Hi Javi, Thanks for the interest. It is possible that Apache Slider will be retiring soon, as a service framework has been integrated into YARN as of Apache Hadoop 3.1.0 (see YARN-5079). I encourage you to take a look at the YARN services work, including a REST API for launching/managing long running services on YARN.
Billie On Sat, Apr 14, 2018 at 2:45 AM, Javi Roman <jroman.espi...@gmail.com> wrote: > Hi Slider community! > > I am an Apache Myriad committer and I would like share with you the > state of Apache Myriad (incubator) here [1]. In particular I would > like comment with you an interesting idea of development approach. The > idea is explained in the section: "Article 4 - Apache Mesos Data > Services a Different Approach". > > The interesting section pasted here: > > ".... a different way of data service development on Apache Mesos: > Probably the current approach is to create data services (such as > Kafka, Spark, Elastic ...) based on Apache Mesos Frameworks. The focus > of development is to create native frameworks for each technology. > Mesosphere does a very good work on that direction with DC/OS SDK > (dcos-commons). > > The different approach explained in this article is to shift the > development focus towards only one Apache Mesos Framework to maintains > (Apache Myriad) and create applications based on YARN by means of > Apache Slider (a "kind" of YARN SDK)" > > The use of Apache Slider as "SDK" for running Apache Mesos data > services by means of Apache Myriad is the main idea. I would like ask > for feedback about this approach, and try to get some collaboration > between both projects. > > What do you think? > > Regards. > > > [1] https://goo.gl/cbZK4y > > -- > Javi Roman > > Twitter: @javiromanrh > GitHub: github.com/javiroman > Linkedin: es.linkedin.com/in/javiroman > Big Data Blog: dataintensive.info >
