Is the Vertigo project still alive?

[Ming Fang]

Vertigo looks very interesting.

However Github activity on this project stopped early July.

Is Vertigo still alive? 

[Jordan Halterman]

Most definitely! Here's the gist:

Development has seemed to stop simply because the plan is for Vertigo 1.0 to be developed on/for Vert.x 3. However, Vertigo development hasn't actually stopped at all. I just haven't committed to that repo in a while. Until very recently, too many Vert.x 3 APIs were still evolving for it to make sense to try to do the initial Vert.x 3 port. So, rather than working on Vertigo code, I've been contributing to Vert.x 3 development and working separately on replication algorithms for Vertigo 1.0.

Vert.x 3 is going to be huge for Vertigo, so I/we have no intention of abandoning it. Here's why Vert.x 3:

- Vert.x will support *automatic* language API generation for libraries like Vertigo. One of the challenges with maintaining Vertigo in Vert.x 2 was the development of separate APIs for each supported language. I certainly don't know all the Vert.x languages, so it would have taken an army to get Vertigo where I wanted it to be (able to be written in any Vert.x language). I've been contributing to language API generation in Vert.x 3 to make sure this can happen sooner than later.

- Vert.x 3 has event bus proxies which will make for a very interesting interface to the Vertigo cluster. Until now, I've had to provide separate Vertigo event bus wrapper interfaces for working with the Vertigo cluster. Vert.x 3 will allow much of that code to be moved into the Vert.x layer.

- Similarly, Vert.x 3 provides cluster-wide data structures, another feature which Vertigo has to provide on top of the Vert.x event bus with Vert.x 2. Again, more code that will be moved up in to the Vert.x layer with Vert.x 3.

One of my primary goals in the design of Vertigo has always been to delegate as much responsibility to the Vert.x framework as possible, and Vert.x 3 is a huge step in the right direction in that sense. Aside from that, I don't anticipate too much of the Vertigo messaging model changing between versions. 

Most of the changes anticipated in Vertigo 1.0 revolve around fault tolerance and consistency. The tentative plan for Vertigo 1.0, aside from porting the existing API to Vert.x 3, is to provide a lightweight, configurable replication framework underneath the existing model. This framework will model Vertigo components as replicated state machines and guarantee fault tolerance and consistency both in terms of nodes and state. In order to accomplish this, I've been working on an algorithm that is fairly similar to the replication algorithm implemented in Kafka 0.8 (with some key differences that allow for better performance). Essentially, this will mean that each component *can* be represented as a log of historical events which can be replayed. The log can be configured according to performance and consistency requirements. This means that components for which state contributed to output, state can be rebuilt upon failures. But I digress.

Actually, stateful processing for Vertigo is a problem on which I've been working for a fairly long time, but it's finally nearing stability. Of course, Vertigo 1.0 will certainly be developed in such a way that these types of performance hogging features will be optional and disabling them will result in no performance impact (Vertigo is very fast in its current form). I'll post an extensive description of the new state and cluster management features here once Vertigo 1.0 rolls around.

Sorry to go on a tangent, but I guess that's all to say "yes!"

--
You received this message because you are subscribed to the Google Groups "vertigo" group.
To unsubscribe from this group and stop receiving emails from it, send an email to vertx-vertig...@googlegroups.com.
To post to this group, send email to vertx-...@googlegroups.com.
Visit this group at http://groups.google.com/group/vertx-vertigo.
To view this discussion on the web visit https://groups.google.com/d/msgid/vertx-vertigo/723017b7-053f-4890-b124-ae9334302799%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Ming Fang]

Jordan

Thank you for the update. This is very good news.

Now I feel confident in choosing Vertigo for our future systems.

I look forward to contribute to this great project.

--ming

[Jordan Halterman]

Love to hear it! :-)

Sent from my iPhone

To view this discussion on the web visit https://groups.google.com/d/msgid/vertx-vertigo/a3c3982e-1e2f-49e6-8965-fce1fb6f9996%40googlegroups.com.

[Oliver Rolle]

Do you have a roadmap when to expect the 1.0 RCs & release?

Can you tell us more about state management of components? I am highly interested in that topic! Do you use checkpoints to persists components states & than play back events to reach its final state?

[Anton Hughes]

Hi, yes am also interested to see Vertigo progress. I think this platform is potentially revolutionary. I really would like to hear if/when v1 is due to be released.

[Jordan Halterman]

Slowly but surely. I've gotten a little side tracked by some related projects and a lot of real work lately. But yes, I continue to work towards that end. 

When that might happen is difficult to estimate at this time. I've been working on the underlying framework for a long time, and it's pretty close to complete. Actually, I expect to be releasing some lower level portions of it this week and testing the complete framework within production environments within a couple weeks. I could theoretically finish the port without it, but with limited time I prefer to continue down this path until it's finished. Once that's done back to Vertigo. Surprisingly, I might still make it around the time Vert.x 3 officially arrives, but maybe not.

To view this discussion on the web visit https://groups.google.com/d/msgid/vertx-vertigo/1ee58f7b-9a3a-4207-8aa0-d1f88d72a49a%40googlegroups.com.

[Carter Youngblood]

I am also very interested!  I have been using version 0.7x in production for over a year now and it's been awesome!  I can't wait for the next one and the problems I can solve with it.  Best of luck, and this is a very exciting project!

-Carter