The paper was prepared by the group that I also worked with before. As I've heard the barrier synchronization was the greatest "barrier" to scaling their solution with HAMA. They had to also overcome memory problems with message buffering. If you're interested, you could ask for the HAMA patch that did the tricks ;-)
On Mon, Oct 17, 2011 at 9:58 AM, Edward J. Yoon <[email protected]> wrote: >> looks like my university (Roma Tre) was deeply involved there, will try to >> have some other feedback personally :) > > Haha, don't abandon us what they say, tommaso. Just kidding. It would > be really nice if we could get feedback from them. :) > > P.S., The paper is published at QDB workshop at VLDB'11. > > On Mon, Oct 17, 2011 at 4:35 PM, Tommaso Teofili > <[email protected]> wrote: >> looks like my university (Roma Tre) was deeply involved there, will try to >> have some other feedback personally :) >> Tommaso >> >> 2011/10/17 Edward J. Yoon <[email protected]> >> >>> http://qdb2011.dia.uniroma3.it/participants/program/p11-SIDLO.pdf >>> >>> -- >>> Best Regards, Edward J. Yoon >>> @eddieyoon >>> >> > > > > -- > Best Regards, Edward J. Yoon > @eddieyoon >
