For Techies: What size server, etc., will TM need to handle Model 3 reservations?

[attathomeguy]

I would bet the have the ordering infrastructure hosted on AWS with auto scaling turned on which is an amazing technology and I have seen it in action and if they are as smart as elon is they will have is hosted through multiple amazon zones and it should be fine

 

[attathomeguy]

They are not using amazon! According to a leaked order URL they are using a company called Quality Technology Services based in the Santa Clara Data Center

 

[ucla107]

They are not using amazon! According to a leaked order URL they are using a company called Quality Technology Services based in the Santa Clara Data Center

what's the leaked URL? I went to ordered and confirmed it at the store this morning, but i was too excited didn't click on the address bar on their ipad so i didn't really see the URL... i do recall that it's something like model3.tesla.com or something though

 

[EinSV]

I work in software development, therefore not the exact match for what is needed to estimate the IT infrastructure need. I would not dare to make a recommendation, but my gut feeling tells me, that Tesla IT will underestimate the demand. I would bet on their server going down under the load in the first 24 hour when the reservations open up for Model 3.

Good call, ZsoZso!

 

[omgwtfbyobbq]

I would bet the have the ordering infrastructure hosted on AWS with auto scaling turned on which is an amazing technology and I have seen it in action and if they are as smart as elon is they will have is hosted through multiple amazon zones and it should be fine

X2! I'm guessing they'll have however many instances serving the sign-up page, which are what could scale the most, and those boxes will send the requests (orders) to an IP with however many (10+?) gateways for the backend processing system.

 

[neroden]

Sure, they can (and probably do) use clustering and fail-over redundant servers. However, there are multiply layers that can become bottleneck. It is one thing to handle the incoming connections in the web server layer, but your DB also need to be able to handle the connection numbers , or you need to use multiple distributed DB servers too. Then you may hit the bottlenecks of the network switches and routers. Large server farms like Google need to build up redundancy and distribution at every level.

Based on the WHOIS record their website seems to be hosted by Akamai:
https://www.akamai.com

This is a cloud service provider, so they could have a service agreement with them to grow the service on demand.

Akamai's great -- and they specialize in parallelizing websites and databases. But I think the bottleneck is actually going to be the financial transactions. Financial transactions which relate to the same account -- namely the account where Tesla is collecting the reservation funding -- *cannot* be done in parallel because they have to satisfy the ACID conditions extremely strictly and need a traceable, nonconcurrent series of timestamps on the transactions. There are ways around this (i.e. actually setting up multiple bank accounts) but I doubt Tesla has actually done this (well, they probably have one account for each country, but I bet they only have one for the US).

 

[HankLloydRight]

As I said, any basic level of web platform can handle tens or even hundreds of thousands of transactions per SECOND.

If tesla gets 100,000 transactions today, that's 1.5 transactions per second on average. Even if the peak load is 1000x that, it's still peanuts for one single server.