Network Performance Dashboard
Een introductie tot Solana
Solana is een nieuwe generatie openbare blockchain technologie. Het platform bestaat uit een netwerklaag die theoretisch, zonder te compromissen met de veiligheid, tot 710.000 transacties per seconde kan verwerken. Het applicatienetwerk kan worden beschouwd als een alternatief voor Ethereum, Cardano en Polkadot.
Solana delivers scalability without compromising decentralization and security. The solution is the implementation of a Proof-of-History (PoH) timestamp system that works as a decentralized clock. In fact, this innovation solves the time problem within decentralized networks since within a decentralized network there is no trusted central point for time measurement. It simplifies network synchronization and therefore enables Optimistic Concurrency Control within a decentralized network. Below is an introductory video explaining this innovation.
More explanation ::
qustion
Create a dashboard displaying Solana network performance over time.
How has the network performed over the past month compared to the rest of the year? Has transaction per second and success rates of transactions gone up recently? Is this because of less botting or fewer users, or new improvements from the Solana engineers? What wallets and programs have paid the most in fees for failed transactions?
Method
I have used the data in this table (solana.fact_transactions)
program_ID = 'FsJ3A3u2vn5cTVofAjvy6y5kwABJAqYWpe4975bi2epH' then 'pyth oracle'
program_id ='9xQeWvG816bUx9EPjHmaT23yvVM2ZWbrrpZb9PusVFin' then 'serum dex v3'
program_ID = 'JUP2jxvXaqu7NQY1GmNF4m1vodw12LVXYxbFL2uJvfo' then 'jupiter aggregator v2'
program_ID = '11111111111111111111111111111111' then 'system program'
program_ID = 'DtmE9D2CSB4L5D6A15mraeEjrGMm6auWVzgaD8hK2tZM' then 'switchboard oracle'
program_ID = 'ComputeBudget111111111111111111111111111111' then 'Compute Budget Program'
program_ID = 'cjg3oHmg9uuPsP8D6g29NWvhySJkdYdAo9D25PRbKXJ' then 'Chainlink Data Feeds Store Program'
program_ID = 'mv3ekLzLbnVPNxjSKvqBpU3ZeZXPQdEC3bp5MDEBG68' then 'Mango Markets'
In the last 30 days, the highest fee has been paid on the chainlink platform In the 6 months of the first year of the pyth oracle platform, the highest fee paid is on this platform
Explanations and conclusions
According to the data I received through the graphs I drew, the activity on the Solana network or the number of successful and unsuccessful transactions has not changed significantly, either during the market crash or before it, and as it was, here the performance of the Solana network is due to The problems it has have been very weak
Solana network was disrupted in the early hours of May 11. The problem created in the consensus mechanism for creating a new block had occurred, and thus the possibility of creating a new block had been lost. This disruption lasted for 7 hours and the problem was solved by restarting the network clusters
The problem arose from here in the Solana network that this morning a group of robots created a large number of invalid transactions (transactions related to NFT issuance) using the NFT token issuing machine of this network called Candy Machine. The number of these transactions is estimated at about 4 million transactions (or 100 gigabytes of data) per second. It should be noted that the Solana network does not have a mempool and transactions in it are very cheap, especially compared to Ethereum. This has made bots able to direct a flood of transactions to the network.
This volume of input data to the network caused a disruption in the consensus mechanism of the validators, in other words, it became impossible to create a new block in the Solana network. The disruption lasted from 1:02 am to 7:30 am Tehran time.
Solana Status' Twitter account reacted to the incident in a post at the same time, asking validators to prepare for a restart. Since this disturbance in the network was of the saturation type (Congestion), it seemed logical to restart the clusters.
The next challenge was that some validators considered this action a form of censorship and were not willing to make this change. Now that two-thirds of the validators had to comply with this code for it to take effect, Twitter Metaplex, which is considered the largest foundation around Solana's development activities, announced in a post that it has been decided to add a code to the network that for every invalid transaction which are usually hit by bots, a fine of 0.01 SOL will be applied to that address. In this way, the repetition of last night's vandalism would be very costly.
In this way, by restarting the clusters and upgrading the validators' node software, the problem was completely solved.
These are the problems of Solana network that its engineers must fix