 Cool. This is Verdean. I'm the engineer manager at Sentinel team, so our team provides software services and data set for monitoring and analysis of the Filecoin blockchain. So we have been working on an effort to store all the historical blockchain data in BigQuery using our software and the full archival snapshot since Genesis. So in this demo, I will just quickly demonstrate how can we query the data. So the data set in BigQuery is a public data set. So anyone with the URL and a valid Google Cloud account should be able to open that query. But you will be you will need to pay for the compute, since we don't want to be responsible for anyone's compute expenses. So once you have access to a data set, you can see it's this one, Lily. So it has all the tables extracted from our Lily software. So we'll use the most common used one, derived gas output. So this table contains all the gas consumption for every single message that's happened over the blockchain. Okay, so here's the table schema. If you'd like to learn more about table schema, feel free to go to our Lily documentation website. It has a very comprehensive model of documents here. Okay, let's go back here. So here I'll try to demonstrate how we can get the base fee and the average base fee and then the average total cost per message for every single day since Genesis. So I'm not sure how people familiar with SQL query, but here we try to group the height by date. So we divide it by 2,800 AD and then pass it back. So all the message on happening the same day will be grouped together. And then we calculate the average base fee round and the total cost total cost is calculated by adding base fee minor tips and overestimate appearance. And we divide and since the unit here is the atom field, so we divided by 10 to 18. And also the work condition here, we only want to filter the successfully accepted messages. So the exit code is zero. Okay, so we can run the query here. Okay, so here's the result. Let's look at the truck information. So you can see that this query scan does a full table scan of derived gas output. So you scan around 60 gig of data. And if you're interested, you can check out the execution graph to see how many rows you scan here. You can see that we scan one billion rows from the table. And it does aggregation, basic computation and aggregation and then generate the results. Okay, so for the result, we can easily visualize the data, quickly explore with sheet, connect with the BigQuery result in the Google spreadsheet. Here we can create a chart that's creating a new sheet. It's like line graph time and we select both the metrics. Let's sort it by time and see how it renders. Okay, you can see it's a little bit weird because in the beginning of the network, the fee might not look normal. So let's try to add some filtering here. So we expect the average cost should not be more than one field. So that's filtered anything beyond that. Apply the filter. So here it is. It's a full historical data for the average gas fee. So you can see after the FEVN launch, we do see an uptick train. But keeping in mind that this query is just for demo purpose. You usually want to filter it by the type of message that's applied to the blockchain instead of doing the average across all messages. Okay, let's look at another example. So we have FEVN specific tables here. You can see the FEVN active states, block headers, contract, et cetera. So let's see. Let's say we want to know how many transactions FEVN transactions per day since the network upgrade FEVN launch. So here we do similar query, but we're just counting the number of rows. And here we filter out the height based on the FEVN launch epoch. So let's run this. Okay, cool. Web result. Let's check it out in the Google Sheet. Again, let's pick our chart. No stacking section sorted by time. Here's all the number of transactions per day. There's some missing dates, we're still backfilling the data, but it should be complete very, very soon. Cool. I think that some of the demo today from the party, feel free to reach out on Slack, feel dash sentinel, or ping me directly on Slack via email. I'll be more than happy to answer your question.