FB's New Real-time Messaging System: HBase to Store 135+B Messages a Month
highscalability.com
highscalability.com
I swear that every couple of months I hear about a situation that basically comes down to this:
Boss: "Hello, Ryan, we would like to do $impossible_thing""
Ryan: "Huh, that sounds kindof impossible. But I guess we could get a team of the smartest people in our industry together and build an entirely new suite of tools to make $impossible_thing into $totally_normal_thing..."
Boss: "Yeah, do it!"
6 months later
Ryan and boss (to internet): "Hey, guys, check out this thing we made. Here, you can have it for free. Have a nice day!"
So just because we couldn't recite an exact response with compilable code doesn't mean we're dumb. We just aren't experts in that particular topic. Hang in there!
I looked at your profile and saw this:
>a failure of a hacker who doesn't have the energy to do much of anything anymore
Failure of a hacker? Pardon my language, but I fucking doubt it. Prove me wrong and tell me why you're a failure, please.
I find myself falling into this ridiculous trap all the time. It's easy to do too... I mean, holy shit, I'm posting on hacker news, a place where multi-millionaires posting about their wildly successful startups is the norm.
I'll find myself thinking that I'm a failure because I don't work at facebook, and I'll probably never be rubbing shoulders with people at the googleplex. I mean...my god, what a total worthless waste of space I am that I'm not a member of the top %0.001 of programmers in the US? Because my office isn't in some uber-cool San Francisco loft? Because cisco isn't employing a team of engineers to try and build switches that can handle the amounts of traffic that my data-center crushing websites produce? Because I'm not routinely completely changing the game in application development?
That's ridiculous. I really really really highly doubt that you're a "failure of a hacker", the fact that you post here pretty much proves that you're not.
HBase has been around for a while, and is part of the Apache Hadoop project. You don't need Zuckerberg signing your checks to work on cool projects along with really smart people.
Out of curiosity, am I wrong about this? Cassandra was developed by facebook, wasn't it?
(I was asking because this was downvoted. I thought maybe I was mistaken about cassandra)
I can't understand why that would matter, and frankly I find the notion very depressing.
Being able to deploy to 500MM people has got to be pretty cool, though.
That's awesome :)
It depends on the developer's favor to pick up a system. If it's me, I will pickup cassandra. The consistency model is not so hard to work with.
Well, I guess we know now! Definitely looking forward to all the future developments of HBase. There are a lot of interesting contribs being made.