Slashdot Mirror

True for consumer CPUs. But for servers, my company regularly makes use of 40-core machines (2x20 core intel CPUs). While our services are all single threaded, with sharding it is possible to utilize all the cores just by running 40 or so instances of the service on each machine. (It works surprisingly well: Sharding is the secret ingredient of the web scale sauce)

If you go to Amazon and browse their instances, you will see many instance types max out at 32 or 40 cpu cores (8xlarge and so on) so this is not uncommon. Amazon can simply resell the instances 1 or 2 cpus at a time if they want. So the fact is that while most applications are still not (heavily) multi-threaded, with each passing year it's possible to do more with a single CPU at least for servers.

Now, when talking desktop and mobile, I have to agree that we are most certainly not utilizing CPUs efficiently. Almost all frameworks (including iOS [UIKit] and Android) require all UI work to be done on the a single "main thread". And while it's possible to use other threads, they will almost certainly not be used efficiently because none of them are allowed to do UI work. The one redeeming factor for mobile is they might be able to save power by shutting off the cores which are not in use.

By the way, I've had very little trouble programming on ARM. First, if you use locks that implies a memory barrier (at least in pthread). Second, all processors since Cortex-A9 guarantee cache coherency just like x86, which is most of the phones on the mobile market running a recent OS (say, almost all devices running android 4.1 or later; and iPhone 4s or later: iPhone 4 was single core).

But is it web scale?

But mongoDB is webscale

> Security seems to be almost an afterthought.

By the time you get done activating security high availability, and integrity checking, you've often lowered the performance to below that of a more mature, well-tuned SQL database. There is a very amusing video that covers some of these issues in passing:

                      http://www.mongodb-is-web-scal...

And of course, MongoDB is web scale!

MongoDB is web scale.

200 comments and nobody has asked whether it's webscale or not. This place is going to the dogs already.

An individual who writes a book cannot realistically make a movie version of that book, because A. they don't have that kind of money, and B. they are not likely to have the required skill set.

I couldn't help but read that in the synthesized tone of "MongoDB is web scale" made with XtraNormal. Anyone who has a computer and can write a screenplay can produce an animated short film. And tools to mark up a screenplay for conversion to an animated film will only get better. Then you can send excessive copyright terms to /dev/null.

However, the law would need to specify a maximum contract duration beyond which those limits kick in so that companies wouldn't license it for [lifetime of copyright minus one day].

Or the director could just avoid doubt by exclusively licensing the motion picture to the studio for 20 years. This would be long enough for the theatrical release, the current home video format, and the next home video format, after which point the copyright reverts to the director.

And my stance is that I first make sure it is correct, then I worry about optimization.

I know too many projects that were done in the reverse. Sure, we have a slick and fast system, but only when it works. It's kind of like saying /dev/null is a fast database.

I understand the concept of big data, I used to do Hadoop back in 2010. But my point is that their example code just it seems to implement spatial hashing in a distributed database, which has been around for a while. I think the summary missed what makes these guy's approach better.

It seems pretty obvious that you should use some type of indexing in the database to select items rather than do some cool O(n^2) operations when you have billions of items.

Also, webscale

Choose MongoDB, its web-scale. http://www.mongodb-is-web-scale.com/

MongoDB is web scale!

See that?

That's you, that is.

http://www.mongodb-is-web-scale.com/