TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

First hand: Scaling under load spikes with Rackspace Cloud Server

4 pointsby zemajabout 15 years ago

3 comments

dacortabout 15 years ago
I've run TweetStats on EC2 for the past year using the great scalr.net to provision new instances.<p>It's expensive as heck (the smallest instance scalr allows is an EC2 large), but it works great.<p>Scalr monitors one of several stats you can specify (RAM, CPU, queue size, HTTP response) and will spin up instances (you can specify a max) as necessary.<p>Granted the app has to be built to support such multiple app servers, which may be easier in some cases than others.<p>Sounds like the cloud server was rather inconvenient, couldn't handle immediate load, and required your manual intervention. Always tough to predict traffic spikes from Twitter. ;)
DennisPabout 15 years ago
"It looked like a few hours of data was lost from the database for some reason"....eh, what? Was this the application's database? Losing a couple hours of data because of a traffic spike is not something I would find remotely acceptable.
zemajabout 15 years ago
Thought I'd give people a first hand look at dealing with this type of situation. Would really like to know how other people would have handled it.