Mongodb – not evil, just misunderstood • Hacker News Board

Re: Mongodb – not evil, just misunderstood

#51

Post by lsh123 » Fri, Dec 06, 2013, 4:28 PM UTC

raverbashing wrote: post #27

tilsammans wrote: post #14

I guess I am old-school but you are listing a bunch of reasons why MongoDB is not evil, yet each and every one of these reasons turns out to be extremely risky business. All of which simply do not apply with relational datastores. The thing I took away from your post is that had you used a 20-year-old relational datastore you would have 0 of your issues anyway. > The advantages of schemaless documents are priceless.…

"A migration is costly but also pretty rare. I migrate PostgreSQL with 100k+ rows as a matter of routine, it's over before you know it." Not if your application is still on development. 100k+ rows? Not complicated. Try with 23M rows. Nobody would risk migrating that table.

pt-online-schema-change, problem solved!

Re: Mongodb – not evil, just misunderstood

#52

Post by goldenkey » Fri, Dec 06, 2013, 5:39 PM UTC

lennel wrote: post #10

goldenkey wrote: post #9

Earlier quoted context omitted.

I actually have first hand experience from merely a small data set of 37,000 records of a couple fields each. I needed to run a batch update to calculate a property of each document. The property is dependent upon a count of other similar documents. Turns out that it was so unbearably slow, that there was no way to do the batch update without waiting hours.. I searched long and hard for a solution to the write/read l…

that is an paragon of maintainability if I ever saw one.

I hope that is sarcasm :-)

Re: Mongodb – not evil, just misunderstood

#53

Post by nasalgoat » Fri, Dec 06, 2013, 7:42 PM UTC

After my experience with MongoDB at scale - I was running one of the largest MongoDB installations in the world according to 10gen - I have since run into the arms of postgres and wish to never repeat the horrors I experienced over there.

I think the main issue is that people want something to be simple, and now that I'm dealing with the somewhat esoteric and opaquely documented postgres, I can understand that feeling. Native replication and auto-failover (via pgpool) is a bit of a black box on postgres, but under MongoDB it was fairly simple.

The problem is, it's complicated for a reason, and that reason is scale. What took me over 100 masters in MongoDB will only need two postgres boxes, so it's worth dealing with the Oracle of Open Source to make it happen.