randyzwitch

Yes, because it’s usually pretty inexpensive (I think I paid $35) relative to having a perfectly lit, clean background picture when one is needed at the last minute

For those of you who want to skip the article and just play with the data, you can view our live GPU-accelerated demo here:

https://opioid-demo.omnisci.com

What are all these new, 'safe' languages written in that they don't eventually have potential memory issues as well? Don't almost all languages provide C/C++ interoperability?

As a DBA, you don't consider yourself a developer?

Because it's irrelevant what the person currently makes. It only matters what the company is willing to pay for the work they are hiring for. Requiring the applicant to tell their current salary is only useful as a means of salary deflation.

Can you point to a commercial GPU database offering that doesn't distribute calculations over multiple GPUs out-of-the-box? That's table stakes, not a point of differentiation.

The p2.16xlarge EC2 GPU instance has 16 GPUs, 732GB of GPU RAM.

(OmniSci/MapD employee here)

In my opinion, I don't see this as undermining the OmniSci commercial offering, but rather opening up GPU accelerated analytics and visualization to a wider audience. A smaller piece of a much larger pie benefits all of the contributors to the RAPIDS project (of which OmniSci is a contributor as well)

Besides VCS unfriendly, the storage space for all the additional space characters provides little utility. For the most part, computers are reading CSV, not humans (though we might scan them).

This article is a bit out of date at this point (2017), but it gets the history correct. To see an updated Tweetmap demo (the demo shown in the carousel), check out:

https://www.mapd.com/demos/tweetmap/

Good point. While this demonstration is for a single bike-sharing program, there are over 200 feeds that conform to the GBFS specification, most updating roughly every 10 seconds.

It is a simple demonstration of how different tools work together. Feel free to do something else.

The data for this example are in aggregate, number of bikes currently at a station along with other selected information about the bike share location. The lowest level information is at the bike_id level, no information is transmitted about which customer is on which bike/bike_id.

https://github.com/NABSA/gbfs/blob/master/gbfs.md

I do, Jowanza does not

For this example, I set up a hosted Kafka cluster on Azure using HDInsight, as I didn't want to mess around with setting up Kafka. It's probably an expensive way to solve this problem long-term, as we're not really using anything exotic that a stock Kafka install wouldn't do out of the box.

https://docs.microsoft.com/en-us/azure/hdinsight/kafka/apach...

The example will obviously work with any API, so we could. But not sure using citybik.es adds anything here for my purposes.

Thanks for reading! I worked with OP (I'm a MapD employee), and yes, for this smallish example this is overkill. However, we're planning on adding a lot more feeds[1] to this pipeline which will make the tools we used a lot more necessary

[1] At the time of writing, there were 213 different bike share feeds https://github.com/NABSA/gbfs/blob/master/systems.csv

Netbooks are the next big thing! Oh wait, we already tried that...

Assuming you don't need to compile software

As long as re-using fast FORTRAN code is better than re-writing it all yourself, people are going to use FORTRAN.

This is what I did, used a desktop with Linux as a remote server of sorts (while also having a monitor attached). As I got more comfortable with Linux, my usage of my MBP dropped

Especially when the company has ties to the founder/CEO.

They are a BI dashboard company with lots of data integration connectors. They've been burning money with ads declaring 'spreadsheets are dead' and silly Domopalooza[1] lifestyle events

[1]https://www.domo.com/news/press/domopalooza-2016-to-feature-...

[2] https://www.domo.com/news/press/kesha-joins-the-domopalooza-...

[3] https://globenewswire.com/news-release/2018/02/15/1349133/0/...

Who would answer questions on newbie-stackoverflow? People who could answer questions would never see interesting content, because by definition it would always be below their skill level.

Weird that a visualization blog post has code, but not the visualizations it produces.

That's unfortunate if true. I have a few unwatched videos from the course, so it still remains for me.

The demo is back up now, thanks for your patience!

Thanks for letting us know, it looks like it might be down temporarily. In the meantime, the other demos work just fine: https://www.mapd.com/demos/

(I'm a MapD employee)

The limitation of 100 million rows is just for 14-day trial purposes; for the paid versions, the lower volumes are quoted as 'Getting Started' prices. For enterprise scale, people can contact MapD to get customized pricing.

As far as the scaling question goes, MapD has a public demo on the website with 11 billion records (and product performance scales somewhat linearly as more GPUs added): https://www.mapd.com/demos/ships/

I found Udacity CS344: Intro to Parallel Programming (CUDA) a great class, not only from a practical standpoint of learning CUDA but also had some decent explanations/whiteboarding behind the algorithms