Yep, that is what we have discussed as well. Currently I'm in the design phase of a real-time query engine on top of Hama. I guess we are calling it "Harq" and it will be somekind of hedgehog.
Once we graduated I will mail arround the Pig and Hive community who is interested in it, because I'm not very experienced in query parsing, execution plans and stuff. That is definitely on our TODO list. I just have to admit that I'd like to have Giraph running on top of Hama, because they have a much more mature graph functionality. But they didn't wanted to, so we are going to build our own. Am 25. März 2012 17:42 schrieb Apurv Verma <[email protected]>: > +1 > In the coming time we'll have to figure out, how to port all map-reduce > algorithms to an efficient BSP model. I was planning to learn to do this > during my summer vacations. > > -- > thanks and regards, > > Apurv Verma > B. Tech.(CSE) > IIT- Ropar > > > > > > > On Sun, Mar 25, 2012 at 7:02 PM, Praveen Sripati > <[email protected]>wrote: > > > Hi, > > > > Does it make to sense to have some tools/frameworks on top of Hama? Like > > Pig, Hive and others for Hadoop. Some of them can make using Hama easier > > and some add additional functionality on top of Hama. > > > > The intent is not to develop the tools immediately, but to start a > thought > > process and discussion around it. > > > > Thanks, > > Praveen > > > -- Thomas Jungblut Berlin <[email protected]>
