160 Request /s for a No DB Call, Simple File Serving on a Single Core System.
Why am I the only one not impressed by this.
My £4 a month server can handle 4.2M requests a day
471–479 of 479 posts
Re: My £4 a month server can handle 4.2M requests a day
#472Earlier quoted context omitted.
Abstractions almost always end up leaky. Spark SQL, for example, does whole-stage codegen which collapses multiple project/filter stages into a single compiled stage, but your underlying data format still needs to be memory friendly (i.e. linear accesses, low branching, etc.). The codegen is very naive and the JVM JIT can only do so much. What I've seen is that you need people who deeply understand the system (e.g. S…
What's wrong with relying on data engineers for data engineering?
Data engineers should be building pipelines and delivering business value, not fidgeting with some JVM or Spark parameter that saves them runtime on a join (or for that matter, from what I've seen at a certain bigco, building their own custom join algorithms). That's why I said it's only economical for big companies to run efficient abstractions and everyone else just throws more compute at the issue.
Re: My £4 a month server can handle 4.2M requests a day
#473tl;dr: a machine with 2 GB of ram that is doing pulling a blog post out of a DB and passing it through Django can do ~50tps
Re: My £4 a month server can handle 4.2M requests a day
#474People tend to severely underestimate how fast modern machines are and overestimate how much you need to spend on hardware. Back in my last startup, I was doing a crypto market intelligence website that subscribed to full trade & order book feeds from the top 10 exchanges. It handled about 3K incoming messages/second (~260M per day), including all of the message parsing, order book update, processing, streaming to we…
I totally agree! I run a data science department at a corporation and it's amazing how much of our job is done on our laptops. I have a Dell Precision. When I need more power (a very rare situation), I can spin up a GPU cloud server and complete my big analysis for under $5.
Re: My £4 a month server can handle 4.2M requests a day
#475Earlier quoted context omitted.
> I'm not sure exactly what you mean by "run out of TCP sockets", but theoretically speaking, the only limitation is how much memory is available to store the necessary info about the socket (like address/protocol info and process info). Probably the 65k port limit since each connection will get assigned a remote port, which can be solved by binding to multiple local ports and using a load balancer in front or using…
The 65k limit is per client address and port . Each client can have 65,535 connections to a single port on your server from each and every 65,535 of their own ports . (65,536? I dunno what would actually happen if you tried to use port 0.)
That would be the end of the Internet.
Re: My £4 a month server can handle 4.2M requests a day
#476People tend to severely underestimate how fast modern machines are and overestimate how much you need to spend on hardware. Back in my last startup, I was doing a crypto market intelligence website that subscribed to full trade & order book feeds from the top 10 exchanges. It handled about 3K incoming messages/second (~260M per day), including all of the message parsing, order book update, processing, streaming to we…
True. I can completely relate with this. I developed an open source crypto exchange connector [0] and created a fun twitter bot [1] on top of that. Currently twitter bot processes all the USDT market trades from Binance (around 260 markets with average 30,000 trades per minute) and calculates OHLC metrics every 15 minutes using InfluxDB. All these installations and calculations are done in a free tier 1 VCPU / 1 GB RAM AWS server (less than 10% CPU and less than 40% RAM usage always).
[0] : https://github.com/milkywaybrain/cryptogalaxy
[1] : https://twitter.com/moon_or_earth
Re: My £4 a month server can handle 4.2M requests a day
#477Earlier quoted context omitted.
What's behind the mouse cursor while you're doing it? Could it be the UX/UI layer keeping state up-to-date? Other possibility, do you have a gaming mouse with 1000Hz polling rate configured?
Yeah, it's a high quality mouse. But the only excuse for this is it's slightly cheaper to make everything USB. PS/2 worked much better. It was limited to 200Hz but needed no polling. Motherboards just stopped providing the port.
This polling is not done by the CPU, this is a common misconception. In a typical modern system the only polling that happens with USB is done by the USB host controller and only when there is actual data the host controller generates interrupts for the CPU to process it. Obviously, when you configure the mouse at higher frequency you will get more interrupts and hence higher CPU usage but that has nothing to do with the polling.
Re: My £4 a month server can handle 4.2M requests a day
#478What's the point of this post? OP is serving a file at 50req/sec. There is not even mention of a dB query. How is that able to relate to any kind of normal app? I guess that the post was written as an answer to the mangadex post [1]. Mangadex was handling 3k req/sec involving dB queries. It was not just a cached Html page. 50req/sec for a Html file is super low which shows that a $4 month server cant do much actually…
People just want to show their works, it's normal.
What I found strange is that not many people seems to be surprised about this and upvote.
Re: My £4 a month server can handle 4.2M requests a day
#479Earlier quoted context omitted.
Does anyone know if any of the current crop of standard databases make use of things like Direct IO (where available on Linux), I’ve seen some write-ups that indicate you can get eye-wateringly fast performance when combined with an nvme drive.
Scylla is built arround direct io.
Specifically, aio (async io).