and online platforms that everyone can easily access and feel comfortable in. This means thinking about stuff like wheelchair ramps, screen readers, and a bunch of other accessibility features.
• Inclusive Language: The words we pick really matter. They can have a big impact on how people feel and connect with us. Make sure to use language that’s friendly and inviting for all, steering clear of any jargon that could make some folks feel left out.
• Ongoing Training: This isn’t a one-time thing, you know? It’s all about keeping the learning going. Having regular training sessions on diversity and inclusion is a great way to ensure everyone’s on the same wavelength. It helps people recognize their biases and stay aware of the diverse perspectives around them.
• Feedback Loops: You know, it’s really crucial to build environments where everyone feels comfortable sharing their thoughts and opinions. Having regular check-ins can really help you see how well your inclusion efforts are working and figure out what tweaks might be needed along the way.
• Celebrate Differences: Organizing events or starting initiatives that showcase different cultures, identities, and life experiences is a great way to unite people and help them feel like they truly belong.
• Supportive Policies: It’s super important for your organization to have solid policies that back diversity and inclusion. Make sure everyone’s in the loop about them!
• Mentorship Programs: Connecting individuals from underrepresented backgrounds with experienced mentors can really make a difference. It offers valuable guidance and can lead to exciting new opportunities!
• Community Engagement: Dive into your local scene! Join up with organizations that are all about inclusion and making everyone feel welcome. Not only does this help you connect with others, but it also shows just how dedicated you are to the cause.
• Track Your Progress: Remember to keep an eye on how far you've come! Whether it's through surveys, data analysis, or other tools, checking in on your progress can really help you see where you’re at and pinpoint what needs a little extra attention.

All of these points can really help foster an atmosphere where everyone feels appreciated and part of the team. Just keep in mind that it’s all about creating an environment where everyone can really shine!

So, there's a bit of competition going on between builders and relays on PBS right now. Those big multi-blob transactions can really drag things out, especially as we approach the end of a slot. It's really important to remember that your 6-blob packages might not make it in on the first try, especially when there’s a ton happening. So, don’t just assume they’ll get included right away! It’s a good idea to space out your submissions early during the time slot and avoid those huge “all-or-nothing” packages just before the deadline. Trust me, it’ll save you a lot of stress! Based on what we've noticed, blob transactions don’t get included in blocks as often as those packed with calldata. This can really dampen the spirits of builders trying to stay motivated. Some teams are really thinking outside the box these days. They’re either going for private submission paths or teaming up with builders when things start to get hectic. It’s great to see that kind of creativity in action! (bitgetapp.com).

---

5) Throttling and backpressure: stop burning money during spikes

Modern batchers, like the OP Stack, can actually control the sequencer's speed, which is super helpful. It keeps those pesky DA backlogs under control and stops your costs from going through the roof!

Sure thing! Just go ahead and turn on sequencer throttling and adjust those backlog thresholds while you're at it. To get things rolling, you might think about starting around 3. You're looking at 2 MiB for the pending DA bytes, and the max intensity is roughly around 12. 8 MiB. Hey, just a heads-up! Make sure you’re capping the size for per-transaction and per-block DA while you’re throttling. It’s an important step! You've got a few different controller options to choose from, like step, linear, quadratic, and even a cool experimental PID mode. Hey, just keep in mind that when you have a lot of power, it also means you’ve got a big responsibility. (docs.optimism.io).

• If you're considering jumping into PID control, just take it easy. Remember, it's all about production gating and being pre-tuned, so tread carefully! Tweaking the tuning can lead to some wild oscillations that might really throw off your SLOs. Just a heads up! Just a quick note: the OP batcher's PIDStrategy is marked as EXPERIMENTAL. So, keep that in mind! (pkg.go.dev).

Runbook Snippet

Hey there! I’ve got a helpful little snippet for you that you can easily grab and drop into your on-call docs. Just copy and paste it--super simple!

# Incident Response Steps

1. **Acknowledge the alert**: Immediately let the team know you’re on it.
2. **Gather information**: Check logs and monitoring tools to pinpoint the issue.
3. **Assess impact**: Determine who is affected and how critical the problem is.
4. **Notify stakeholders**: Update the relevant parties about the situation.
5. **Implement a fix**: Apply a temporary solution if possible, or escalate if needed.
6. **Document the incident**: Keep a detailed record of what happened for future reference.
7. **Review and follow-up**: After resolving the issue, analyze what went wrong and how to improve.

For more info, check out our [incident management guide](https://example.com/incident-management).

Sure thing! I’ve been trained on data all the way up until October 2023, so I've got some pretty solid info to work with! If you need any changes or adjustments, just let me know!

• Trigger: If the price of Pb hits X gwei for Y blocks, or if the inclusion latency bumps up past Z slots.
• Action A: Let’s go ahead and cut the target blobs and transactions down by half. We’ll also shorten the channel duration to around 5-10 minutes. Oh, and don’t forget to enable throttling at a low threshold!
• Action B: Let’s switch the data availability (DA) route to “auto.” Also, we should put in place those calldata floor guardrails (after 7623) to keep any wild block payloads in check.
• Action C: If the backlog stretches past 30 minutes, let's hit the pause button on making new L2 batches for the low-priority stuff and concentrate on clearing it out instead. (docs.optimism.io).

---

6) Capacity planning under Pectra and December’s PeerDAS/BPO

Here are some anchor numbers you can easily plug into a spreadsheet (in binary units):

• Pectra (live): So, the idea is to handle 6 blobs for every block, which adds up to roughly 43,200 blobs daily. Yeah, that’s about 5 blobs every second! 3 GiB per day. So, if we hit the limit with 9 blobs, we're looking at cranking it up to about 7. 9 GiB daily. Check it out here.
• Post-PeerDAS BPO1 (Dec 2025): In this next phase, we're shooting for 10 blobs, which breaks down to roughly 72,000 blobs every day--that's about 8 blobs an hour. 8 GiB daily. So, if we reach the maximum of 15 blobs, that would bring us up to around 13. 2 GiB a day.
• Post-PeerDAS BPO2 (Dec 2025): At this point, we're aiming for 14 blobs. If all goes well, that should get us close to 100,800 blobs each day, adding up to about 12 in total. 3 GiB each day. So, if we're maxing out at 21 blobs, that basically drops down to around 19. 8 GiB a day. Get more details here.

So, we’ve got this payments rollup that can handle about 20 transactions per second. The average transaction size is around 200 bytes, and the Brotli compression ratio is sitting at 0.
6:.

Alright, so here's how we can figure out the effective bytes per second: just take 20, multiply it by 200, and then multiply that result by 0. Simple enough, right? So, if you break it down, that comes out to roughly 2,400 bytes per second.

• When it comes to daily data availability, we're seeing about zero. 2 GiB. So, when you're using Pectra, you're really only tapping into less than 4% of what you could potentially handle. The majority of the costs you’re seeing are actually coming from execution gas. You'll be able to fill up one blob roughly every 53 seconds, so it makes sense to plan your channels for about 1 to 5 minutes. Take a look at this link: (docs.optimism.io). It’s worth checking out!

So, we’ve got a social rollup here that’s handling around 120 transactions per second. The average transaction size is about 600 bytes, and interestingly, the ratio is sitting at 0. 5:.

Alright, let’s simplify this a bit. If we crunch the numbers, we’re looking at an effective rate of approximately 120 times 600 times 0. So, when you break it down, 5 equals about 36,000 bits per second, which is roughly 31. Pretty interesting, right? 25 KiB/s. So, if we break it down, that's about 2 every day. 6 GiB. With Pectra, you’re currently utilizing around 49% of the target capacity. Generally, the blob cost tends to stay pretty low, but every now and then, you’ll see those unexpected spikes that can really shake things up. If you're looking to optimize your setup, it’s a smart move to go with multiple blobs--like 5 or 6 of them. And when it comes to channels, aiming for around 5 to 10 minutes works really well. Trust me, you’ll notice the difference! Oh, and make sure to switch on that throttling with a quadratic controller too! Once BPO1 rolls out, you should notice that it drops to about 30% of the target. This change will definitely help cut costs and make everything run a lot more smoothly! For more info, be sure to swing by (blog.ethereum.org). It’s got all the juicy details you might want!

Checklist for December:

• Wrap up holiday shopping: Don’t forget to pick up those last-minute presents for your family and friends! Hey, just a quick reminder--make sure to mark off each name on your list!
• Wrap presents: Let your creativity shine when it comes to wrapping! Grab some festive paper, play around with colorful ribbons, and don’t forget to throw in a sweet handwritten note for that personal touch. It makes all the difference!
• Send out holiday cards: A simple note can really brighten someone’s day! If you're running low on time, think about sending digital cards! They can be a quick and convenient way to show you care.
• Spruce up your home: It's that time of year again! Grab your holiday decorations and add some festive cheer to your place. A little sparkle can make all the difference! Let’s get into the holiday spirit with some lights, ornaments, and maybe even a wreath! It's time to create that warm and cozy vibe!
• Plan your holiday meals: Take a moment to brainstorm what you want to serve at those festive get-togethers. Hey, make sure to remember everyone’s favorites!
• Plan family hangouts: Let’s pin down some dates for those fun family get-togethers we’ve all been missing! There's nothing quite like getting together and celebrating!
• Review your travel plans: If you’re planning a trip, make sure to double-check your reservations. It’s also a good idea to start packing your bags ahead of time!
• Look Back on the Year: Set aside a moment to really think about everything that’s happened over the past year. So, what have you achieved this past year, and what are you excited about as we head into the new one?
• Kick off your New Year’s resolutions: Take a moment to write down the goals you’d like to tackle in the upcoming year. It's such a great way to kick things off anew!
• Make time for yourself: With everything going on, it’s so important to set aside some time just for you. Whether it's a cozy night with a steaming cup of hot cocoa or just some peaceful time with a good book, you totally deserve to treat yourself!

Hey, just a quick heads-up! Make sure your CL/EL clients are all set for Fusaka because PeerDAS is really stirring things up with some awesome networking and sampling opportunities! Hey, just a heads up! You'll want to bump up your batcher thresholds since we're moving from 10 to 14 blob targets. But make sure to leave the spike runbooks alone; no need to tweak those. Hey there! Just a quick heads up: make sure to update your budget forecast to include those BPO dates. And don't forget to plan for a ramp-up in utilization during December and January, since that time tends to get pretty hectic. (blog.ethereum.org).

---

7) Operational realities most teams miss

Blobs usually hang around for just about 18 days. If you're looking to dive into some historical data for your fraud-proofing or audits, it might be a good idea to partner up with a third-party blob archive like Blocknative. Alternatively, you could always save your own data to platforms like S3 or Arweave. Just some options to keep in mind! Honestly, don't hold your breath waiting for L1 to have those blobs ready by day 19. (blocknative.com).

Just a heads up, Type-3 transactions can’t include creating new contracts since they need to be non-nil. Just a heads-up: keep that in mind when you're putting together your posting contract path! (eips.ethereum.org).

Hey builders and relays! Just a heads up--when you're dealing with high-blob payloads, timing can really be a tricky thing. So, keep that in mind! To keep things smooth and avoid any hiccups, it’s a good idea to mix up where you send your submissions. Try using different relays and builders instead of putting all your eggs in one basket! And hey, try to avoid those last-minute multi-blob sends, especially the late ones. They can be a real headache! (github.com).

Hey there! So, for your monitoring pack, make sure to keep tabs on a few key things. First off, check the blob_base_fee and blob_gas_used to see how they're stacking up against the target. Also, see how often you're getting included on the first go - that's always a nice win! Don't forget to keep an eye on your resubmission counts, too. And let’s not overlook that annoying backlog of “unsafe DA bytes” - it’s definitely worth keeping track of. Happy monitoring! "Make sure to set up alert thresholds based on how long your channel runs and the throttle limits. This way, you can keep everything under control and stay ahead of any issues." (docs.optimism.io).

---

8) Best emerging practices we see working

Alright, so here's the plan: let's use span batches (you know, the OP Stack “batch‑type=1”) and really ramp up the compression with brotli set to 10 or 11. It'll give us some solid results! Try combining this with multi-frame channels so you can aim for about 5-6 blobs per transaction when the volume is good. (docs.optimism.io). Make sure to keep that OP_BATCHER_MAX_CHANNEL_DURATION on the shorter side, around 10 to 60 minutes, especially during peak hours. This helps to keep latency in check and ensures everything runs smoothly! Feel free to ease up a bit during the quieter times if you need to fill those blobs. (docs.optimism.io). Alright, so let’s get the throttle-on-backlog set up with some reasonable defaults. And hey, don’t overlook adding in a manual “circuit breaker” for those really intense moments--like, make sure you can hit pause on the low-priority traffic when it gets too crazy! If you're not super familiar with control theory, going with a quadratic controller is generally a smarter choice than using a PID controller. (docs.optimism.io). So, here’s the deal: if you’re using the batcher, make sure to flip it to the “auto” DA mode. This way, it can automatically shuffle between blobs and calldata depending on the L1 pricing. Oh, and just a quick note--the minimum for calldata went up thanks to EIP‑7623. This change was made to help keep those block sizes in check, so be sure to consider that when you're figuring out your thresholds! (docs.optimism.io). Want to stay ahead of the game? Start by mapping out those “event calendars”--you know, like airdrops and L2 token claims. It’s super helpful to plan these out in advance. Plus, be sure to cut down your channel duration and blob target hours ahead of time. It’ll save you a lot of hassle later on! In the past, we've seen price spikes often come from things like inscriptions and airdrops. The cool thing is, these events tend to be pretty predictable, which means you can actually plan for them and protect yourself financially. (coindesk.com). Hey team! Just a quick reminder to plan ahead for our BPOs. Make sure to set your December change windows in the runbooks. Also, it’s a good time to re-baseline those dashboards to hit those 10/14 targets. And don’t forget to put your batcher through a solid test with under 10-14 blobs in staging. You got this! (blog.ethereum.org).

---

9) Putting it all together: a minimal, cost‑aware schedule

• Off-peak (when the blob base fee is at its lowest point; Peg is below 5 gwei): You can usually count on the channel taking around 10 to 20 minutes. Try to keep it to about 5 or 6 blobs for each transaction. So, multi-frame is all set up, tips are sitting pretty at 2 gwei, and we've got the quadratic throttle turned on. Plus, the auto DA is configured to "prefer blobs." Everything's lined up nicely! “Take a look at it here: (docs.optimism.io).”
• Peak (you're hitting close to your goal with utilization; Pb is on the upswing):
• When things get busy, the channel usually slows down to about 5 to 10 minutes. Try to aim for around 3 to 4 blobs each time you make a transaction. Also, make sure to set a resubmission timeout of at least 180 seconds. We’ve lowered the throttle thresholds, and we’ve also put some alerts in place to keep an eye on inclusion latency. More details here: (docs.optimism.io).
• Spike: When the gas price (Pb) goes over 10 gwei for more than 30 minutes and keeps trying to get included. When things get busy, you can usually count on a channel time of around 2 to 5 minutes. We're limiting it to just 1 or 2 blobs for each transaction. It's a smart move to turn on the "auto" (calldata fallback) feature and set some ceilings. Before jumping back to your usual routine, it’s worth pausing any low-priority traffic and clearing out any backlog. This way, you can ensure everything runs smoothly when you’re back in action! If you want to dive deeper into this topic, feel free to check out this link: docs.optimism.io. It’s got all the info you need!

Budget Guardrails You Can Sign Off On:

Managing your budget can be a real game changer, and having a few clear guidelines in place can really help you out. Here are some budget guidelines that you can feel good about approving:

1. Set Clear Spending Limits: Take a moment to figure out how much you're comfortable spending in each category, and make a plan to stick to it! This little step can really help you steer clear of overspending and keep your finances in check. It’s all about finding that balance!

2. Focus on What You Need First: Before splurging on the things you want, make sure you've covered your basics. Think bills, groceries, and saving a bit for the future. It's all about getting those essentials sorted out first! Every now and then, go ahead and treat yourself! It's important to indulge a little, but don't forget to keep your eyes on the things that truly matter in life.

3. Keep an Eye on Your Spending: Try using an app or even a basic spreadsheet to help you see where your money is going every month. It can really open your eyes to your spending habits! This is a fantastic way to catch trends and tweak things when necessary.

4. Take a Look Often: Make it a habit to review your budget at least once a month. This will keep you on track and let you adjust things if needed before next month sneaks up on you.

5. Plan for Unexpected Costs: It’s a good idea to stash away a bit more cash each month for those surprise bills that pop up out of nowhere. Trust me, you'll be glad you did! Believe me, they always show up when you least expect them!

6. Remember to Save: Try to set aside a little bit of your paycheck every month. It doesn't have to be a huge amount--just something to help you build up your savings over time! Having some savings tucked away, whether it's for a rainy day or that dream vacation you've been eyeing, can really help ease your mind.

7. Stay Flexible: Life can throw some surprises at us! If you ever need to tweak your budget, just roll with it and don’t sweat it. Make sure to take a step back and check in with yourself every now and then. It’s super important to stay focused on those long-term goals you’ve set!

If you stick to these guidelines, you'll have a great foundation to confidently handle your budget. Happy budgeting!.

• Base forecast: So, here’s the scoop on gas costs. Most of the expenses are gonna come from how well we execute things. We're looking at a range of about $0. 50 and $3. In regular situations, you're looking at about $00 for each batch transaction, and honestly, blob fees are pretty much negligible--almost zero, really.
• Stress budget: Let’s put together a little “spike reserve” that can hold us over for about 2 to 6 hours, aiming for a price between 10 and 50 gwei. So, what this really means is that we should probably plan to set aside a few hundred bucks to a couple thousand for every event. This is especially true for those medium rollups. And if we’re diving into high-volume L2s? Well, we might need to dig a little deeper into our pockets! (ethresear.ch).

---

10) What 7Block Labs can do for you

• Cost models: We're going to whip up a personalized workbook tailored specifically for you! It'll include your transaction mix, compression ratios, and all the on-chain telemetry you need. Oh, and don’t forget, we’re also adding in those December BPO scenarios, specifically 10/15 and 14/21. Check it out here.
• Batcher hardening: We take a close look at your multi-blob/channel setup and make some tweaks to optimize things. We'll adjust the throttling curves and whip up some runbooks to help keep everything running smoothly. We're going to set up alerts for both the blob_base_fee and inclusion latency. This way, we can make sure everything stays within those safe head SLOs. If you're looking for more info, you can check it out here.
• Archival and compliance: We're going to create a blob archive pipeline to make sure your audits and fraud proofs are safe and sound, even beyond day 18. Learn more here.

---

Appendix: Reference facts you can depend on

Great news! Dencun (EIP-4844) is officially set to launch on the mainnet on March 13, 2024. Mark your calendars! Hey, just a quick note for you! The blobs are set to be 128 KiB and they'll be sticking around for roughly 18 days. Just a heads up--type-3 transactions will have a max fee per blob gas. But remember, you can’t use them for creating contracts. If you want to dive deeper into the details, just click here!

Hey, just a heads up about Pectra! It’s set to launch on May 7, 2025, and they’re tweaking the blob target and maximum to 6/9. Exciting stuff ahead! Oh, and by the way, the calldata floor is about to get a nice upgrade with EIP-7623! If you're interested in digging deeper, you can check it out here. It’s got all the details you need!

Hey, don't forget to mark your calendar for December 2025--Fusaka is going to be something special! With this update, both PeerDAS and BPO are set to level up to 10/15, and then they'll soar even higher to 14/21. It's going to be an exciting time! Want to dive deeper? Check it out here!

---

Notes on data cited

When we’re checking out price trends and those unexpected shifts, we really lean on some key sources like EF/EIP. Plus, we also dig into independent analyses from folks over at Galaxy, The Block, and CoinDesk or Cointelegraph. They’ve got some great insights! These sources really help us improve our stress tests. Hey, quick tip for you: make sure to double-check the current blob_base_fee when you're scheduling, especially while you’re at it during runtime. Just a little reminder to keep everything running smoothly! (ethmrc.com).

---