Greeting

President of the HPCI Consortium Nakajima

In the General Assembly of HPCI Consortium held on May 24, 2016, I was elected as the President for next two years. As the President, I will do my best for the Consortium and HPC community in Japan, together with the member of the Board.

I have been personally involved in the Consortium since its establishment as a board member and then the auditor. This means I have been with the Consortium and its pre-establishment organization for almost six years to see its growth from the cradle. Therefore it is impressive for me to see that HPCI is working well with K-computer and other many supercomputers as the platform for a wide spectrum of HPC researches and the Consortium is also working well as the organization for the integration of community’s will. However, we have to remember that “working well” sometimes means that we are in a local optimum from which we hardly jump to the real optimum adapting the change of the environment surrounding us.

In fact, we are facing the retirement of K-computer being our key-stone of HPCI in very near future, bringing a window without flagship machines until post-K machine is developed in Flagship 2020 project. In the window as well as the period after the post-K starts its operation, it is not guaranteed that the framework having been “working well” will also work well. In addition, when we have the post-post-K computer, we will be in the post-Moore era in which it is unknown whether the current scheme of HPC is still working well.

Seeing such circumstances, I strongly feel that the Consortium should make its proposals and raise various issues more aggressively. Though as the players of our game of HPCI we have to take care of the feasibility of our proposals, we must not interpret the word feasible as least-conflictive. That is, we have to present the ways which we believe to be best for our high-performance computing and its community though some of them could bring conflicts. For such aggressive proposals and issues possibly with conflicts, it is essentially necessary for our Board to have opportunities to discuss the matters not only with the members of the Consortium but also people inside/outside our community as many times as possible. I would like to have such opportunities by means of the meetings for which we invite members and their colleagues to the board meeting room following the good practice made by our former President, Prof. Kozo Fujii, and the town-meetings which we had frequently held in the first year of the Consortium.

Finally, we board members would like to solicit your kind support and cooperation to our job which we will perform with our largest effort for the members of the Consortium and our community.

To Page Top