Author: Zack Pokorny, Galaxy; Compiled: Whitewater, Abstract: Within 150 days after the implementation of EIP-4844 (as of August 10, 2024), 2225958 blob have been purchased with an average cost of $1.59 per blob, and 1104315 blob carry layer 1 transactions with an average cost of $5.22 per transaction. Overall, Ethereum generated 2692.39 ETH and 9318794 USD in revenue from the blob. 89.45% of the total fee income of 2408.41 ETH or blob has been destroyed; The rest will be provided to the verifier in the form of priority tips. Rollups purchased approximately 285 GB of blob data in total, using only about 76% of its capacity. The size of Blob is fixed at 128KB, and a maximum of 6 Blob can be processed in an Ethereum block. The data contained in each blob is transient and will be automatically deleted from most Ethereum nodes after about two weeks. Rollup spent a total of $3549430 on the blob, with a cost of $16473 per GB used and $12458 per GB purchased. These data cover the period from March 13, 2024 to August 10, 2024. Rollup costs have significantly decreased, with Arbitrarum, OP Mainnet, Base, zkSync, Linea, and Scroll paying an average of $5.564 million in operating costs per day after Dencun. By contrast, in the 150 days prior to Dencun, the daily cost of using calldata (another way to store arbitrary data on Ethereum) was $1.07 million. The profit margin of Rollup has relatively improved, with an optimistic Rollup profit margin (using Base, OP Mainnet, and Arbitrarum as agents) increasing from 22.65% in the first 150 days of Dencun to 92.3% in the first 150 days of Dencun; The profit margin of zero knowledge Rollup (using zkSync, Scroll, and Linea as agents) has increased from 27.27% before Dencun to 66.7% after Dencun. Although the daily value of expense income has decreased by 42% compared to before Dencun, the absolute Rollup profit margin has improved after Dencun. Rollup earns more bottom line income than Dencun did before. After Dencun, the activity on the leading L2 immediately rebounded. However, the increase in transaction volume is accompanied by an increase in failure rate. Most failed transactions come from highly active addresses, possibly bots. The low cost on L2 may drive an increase in robot activity. The transaction failure rate of ordinary users who have not sent a large number of transactions is only slightly higher than the level before Dencun. After Dencun, Ethereum's revenue significantly decreased and the supply of ETH was destroyed. The total revenue is 69% lower than the average 150 day rolling total before the upgrade; The amount of ETH destroyed is 84% lower than the average 150 day rolling total before the upgrade. The key definition of Blob - a binary large object or "blob" is a temporary data storage space for aggregating data on Ethereum consensus clients. Blob was introduced as part of EIP-4844. Storage can be used for any type of data, just like the calldata space that was dependent on before EIP-4844. However, Ethereum developers plan to use Blob to store transaction data for aggregation. The target rate for each Ethereum block is three Blob, with a maximum quantity of six. Blob sideling "refers to the object wrapper of a blob that carries transaction data in a block. Blobdata "refers to the data stored in Blob. Blob Carrying Transaction - Blob carrying transaction or "Type 3" transaction is an EIP-4844 transaction that includes a reference to the blob but does not include the blob itself. Blob propagates through the consensus client Blob sideling and is not provided to the executing client. Transactions carrying Blob are processed on the execution layer (Layer 1) of Ethereum, which includes fees related to Blob in addition to the typical base and priority fees of Ethereum transactions. Blob Capacity - The maximum amount of blob data that can be stored in a blob. Each blob can hold up to 128 KB of data. Blob buyers must pay for the full capacity of the blob, regardless of whether they fill it or not. Blob capacity usage - Rollup actually fills each blob with a share of 128 KB capacity filled with batch transaction data. Calldata - a dedicated data storage space attached to each Ethereum transaction. This dedicated space can be used to store any type of data, and prior to EIP-4844, Rollup typically used it as a storage space for transaction data. After EIP-4844, some Rollups are still using Calldata. Batch processing - a set of Layer 2 transactions that are "aggregated" together and submitted to Ethereum as a single transaction. Bundling transactions in batches is a method for L2 aggregation to reduce Ethereum user feesBatch processing is currently mainly stored in blob, previously stored in transaction call data. Ethereum Execution Layer - also known as Ethereum Layer 1 (L1), the execution layer is a part of the Ethereum network used to process transactions and execute smart contracts. It also includes EVM, the execution engine of Ethereum, which is used to execute rules and pricing for all on chain operations. Ethereum Consensus Layer - A part of Ethereum used to implement Proof of Stake (PoS) consensus algorithm, enabling the network to reach consensus based on verified data from executing clients. The consensus layer is where blob data is stored. The background of EIP-4844: Ethereum's Dencun upgrade was successfully activated on March 13, 2024 at 1:55 PM UTC (beacon slot 8626178 and epoch 269568). The nine Ethereum Improvement Proposals (EIPs) for upgrading include EIP-4844. This EIP provides a solution to reduce the operational costs of Ethereum aggregation by using blob and type-3 transactions. In terms of aggregation, the data storage cost of Blob and Type-3 transactions is lower than that of Calldata and Type-2 transactions. Unlike calldata stored in the Ethereum execution layer, Blobdata is temporarily stored in the consensus layer and cannot be queried for smart contracts through the Ethereum Virtual Machine (EVM). This reduces the dependency of aggregation on calldata and Ethereum execution layer, thereby reducing the computational burden of storing aggregated batch data and publishing it to Ethereum, thereby lowering costs. Then, Blobdata is published to Ethereum through type 3 transactions that include references to the blob stored on the consensus layer, but do not include the blob itself or the data it contains. The total cost of storing summary data in the call data of transactions and publishing it together with Type 2 transactions is calculated as follows: (Basic cost * Gas used for calling data)+(Priority cost * Gas used for calling data). In contrast, the total cost of Type-3 transactions on Ethereum that include Blobdata is calculated as follows: (Blob Basic cost * Blob Gas usage)+(Basic cost * Calldata Gas usage)+(Priority cost * Calldata Gas usage). Although the blob basic cost is included in the total cost calculation, Type 3 transactions are usually cheaper than Type 2 transactions because there is less call data gas used in Type 3 transactions, and the costs related to blob gas are usually higher than those related to calling. The cost related to data gas is cheap. Blob has its own fee market, independent of any Ethereum L1 transaction type fee market. The Blob fee market has its own parameters to set the basic fee, which are only affected by the demand for the blob in the previous block. The activities that occur on Ethereum L1, such as user exchanges on DEX, will not affect the cost of the blob. The activity on Ethereum L1 does indeed affect the basic fees for transactions carrying blob, but it will not affect the basic fees for blob. The basic fee for blob in an independent market is set at 1 wei, and as the demand for blob exceeds the target quantity of each block and reaches its maximum value, the fee will also increase accordingly. In gas units, a blob requires 131072 gas. Calculated at an ETH price of $3000, this means that the basic cost of the blob may be as low as 0.000 billion 00131072 ETH (12 zeros) or 0.000 million 039322 USD (9 zeros). Therefore, the first part of calculating transaction fees for Type 3 mentioned above is sometimes actually free. This report investigates the usage status of Ethereum blob and the impact of EIP-4844 on the aggregation economy from the perspectives of consensus and execution client storage and publishing. It does not include any analysis of the other eight EIPs included in the Dencun upgrade. Starting from the perspective of the Ethereum consensus layer, this section introduces the overview of Blob on the Ethereum consensus layer. It includes information about the number of purchased Blob, the total rollup expenditure on the consensus layer Blob, and the purchased and used blob capacity. The following table provides a high-level overview of the top 10 Ethereum rollups by the number of purchased blob. In the 150 days since the activation of EIP-4844, they have purchased a total of 1939657 blob (with a purchased blob capacity of 248.2 GB, accounting for 87% of the total blob demand) and spent $2.1 million (603.6 ETH) on the blob. Paradex, StarkNet, and Arbitrarum used the largest share of purchased blob capacity, with an average utilization rate of 95.23%. Base spent the most on blob during this period, spending $811000 or 232.5 ETH on 619516 blob. The cost of a single blob is calculated in ETH using the following formula, where the blob gas usage is 131072 gas: (blob basic fee * blob gas usage)/1e18. The cost of the blob can also be calculated in US dollars by multiplying the result of this formula by the ETH price when the blob is included in the Ethereum blockThis is the ETH price when executing blob carrying transactions on the chain. As mentioned in the previous section of this report, the total cost of using blob is the sum of all three components calculated for Type 3 transaction fees mentioned above. The cost of the blob itself is only the first step in calculating the total transaction cost. The other two values related to transaction costs carried by blob will be introduced in more detail in the next section of this report. Since May 28, 2024, Blob has been receiving continuous demand from Rollup, with between 15000 and 20000 Blob per day. This date is important because it marks the launch based on Rollup Taiko. The unique feature of rollup is that transactions on them will only be confirmed after being confirmed on Ethereum L1. This requires them to constantly purchase blob to keep the blockchain running. They are different from optimism and zero knowledge aggregation, which provide soft confirmations of transactions on aggregation and then confirm these transactions on L1 at some point in the future. This allows the sorter to have greater flexibility in terms of the number of purchased blob and L1 transaction frequency. Although Taiko is a relatively new aggregation, it is the third largest spender on blob in the consensus layer and the largest spender on blob carrying transactions in the execution layer. Base and Taiko are the two platforms that purchase the most blob per day, with an average of 8667 blob purchases per day since June 1, 2024 (accounting for 47.4% of all blob demand). During this period, Base purchased an average of 5094.5 blob per day, while Taiko purchased an average of 3572.5 blob per day. Despite being released 75 days after the blob was launched, Taiko remains one of the top three aggregation platforms for blob purchases since Dencun. From a data demand perspective, since June 1, 2024, Rollup has been continuously purchasing 2 to 2.5 GB of blob per day. However, they only used 1.5 to 2 GB of data to fill the blob they purchased. This highlights that Rollup only used 71.91% of the capacity they paid for during this period. June 20, 2024 and June 21, 2024 are the two days with the highest usage of Blobdata, at 2.66 and 2.75 GB, respectively. June 20, 2024 was the airdrop of Layer Zero on Arbitrum, which caused a surge in activity on Arbitrum. The demand for blob capacity is highest from March 27, 2024 to April 3, 2024, at approximately 2.7 GB per day. During this period, Blob subscriptions were very popular, which created an artificial demand for Blob. Blobscripts allow users to embed any data into Blob, regardless of the intended use of carrying aggregated transaction data. Within 150 days after Dencun, an average of 21.1% of blob capacity was unused per day. The following figure highlights the composition of daily blob capacity between blob capacity and unused capacity. Nine days worth noting, rollup blob expenses exceeded $100000 per day. The notable features of these days are the Blobscript frenzy from late March to early April and the Layer Zero airdrop day on Arbitrum on June 20th. These nine days spent a total of $3542579 on Rollup, averaging $394000 per day (815714% higher than the average spent on all other days). In addition, these nine days accounted for 99.8% of Dencun's total blob expenditure of $3549430 since its launch. Excluding these outliers, the average daily expenditure of Rollup on the blob is only $48.25. It should also be noted that these days there are an average of 2.9 blob per Ethereum block, which is 52% lower than the maximum blob count per block (6), 4% lower than the target count (3), and 40% higher than the daily average blob count per block since the release of EIP-4844 (2.06). This highlights that when the demand for blob continues to reach or exceed the target rate, the cost follows a parabolic curve. Another way to measure blob cost per GB is based on the amount of space purchased and ultimately filled by aggregation. By using the summarized daily Blob expenditure and usage data, we can understand the cost of purchasing and filling 1GB of Blob space. The following figure compares the daily Blob expenditure with the required daily Blob capacity and the Blob capacity used to measure the cost per GB of Blob. At the peak of the Blobscript craze, the cost of blob reached $503000 per GB of blob capacity used and $207000 per GB of purchased blob capacity. On the day of the Layer Zero airdrop on Arbitrarum, the cost of the blob reached $410000 per GB of capacity used and $330000 per GB of purchased capacity. Please note that the cost per unit of capacity used is greater than the cost per unit of capacity purchasedBecause the summary did not utilize the full capacity of the blob in these days. During the Blobscript craze, the average blob capacity used was only 50%. Sometimes, up to 63% of purchased blob capacity is not aggregated for use. On the Layer Zero airdrop day of June 20th, 19.5% of blob capacity was not aggregated and utilized. In the 150 days since Dencun went live, Rollup has spent a total of $3.55 million and 1020 ETH on consensus layer blob spending. ETH spent on blob will be removed from circulation, just as the basic fees for transactions executed at the Ethereum execution layer will be destroyed. The following values represent the amount of ETH removed from the circulating supply due to the Blobdata activity of the consensus layer. It does not reflect the amount of ETH destroyed through blob carrying transactions at the execution layer - more information about this will be introduced in the next section of the report. Overall, since the launch of EIP 4844, Rollups has spent $12458 per GB of blob space purchased and $16473 per GB of blob space used. In the next section of the report, we will analyze the blob pattern from the perspective of the Ethereum execution layer. This includes components such as the number of transactions carried by the executed blob, the cost of executing these transactions, and the average number of blob transactions carried by each blob. Please note that these values do not include the cost of Blobdata in blob carrying transactions, but rather focus on the values of the last two components of the total blob fee calculation, namely: (basic fee * Gas used for call data)+(priority fee * Gas used for call data). This report will later share an analysis of the cost difference between the third type of transactions, including Blobdata fees, and the second type of transactions that write summary transactions into call data. The table below summarizes the top 10 Ethereum tokens used for blob transfer transactions on the Ethereum execution layer. Taiko spent the most on blob transfer transactions, with a significant increase in proportion, investing 631.67 ETH, resulting in 423.58 ETH being burned out of the basic fee (totaling $2.2 million). ZkSync spends the most on each blob transfer transaction, with 0.0045672 ETH per transaction ($15.78 per transaction). Starting from May 28, 2024, Rollups will continuously execute 8000 to 10000 blob carry transactions per day, with an average of 2.02 blob transactions per Ethereum transaction (total number of purchased blob/total number of transactions carrying blob). Taiko carries the most blob transactions to Ethereum every day, with an average of 3558 transactions per day. Since the launch of Blob, the average daily transaction expenditure for all Rollup Blob carry transactions has been $341000, and since Taiko's launch on May 27th, the average daily expenditure has been $418000. It is worth noting that during this period, the gas prices of all Ethereum transactions have been on a downward trend. As of August 10, 2024, the average daily gas price used is 3.5 gwei, which is 67.2% lower than the daily average of 356 days and 94.5% lower than the average gas price on March 13, 2024. Since Dencun went live, Rollup has spent a total of $5.77 million on blob carrying transactions. The execution layer cost of Blob Carrying transaction expenses, divided by expense type, is mostly basic expenses, accounting for 83% of the total payment expenses. The basic fees on Ethereum will be destroyed and permanently removed from the circulating supply, while priority fees will be paid as a "tip" to validators. Since Dencun's launch, validators have received a priority fee of $974800 from Rollup. Taiko paid the highest priority fee, which was 719700 US dollars. This accounts for 74% of the total priority fees paid by Rollup. Of Taiko's total expenditure on executing layer blob costs, 33% is used to pay priority fees. It pays a higher proportion of the total cost of the execution layer blob in the form of priority fees than all other rollups. StarkNet has the least expenditure on priority fees. Of StarkNet's total expenditure on executing layer blob costs, 0.49% is used to pay priority fees. By measuring the base and priority fees paid in ETH units, we can accurately determine how much ETH is consumed through blob based transactions. A total of 1389 ETH was consumed through executing blob hosted execution layer transactions. The priority fee of 283.97 ETH has been paid as a tip to the verifier. The next section of this report will introduce more information about ETH destruction and the income obtained by validators. The impact on Ethereum validators and ETH supply has been reduced due to the introduction of blob, resulting in a decrease in total revenue generated by Ethereum and ETH destroyed during Rollup activities. The following analysis focuses on 1) changes in destroyed ETH, 2) income directly paid to validators, and 3) total income obtained by Ethereum before and after the implementation of EIP-4844For this purpose, we compare the 150 day rolling sum of each data point before EIP-4844 with the direct sum of the same metrics for 150 days after EIP-4844. The rolling total of 150 days starts from January 1, 2022, so it captures values before Dencun, which can be traced back to August 5, 2021 at the earliest. This analysis captured data from 24 different Rollups. As mentioned earlier, the blob and basic fees for Type 3 transactions will be destroyed, and priority fees will be paid as tip income to the verifier. The values before Dencun do not include the revenue contribution and ETH destruction of zero knowledge (ZK) proofs, which are a type of data stored on Ethereum by ZK rollups. Our analysis only includes the protocol revenue and destruction of bulk submissions for optimistic rollups and ZK rollups transactions. Please note that Dencun's previous values will cease on March 13, 2024 (the day EIP-4844 goes live). The following figure compares the total amount of ETH destroyed within 150 days after introducing the blob with the historical rolling total of ETH destroyed by rolling up call data before Dencun. Although some rollups after Dencun still rely on calldata, the usage of rollups is much less extensive. As of August 10, 2024, the estimated Ethereum protocol fee income generated by Rollup using calldata in the past 30 days is 39.69 ETH, compared to 13163 ETH in the 30 days prior to Dencun. After Dencun, the amount of ETH destroyed from the base fees and blob fees of Type 3 transactions is lower than any 150 day rolling total of data published using calldata from January 1, 2022. Starting from March 13, 2024, 2408 ETH has been destroyed, while Dencun's minimum rolling total for the previous 150 days was 3286 ETH. Before EIP-4844, the average rolling total of ETH destroyed under batch release of calldata was 15052 ETH over 150 days. This indicates that after the activation of EIP-4844, the amount of ETH burned from bulk submissions of aggregated transactions on Ethereum has significantly decreased. The following figure highlights the value of direct payment to validators before and after EIP-4844. A total of $974876 has been paid to the verifier through the Type 3 transaction priority fee. This is the level seen in May 2022 and June 2023 using the total of 150 day rolling priority fees from batch submissions from calldata before using blob. When using Calldata, batch submissions generated an average of $1.196 million in priority fees over a rolling period of 150 days. The validator's daily income from the priority fee of type 3 transactions containing Blobdata is lower than that of type 2 transactions containing Calldata. Finally, we can compare the total revenue obtained from the blob with the total revenue obtained from batch submissions of calldata. The calculation method for the total revenue obtained from blob is: (Blob basic fee * Gas used by Blob)+(basic fee * Gas used by Calldata)+(priority fee * Gas used by Calldata). The calculation method for the total revenue obtained from batch submissions of calldata is: (basic fee * Gas used by Calldata)+(priority fee * Gas used by Calldata). The total revenue obtained by Ethereum from the basic and priority fees of Blobdata and Type 3 transactions is 9318794 US dollars, while the average total revenue obtained from Type 2 transactions and the use of calldata is 29.92 million US dollars over 150 days. The table below breaks down the top 25 summaries used in the above analysis by Rollup type and their respective contributions to Ethereum protocol fee revenue after Dencun. In total, we evaluated 18 optimistic summaries, 6 zero knowledge summaries, and 1 based summary. It is worth noting that the only one in our analysis based on aggregated Taiko contributed 74% of all tips paid to validators and over a quarter of all ETH destroyed under EIP-4844. These 25 aggregates account for 81% of the destroyed ETH, 98% of the tips paid to validators, and 83% of the total revenue generated from all blob and type 3 transactions executed on Ethereum. The goal of EIP-4844 is to reduce the operating costs of Ethereum Rollup, making it easier to operate and use. The following section evaluates the impact of the upgrade on Rollup's economy and activities. It uses Arbitrarum, Base, OP Mainnet, Linea, Scroll, and zkSync for analysis. These are the optimistic Rollup and ZK Rollup ranked in the top three based on the total value of network lock-in. Advanced Economics: Firstly, after the launch of EIP-4844The operating costs of Rollup have significantly decreased, except for Blobscript frenzy, Layer Zero airdrops, and the market volatility day on August 5th. The following figure shows the daily total cost generated by Rollup, including Blobdata, publishing Blobdata through Type-3 transactions, batch submission of calldata, and zero knowledge proofs (in the case of Scroll, Linea, and zkSync). Since the activation of the blob, including the number of abnormal days and the time it took for rollups to implement Blobdata, these rollups have paid an average of $5.564 million in operating costs per day. In contrast, before the implementation of EIP-4844, the daily average cost during the 30 day period was $1.9 million, the daily average cost during the 90 day period was $1.27 million, and the daily average cost during the 150 day period was $1.07 million. The reduction in operating costs is accompanied by a decrease in revenue resulting from these consolidations. The day Layer Zero was airdropped on Arbitrum and August 5, 2024 were obvious exceptions. Arbitrum earned most of its income during these abnormal days. In the past two days, the total revenue of the six aggregates was 6.09 million US dollars, of which Arbitrarum received 4.63 million US dollars (76%). Since the activation of the blob, including abnormal days and the time spent on implementing Blobdata, these summaries have generated an average of $691300 in revenue per day. By comparison, before the implementation of EIP-4844, the daily average revenue during the 30 day period was $2.33 million, the daily average revenue during the 90 day period was $1.46 million, and the daily average revenue during the 150 day period was $1.2 million. Although the income has decreased, the absolute value of these aggregated profit margins has increased. The following figure shows the observed summarized bottom line profit margin, calculated as revenue minus all operating costs. Base is the only optimistic summary in our analysis that has experienced a net loss since the launch of EIP-4844. On June 20, 2024, the consolidated loss was $175000. The loss has been fully recovered and some have been recovered in the following three days. Scroll is the only ZK summary with a negative profit margin. On March 26, 2024, Rollup incurred a loss of $25000, but on the following day (March 27, 2024), it generated a net profit of $150000. On May 28, 2024, Rollup incurred a loss of $6700, a loss of $8000 on June 5, 2024, and a loss of $9000 on June 6, 2024. Since activating the blob, including the number of abnormal days and the time it takes for rollups to implement Blobdata, these rollups can earn an average of $553000 per day. By comparison, before the implementation of EIP-4844, the daily average revenue within 30 days was $685000, within 90 days was $389000, and within 150 days was $324000. The Rollup profit margin expressed as a percentage indicates a relative improvement in the share of its revenue used for operating costs. After EIP-4844, Rollup earned more income than before. The following figure separates the profit margins of optimistic Rollup and ZK Rollup to highlight how the use of Blob affects the profit margins of each type of Rollup. Since March 13, 2024, the average daily percentage profit margin of optimistic Rollup has been 92.3%; Since the same time, ZK Rollup's average daily percentage profit margin has been 66.7%. The additional cost of proof paid by ZK Rollup has relatively reduced its profit margin. The Rollup activities before and after EIP-4844 showed an immediate increase in user activity on the six aggregates observed after EIP-4844 activation, and maintained high usage for the following 150 days. The following figure highlights the direct impact of EIP-4844 on these aggregated trading activities. During the period from December 1, 2023 to March 12, 2024, an average of 3.285 million transactions were conducted per day, compared to 6.656 million transactions conducted within 150 days after EIP-4844. This indicates that trading activity has more than doubled. The same trend can be observed in all Ethereum aggregates. The cost reduction of using these networks is the main reason for the increase in rollup usage after Dencun. The following figure highlights the significant reduction in transaction costs for each network. Arbitrarum achieved the maximum reduction in transaction costs by using the daily median transaction fee. From December 1, 2023 to EIP-4844 activation, the median cost per transaction for this network is $0.37 on average; Within 150 days after EIP-4844, this rate decreased by 94% to $0.02. The median transaction fee for Scroll has the lowest decrease, at 58%, dropping from $0.74 to $0.31. The observed increase in summary activities has led to an overall increase in transaction failure rates. It is worth noting that the proportion of transaction failures on Arbitrarum, Base, and OP Mainnet networks has significantly increased. Using a seven day moving average of daily failure ratesWithin 150 days after Dencun, the failure rate of Base was as high as 21%, Arbitrarum was 15.4%, and OP Mainnet was 10.4%. The failure rate is mainly driven by high activity addresses (possibly robots). The Galaxy Research Dune query highlights the failure rate of addresses attempting 100 or more transactions per day. Since Dencun activation, using a seven day moving average, the failure rate of these addresses has reached as high as 41.6% on Base, 20.87% on Arbitrarum, and 12.85% on OP Mainnet. In contrast, the failure rate of low activity addresses attempting five or fewer trades per day using the same seven day moving average during the same period was the highest among all observed networks at 4.02%. The transaction failure rate of OP Mainnet in low activity addresses is lower than March 13, 2024, while the transaction failure rate of Base in low activity addresses has only been slightly higher since the same time. Strangely, after Dencun, the transaction failure rate of low activity addresses on Arbitrarum increased by 545%. The decrease in transaction costs for Rollup and the concentration of transaction failures in high activity addresses rather than low activity addresses suggest that robot activity may be the root cause of the increase in these Rollup failure rates after Dencun. Conclusion: The introduction of EIP-4844 and Blobdata has greatly improved the operation and cost-effectiveness of Ethereum rollup. However, this development has shifted some of the revenue sources from Ethereum L1 and ETH consumption rates to rollup operators (also known as sequencers). Under the previous data availability (DA) calldata model, Ethereum captured 77% of the value generated by rollup. With the implementation of EIP-4844, Ethereum now captures 8% of the value generated by optimistic rollups and 33% of the value generated by ZK rollups, totaling (1-Rollup percentage profit margin). This transformation aligns with Ethereum's mission to become an efficient DA layer, as rollups are the primary venue for paying users and trading activities. Although the cost of Ethereum DA has improved, overall it is still expensive, with a rollup cost of $16473 per 1 GB of Blobdata used and a cost of $12458 per 1 GB of Blobdata purchased. The high cost was caused by several days of skyrocketing blob costs. When the demand for blob is low, the cost can be ignored. Excluding blob costs, Type 3 transactions only cost an average of $5.22 per transaction. Some people are concerned that if the demand for blob increases and remains high, the cost may rise. One possible solution could be to increase the maximum and target blob count for each Ethereum block