What is Arbitrum ?

Arbitrum is a suite of Ethereum scaling solutions that enables high-throughput, low cost smart contracts while remaining trustlessly secure. Arbitrum uses optimistic rollup, which allows Ethereum smart contracts to scale by passing data between smart contracts on the Mainnet and those on the Arbitrum Layer 2.

Official Resources

Website | Twitter | Discord | Whitepaper

Author

Chinmay Farkya

Twitter

Discord - Chinmay#4134

Twitter Thread explaining this Dashboard

Overview

1. We will use Arbitrum’s transactions table to sum up number of transactions till date, total unique addresses/users that have transacted on the blockchain, and total unique contracts deployed on the blockchain.
2. For total accounts, we will count number of distinct “to addresses”.
3. For total contracts, we will count number of distinct “contract addresses”.

Transaction Statistics

1. We use the transactions table to find the number of transactions, and filter it for the required timeframe.
2. The 24 hour counter sees last 24 hours from the current timestamp, and similarly for the 7 days and 30 days Volume.
3. The daily, weekly and monthly trends of Transaction volume divide all transactions according to the day/week/month their block was mined.
4. These bar graphs show only data starting from June 15th, again because other data isn’t available. It will automatically update to include historical data whenever it is available.
5. Furthermore, these graphs show Volume of successful as well as failed transactions as per day/week/month.
6. Successful transactions are counted when status = “SUCCESS”.
7. Failed transactions are counted when status = “FAIL”.
8. We have used pie charts to show percentage of successful txns in the Last 24 hours, 7 Days and 30 Days, using the same method.
9. The transaction failure rate shows percentage of failed Txns by Week. Failure rate is calculated as (failed txns in a week)/(total txns in that week)*100.

Summary

1. The transaction data available in the Arbitrum Core tables on flipside crypto is only for the time since 15th of June, 2022.
2. The data recorded is proper only since the 23rd of June, 2022.
3. That is why the total transactions, contracts, and accounts data appears skewed.
4. The Arbitrum tables are in lite mode right now, which will be backfilled by the flipside team to include complete historical data eventually.
5. I have set these counters to refresh every 3 hours, so that they fetch complete data automatically whenever it is available.

Summary

1. For the Failure Rate of Transactions, timeline has been set to after June 19 to make the visualization depict correct broader values. The data before that is skewed anyways.
2. For Daily - The highest number of Daily transactions was on June 29 ( 243.25 k ). The highest number of Successful Txns was on June 29 ( 231.23 k ). The highest number of Failed Txns was on June 29 ( 12.01 k ).
3. For Weekly - The highest number of Weekly transactions was during Jun 27 - Jul 3 ( 1.109 M ). The highest number of Successful Txns was during Jun 27 - Jul 3 (1.072 M ). The highest number of Failed Txns was during Jul 18 - Jul 24 ( 42.179 k ).
4. These Daily and Weekly Trends are in sync with the Odyssey Event. Reference : Odyssey Event
5. The number of users then declined, maybe because the event was paused. Reference : Odyssey Event Paused
6. Monthly txns graph doesn’t have much data.
7. The highest Weekly Txn Failure rate was during Jul 11 - Jul 17 ( 6.227 % ).

User Statistics

1. Active Users refers to the unique addresses that transacted at least once during the given period.
2. We will use the transactions table to count the number of distinct “from addresses” that have transacted during the timeframes of last 24 hours/ 7 Days/ 30 Days.
3. The 24 Hour counter sees last 24 hours from the current timestamp and similarly for 7 days and 30 days.
4. Active Users refers to the unique addresses that transacted at least once during the given period.
5. The number of Active Users is segregated into daily/weekly trends according to the time when the block including their transaction was mined.
6. The data in bar graphs is incomplete before June 15. It will automatically update when other data is incorportaed.

Summary

1. The highest active users in a Day was on June 29 ( 57.181 k ), when the Arbitrum Odyssey event gathered interest of users.
2. The highest active users in a Week was during the week June 27 - Jul 3 ( 209.487 k ), due to the Odyssey event. Reference : Odyssey Event
3. The number of users then declined, maybe because the event was paused. Reference : Odyssey Event Paused

Summary

1. New Users data is hampered because these so called “new users” may have interacted before June 23, and those transactions are currently not recorded in the table.
2. Monthly data will not make sense here until complete data is backfilled.
3. Any highs and lows here are meaningless unless all data is backfilled, because these new users may not actually be new to the blockchain, many of them are only new because data prior to 15 June is not incorporated.

Most Active Addresses

1. We use the transactions table to figure out who are actively transacting on Arbitrum.
2. We find the “from addresses” with most number of transactions in the required timeframe.
3. This is done by aggregating all transactions of a particular address and then counting them.
4. The transactions are filtered out according to the timeframe using the block timestamp.
5. We have limited the scope of this query to Top 50.
6. 7 Days Timeframe includes transactions in the last 7 Days before the current timestamp, and similarly for 30 Days.
7. All time table includes all transactions since inception, but Data is available only after June 15th.

Most Active Contracts

1. We use the Arbitrum Event Logs table to figure out what are the most popular contracts on Arbitrum.
2. We find the “contract addresses” with most number of events in the required timeframe. This includes all types of events recorded in the events table.
3. This is done by aggregating all events of a particular contract and then counting them. This includes all types of events recorded in the events table.
4. The events are filtered out according to the timeframe using the block timestamp.
5. We have limited the scope of this query to Top 50.
6. 7 Days Timeframe includes events in the last 7 Days before the current timestamp, and similarly for 30 Days.
7. All time table includes all events since inception, but Data is available only after June 15th.

Gas Statistics - Comparison with Ethereum Mainnet

1. We will use the Arbitrum transactions table to find out Gas Price Paid for all transactions in a required timeframe, and then take average of all these values segregated into 24 hours/ 7 Days/ 30 Days.
2. 24 Hour counter includes last 24 hour from the current timestamp, and similarly for 7 Days and 30 Days.
3. We will use the Ethereum transactions table to find out Gas Price for all transactions in a required timeframe, and then take average of all these values segregated into 24 hours/ 7 Days/ 30 Days.
4. 24 Hour counter includes last 24 hour from the current timestamp, and similarly for 7 Days and 30 Days.
5. For Ethereum, we have left out any abrupt values while calculating the average.

TPS - Comparison with Ethereum Mainnet

1. We use Ethereum and Arbitrum transactions tables to find number of transactions per second by segregating transactions over Day.
2. We take total transaction on each day and divide it by 86400 to get TPS.
3. We have left out the current date to avoid anomaly in the line chart.
4. ALso, we have set the block timestamp for both networks to be after June 1, 2021 for better comparison. Arbitrum network was launched on this date.
5. Although the chart is supposed to show all data, but since these tables do not have data before June 15, 2022 , it is incomplete. It will be eventually backfilled and automatically updated.
6. The graph shows high TPS for Arbitrum during the Odyssey Event.

Conclusion

1. All the observations and analysis made in this dashboard is according to the standing on Aug 3, 2022.
2. All values will rapidly change and older data will be incorporated soon.
3. Till that time, this dashboard will serve as the overview of Arbitrum’s performance from June 15, 2022 to whenever you read this.
4. Right now, Arbitrum is doing well on all fronts : Activity, Onboarding New users, Low Gas Fees. Hope it will do better.
5. This dashboard shows that Arbitrum has been successful at gaining its share among Layer 2s.
6. We will eventually revisit this dashboard to see how Arbitrum is faring against Ethereum in terms of Gas Fees and TPS.

Here is a link to my Twitter thread explaining this dashboard :

Method

1. We will use transactions table to find out new users on the blockchain.
2. New Users refers to the Active addresses during a timeframe, that have never interacted on the blockchain before that timeframe. New users on Jul 1,for example, have never transacted before Jul 1, in the recorded history.
3. First we take the Union of Distinct “from addresses” and “to addresses”.
4. Then aggregate the transactions of a unique address.
5. Now find out the first time that these addresses ( individually ) transacted using the minimum of block timestamp (day/week/month when the block consisting of this first transaction was mined) for all these addresses.
6. Now segregate these addresses according to their joining times and required timeframes ie. 24 hours/ 7 days/ 30 Days.
7. The bar charts are based on the same method. Only segregation is based on Daily/Weekly.
8. The data is incomplete before June 15. It will automatically update when relevant data will be incorporated.

Summary

1. In last 7 Days Activity, ( 0xef…faf ) was the most active address with 20,622 transactions, accounting for 11.9 % of the sum of transactions of the top 50 addresses. Its count was more than 3 times of the count of the second most active address ( 0x1e…fa3 ).
2. In last 30 Days Activity, ( 0xef…faf ) was the most active address with 83,767 transactions, accounting for 12 % of the sum of transactions of the top 50 addresses. Its count was more than 2.5 times of the count of the second most active address ( 0x14…7b9 ).
3. In all time Activity, ( 0x71…564 ) was the most active address with 222,946 transactions, accounting for 15.6 % of the sum of transactions of the top 50 addresses. Its count was more than 1.15 times of the count of the second most active address ( 0xf8…b8b ).

Note : This data is calculated according to the standing on Aug 3, 2022. The data will rapidly change and these observations may be proven inaccurate.

Summary

1. In last 7 Days Activity, ( 0x82…ab1 ) was the most active contract with 484,773 transactions, accounting for 18.9 % of the sum of transactions of the top 50 contracts. Its count was more than 1.22 times of the count of the second most active contract ( 0xff…cc8 ).
2. In last 30 Days Activity, ( 0x82…ab1 ) was the most active address with 1,800,725 transactions, accounting for 18 % of the sum of transactions of the top 50 contracts. Its count was more than 1.21 times of the count of the second most active contract ( 0xff…cc8 ).
3. In all time Activity, ( 0x82…ab1 ) was the most active contract with 3,414,852 transactions, accounting for 18.3 % of the sum of transactions of the top 50 contracts. Its count was more than 1.26 times of the count of the second most active address ( 0xda…7eb ).
4. This means that ( 0x82…ab1 ) has consistently remained popular.

Note : This data is calculated according to the standing on Aug 3, 2022. The data will rapidly change and these observations may be proven inaccurate.

Summary

1. Gas fees is the actual ETH paid for a transaction. We use the transactions table of Arbitrum and Ethereum to fetch Txn Fee and then average these values over Day.
2. We have limited the results to only successful transactions and to timestamps after June 1, 2021, when the Arbitrum network was launched.
3. We have left out abrupt values of gas bids to normalize the chart.
4. Though right now data is incomplete, we only see data since June 15 in the line graph. It will be incorporated automatically when data is backfilled by flipside team.
5. The left Y-axis shows Gas fees for Ethereum on that Day. The right Y-axis shows Gas fees for Arbitrum on that Day. Both axes have Unit $ETH.
6. The highest Gas fees of Ethereum was on June 29 ( 0.0049 ETH ). The highest gas fees of Arbitrum was on June 29 as well ( 0.0017 ETH ).
7. We know that Gas Fees increases when there is congestion in the network. Because of the Odyssey Event, there waas unusually high activity on the Arbitrum network, and thus the high gas fees. Reference : Odyssey leads to high Gas Fees

Summary

1. Gas fees is the actual ETH paid for a transaction. We use the transactions table of Arbitrum and Ethereum to fetch Txn Fee and then average these values over Week.
2. We have limited the results to only successful transactions and to timestamps after June 1, 2021, when the Arbitrum network was launched.
3. Y-axis shows the Average $ETH value of gas fees during that week, for both Ethereum and Arbitrum.
4. This line chart is supposed to bring in place perspective of the gas fees difference between the two networks, as against the previous line chart.
5. We have left out abrupt values of gas bids to normalize the chart.
6. Though right now data is incomplete, we only see correct data since June 15 in the line graph. Correct data will be incorporated automatically when data is backfilled by flipside team.
7. We know that Gas Fees increases when there is congestion in the network. Because of the Odyssey Event, there was unusually high activity on the Arbitrum network, and thus the high gas fees. Reference : Odyssey leads to high Gas Fees