"Brothers! Aptos node registration is here, if you grab it, you will be a big hair! If you win, you will be a prostitute! Get up!"

In the past two months, such content has often appeared on Twitter. From Aptos AIT 1 to AIT 2 (AIT is the Aptos network test, which is divided into 4 stages), some big Vs have spared no effort to promote the registration of Aptos nodes, claiming that this emerging public chain has the endorsement of Meta and Diem and the endorsement of top institutions. With the support of huge funds, it must show unlimited potential.

AIT 2 is currently in progress and will end on July 23, Beijing time. At that time, Aptos will airdrop token rewards to qualified nodes.

But in fact, the probability of Aptos nodes being selected is very small. Taking Aptos AIT 1 on May 14 as an example, the official disclosure revealed that about 30,000 nodes registered, and only 100 nodes were selected in the end. When it came to AIT 2, many Aptos believers rushed in. Taking into account the enthusiasm for participation, the project party also increased the number of candidate nodes at this stage from 100 to 200.

So,

So,What is the actual status of Aptos AIT 2, which has been running for a while? What are the costs and benefits of participating in node pledge? What problems might you encounter?I communicated with SID, the node operator who participated in Aptos AIT 2, on the above issues. The following is the content arrangement:

Odaily: As a non-Web3 practitioner, how did you pay attention to Aptos? Why did you want to participate in Aptos' testnet?

SID：Although I don't work on Web3 full-time, I have been following the development of Web3. I have 4 years experience in cryptocurrency trading and I am also a development engineer, so I am very interested in Web3 technology and am learning Solidity language.

Therefore, when I saw that Aptos had the endorsement of Meta and Diem, and entered the public eye with huge financing and support from top institutions, I was quite interested to learn about it.

I was very interested in Diem's ​​development language Move. In my opinion, Move is relatively rigorous and has been heavily modified for the security holes that are prone to occur in Solidity. What kind of sparks can such a combination collide with a language that does not seem to be delicate and mature enough but is very distinctive, coupled with a mature and emerging team?

In addition, Aptos, as an early Layer1 project, officially claims that it has: safe, scalable, and upgradeable smart contracts. As a development engineer, I certainly want to participate.

Odaily: As far as I know, the competition for Aptos nodes is fierce, and only 200 were selected in the end. Do you think there is any tendency for Aptos to select nodes?

SID：At present, the official number of AIT 2 nodes has not been announced. I borrowed third-party tools and community node distribution tools to judge that 30,000 nodes should not be much less than AIT1, and the specific number has to be officially announced.

I personally think that node selection is related to geographical location. Since Aptos is doing Layer 1, it is of course hoped that the nodes will be distributed in different regions of the world. Because different regions will have different network delays and other objective problems, it is more conducive for Aptos to verify the stability of the entire system.

Odaily: As a verifier to participate in the node pledge test, do you have high requirements for hardware equipment? Need a programming background?

SID：The officially required hardware must be at least 4-core 8G, 300G hard drive, but I don't use that much at the moment. No programming background is required, but a basic knowledge of Ubuntu and Docker is required.

Odaily: Is there any scoring standard for running the testnet? How to determine whether to receive token rewards? How many eligible nodes are there currently?

SID：There are three scoring requirements for running a node, LIVENESS network activity, PARTICIPATION proposal participation and voting in the consensus round.

• Complete all requirements on time (July 23rd deadline)

• LIVENESS requires >= 95%

• PARTICIPATION >= 95%

• Vote in at least 25% of consensus rounds

The above requirements are indispensable.

At present, the actual number of participating nodes is 212, and about 111 meet the standard. The number of qualified users increases every day, but according to Aptos' algorithm mechanism, the value of PARTICIPATION is more difficult to increase at the later stage. According to the current fluctuation of PARTICIPATION, it is estimated that it is difficult for users with less than 90% of PARTICIPATION to reach the qualified standard. There are also nodes in the group who said that if the AIT1 standard is followed (50% reward for PARTICIPATION 5%-94%), the final toss is no different from lying flat.

Odaily: How much did you spend in the audition stage? How much does it cost to run a node for a day? Where is the main cost?

SID：The cost of each small partner participating in the test is different, which depends on the cloud service provider you choose to build the test node. Of course, I personally recommend that the small partners who participate in the test choose a brand cloud service provider, at least to ensure its safety and stability, but the cost will also increase. slightly higher.

After the stimulus test started, my server traffic cost was about 40 U per day. It takes 10 days to run, plus the cost of the server, the overall test cost is quite high.

In the first two days of the incentive test, the traffic fee increased significantly. Fortunately, Aptos officially released the optimization plan in time, and the traffic dropped significantly.

The main cost is traffic, and the traffic cost is different depending on the node area and cloud service provider.

Odaily: The test network is running until now, have there been any problems? How did the authorities solve it? Will this affect your normal node operations?

SID：Did run into some issues:

• I encountered node panic and status synchronization blocking problems, but the official updated the latest version of testnet in time. As long as the personnel participating in the test update the nodes in time, they will basically not be affected much. Here, I would like to thank "Boyu_Chen" from the community for synchronizing the official information in time.

• The high traffic problem, I personally think it should be a DDos attack. Two days before the start of the incentive test, it was indeed affected by the high traffic problem. On July 15, the official made an optimization, switching the consensus signature information from json to BSC, and the update is complete. The traffic dropped significantly after the release. But some people's scores did not rise but fell.

• Optimized to improve bandwidth consumption of disk writes.

In my opinion, these problems have affected the overall node operation to varying degrees, and have also increased the budget for testing. During the 1-2 days of the DDOS attack, I had to pay more than 70 U of traffic fees per day. Some users were caused by these problems. The PARTICIPATION data dropped. My data started to drop after the high traffic problem occurred. Although the latest version of the testnet has been upgraded in time, it was still affected a lot. After a few hours, the data rose steadily and slowly.

There is a problem here. Many community node testers have been actively participating in the test, but the value of PARTICIPATION has not increased significantly, but has decreased. Maybe it has something to do with the network environment, but so far I haven't seen a clear official answer. Some users have also tried to make some changes, such as increasing bandwidth, improving server configuration, and switching regions, but it seems that they have not solved the problem of PARTICIPATION not increasing.

But since it is a test, there must be problems exposed, and solving the problem is the ultimate goal. In general, Aptos runs according to the official test documents, and the core functions of the test have no serious bugs, and the overall test is relatively smooth.

Odaily: It is understood that there will be Aptos AIT3/4 testing in the later stage. Do you plan to participate? Why? Is there any advice for users who have not participated?

SID：The functions tested each time are different, and the testing standards are also different. Of course, we will participate (provided that we are lucky enough to be selected).

I personally feel that it will cost a certain amount of money to participate in the node test this time. There will be a lot of uncertainty in the process of participating until the final selection. If you want to participate, you should hold an optimistic attitude of experience and learning. I also want to remind everyone here that there are currently communities selling Aptos accounts, and the accounts are priced according to the price converted into tokens from the operating costs. There is a one-year lock-up period, so users who have not participated in the node pledge should identify it by themselves to prevent being cheated.

In addition, a reminder of a little risk: According to SID’s estimation of the total input cost and the total number of Aptos tokens obtained in the end, the current theoretical currency price range is lower than 30-39 U, but there are still great differences in this price. Be cautious and it is best to get tokens from official sources.