If you've ever used elasticsearch and needed monitoring and alerting for your cluster without the overhead of running tools in the cluster itself, you'll realize there's not many options. So came up with the idea to separate monitoring and alerting for elasticsearch clusters to create Pulse. Close to launch. https://www.espulse.com/

You could easily do this with Elasticsearch. I've created a search UI(http://www.github.com/romansanchez/calaca) for Elasticsearch, which is what I think you're trying to do. You may want to recreate it on your own for learning purposes.

Is the demo down? Looks very useful by the way.

Working for us! Here's what we see when we click the "Try the demo" button: https://i.imgur.com/ZUi3sib.png

Have you tried elasticsearch?

Couldn't find in the docs. Does this support epoch timestamps? Does it support more granular x-axis ticks, eg hours, min, sec?

I would honestly just use this to start off and save time, but once things start picking up, data export and in house user mgmt would be the move.

Has any body heard of TIBCO? They've been doing this (flow-based programming) for years(>10).

Does anyone know if the software used for HFT is built in house or is there a particular vendor that specializes in this particular market?

From what I have read, its all in house, the software (including the mathematical algorithms) is the most important center point of a HFT shop. No vender would sell a competitive software system because they would make more money using it themselves.

In-house. very very much in-house.

I would say this pretty much killed Carsabi.

Actually not. Enough of their content is direct from dealers that it looks like they will be able to bounce back.

Yeah, they still have quite a lot of content, and to be honest craigslist listings were some of the worst on there. Still, it was handy to just have one go to place for car shopping.

I'm curious if videos really do scare people away. If you take a look at dropbox and cloudflare they do a really good job at using a video to demonstrate their product. Now taking into consideration the size of dropbox's and cloudflare's users, I'm more inclined to the idea that "videos work".

It's not that videos scare people away, but they need to be augmented. Would Dropbox convert better if they had descriptive text on their home page, or at least made the features link more visible than the one in the footer? We'll never know unless Dropbox tests it and publishes the results.

When I think of software sold to businesses, I'm reminded of the place I used to work where the president had the IT department disable the speakers on all new PC's. He didn't want employees sitting around watching videos or listening to music on company time. It may be hard to fathom for some people, but there are offices where playing videos (even demos) is frowned upon - especially in a cube farm.

If you want to make conversions, think of all the barriers between you and the sale, and start knocking them down one by one. Making your products key features readable and crawl-able for search engines is an easy one.

As someone from the affiliate space where EVERYTHING gets splittested I can tell you that videos generally own all other forms of website content when trying to get the user to do something (download, signup, buy).

>Would Dropbox convert better if they had descriptive text on their home page, or at least made the features link more visible than the one in the footer? We'll never know unless Dropbox tests it and publishes the results.

of course DB tested their homepage a ton

Not sure about scaring people away, just pushes time-conscious people away, which is true for me personally

They don't scare me. I just don't bother watching them if I don't know what will be explained in them beforehand. I am a quick reader, but I cannot accelerate your video.

If video is your only explanation, I'll never come back.

I don't think I have ever seen Dropbox video, even though I have been using it for years.