Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

What does it take to run a web app with 5K – 10K users?

news.ycombinator.com

Post reply on HN
Print view View original

61–70 of 104 posts

Re: What does it take to run a web app with 5K – 10K users?

#61

Post by jacquesm »

Gurkenmaster wrote: post #58
karterk wrote: post #2

The most difficult thing is going to be getting to 10K active users :) These days RAM is cheap and SSD storage is also widely available. For a very long time, one of my side projects with 50K users was hosted in a EC2 small instance. With that out of the way, here are a few things you will need to take care of: * Security (especially passwords) - Rails should take care of most of this for you, but you should ensure t…

> These days RAM is cheap Ironicially it's a completly different story on the consumer market. I bought my two 4GB sticks about 2 years ago and now they cost twice as much. 4GB can cost you $40 which is not cheap at all.

> 4GB can cost you $40 which is not cheap at all.

That's totally negligible on the cost of the rest of the machine. Imagine $160 will get you 16G!! That's an absolutely enormous amount of memory, most power users would be more than satisfied with that.

Not all that long ago that amount of RAM would cost more than a brand new mid sized car.

Re: What does it take to run a web app with 5K – 10K users?

#62

Post by jacquesm »

It mostly depends on what the users do. If the interaction is minimal and most pages can be rendered once and then served from cache you can run this on a laptop. If interaction is intense and everything has to be generated on the fly it could tax a mid range server. It all depends on the use case.

Re: What does it take to run a web app with 5K – 10K users?

#63

Post by metaphorm »

I can think of three things that you'll really really really want to be on top of while scaling up

1. your architecture must allow for vertical scaling. this means upgrading your hardware to beefier, stronger, faster machines with more CPU power and more memory. vertical scaling is often a very cost effective of improving performance.

2. your architecture must allow for horizontal scaling. this means being able to provision and deploy new instances of your application servers very easily, using an automated process. more servers running in parallel is a very effective way of handling increased load.

3. you must be able to monitor and protect your systems. https everywhere. highly secure passwords everywhere, and you should rotate your passwords on a regular basis. log everything and set up services to monitor your logs and notify you when weird/bad shit happens.

Re: What does it take to run a web app with 5K – 10K users?

#64

Post by gizmo686 »

reymus wrote: post #42
tlack wrote: post #34

Earlier quoted context omitted.

Except you won't know if your server ever sent it to Mandrill. :) Always be extremely verbose with logging!

I have always heard the opposite, that too much logging is as bad as no logging. I see the point of having the logs to be able to find out what happened, but what happens when there' s so much logging that the information needed is just buried into huge amount of noise?

My favorite systems to work with are the ones with overly verbose logs, where the overly verbose parts were clearly tagged and could be filtered out. Generally, we would never look at the verbose lines, and even when we did, we would normally have some idea what we were looking for, and be able to filter somewhat for it.

Re: What does it take to run a web app with 5K – 10K users?

#65

Post by wahsd »

phacops wrote: post #33

I have a little over 10k DAU for http://cronometer.com/ and it runs on 2 fairly small cloud instances + a cloud database. A single host can handle the load fine, but I keep 2 running to survive failures. And it took me nearly 4 years to get that many users. We can’t all grow like facebook!

I had never even heard of your site. Could it be you need to do a little marketing?

Re: What does it take to run a web app with 5K – 10K users?

#66

Post by arihant »

I'm not sure how system intensive Rails app is, but we ran a Facebook game with 20K daily active users on a $8/mo Dreamhost, back in 2008, with a fantasy league and all. My point is, unless you're doing something extraordinary, I doubt you would need more than a VPS.

As others have mentioned, multiple of those users can be hosted on an EC2 small instance. I suggest you start there. When moving to production, a bigger challenge is security, both in terms of intrusion and data protection. Making sure you have good rollback feature built into your rollout regime, because things can be fatal with real users. If you're using something that's basic like Heroku or EC2, you can scale way beyond that user strength with a click of a button. Scaling up would be least of your worries, at least for a few weeks.

If you're unsure, go with Heroku. Once you understand your system use, you can very easily switch to AWS and reduce costs.

Re: What does it take to run a web app with 5K – 10K users?

#67

Post by phacops »

wahsd wrote: post #65
phacops wrote: post #33

I have a little over 10k DAU for http://cronometer.com/ and it runs on 2 fairly small cloud instances + a cloud database. A single host can handle the load fine, but I keep 2 running to survive failures. And it took me nearly 4 years to get that many users. We can’t all grow like facebook!

I had never even heard of your site. Could it be you need to do a little marketing?

True, it started as a hobby, so it’s been fully bootstrapped and I’ve spent less than $2500 on actual marketing to date. Advertising in the Health / Diet space is super pricy, the ROI still doesn’t make sense. CPA >>>>>> LTV

Re: What does it take to run a web app with 5K – 10K users?

#69

Post by exelius »

reymus wrote: post #42
tlack wrote: post #34

Earlier quoted context omitted.

Except you won't know if your server ever sent it to Mandrill. :) Always be extremely verbose with logging!

I have always heard the opposite, that too much logging is as bad as no logging. I see the point of having the logs to be able to find out what happened, but what happens when there' s so much logging that the information needed is just buried into huge amount of noise?

This was true before Splunk. If you logged too much, your logs could start to outstrip the assumptions behind your log rotations and cause trouble. Now the common wisdom is to just log everything so you can Splunk it later if you have a problem. Verbose logging + Splunk have made production incident identification so much easier than it used to be.

Splunk DOES charge by the GB, but it's not very expensive in the long run.

Re: What does it take to run a web app with 5K – 10K users?

#70

Post by clogston »

chrissnell wrote: post #52
landr0id wrote: post #47

Earlier quoted context omitted.

> Logging, logging, logging - I can't stress on this enough. When things break, logging is crucial in piecing together what happened. Use log rotation to archive old logs so they don't hog the disk space. How do most people manage activity logs? Currently what we have set up is the user id (if the user is logged in), IP address, URL they hit, user agent, and timestamp are all inserted into an activity logs table. For…

Off-machine logging. There are commercial services (we're using Papertrail but there are tons of them), roll-your-own solutions (Elasticsearch-Logstash-Kibana), and simple solutions (syslog). For an easy and simple solution, spin up a second instance and send logs to it via rsyslog over a private network interface. Most mature frameworks provide a method to send logs over syslog. It's UDP and very lightweight. Anothe…

We're running the Elasticsearch, Logstash, Kibana (ELK) stack with the recommended approach i.e.:

  logstash client
                   \
  logstash client --> redis -> logstash "server" process -> elasticsearch
We have a high logging load (we log every request) due largely to IRS requirements. I've been really happy with it over the past 6 months but something that cannot be overstated is that you'll really need to become familiar with each one of the technologies used as each requires it's own setup and configuration. Not being familiar with any of them, it took me a solid 3 days to get to where the whole thing was usable and performant. Troubleshooting it is a breeze, and the whole system scales really easily, but a lot of that time was invested up front.
Post reply on HN