こんばんは，M1の古畑です．
2020年11月25日(JST)から2020年11月27日(JST)にかけて開催されたCANDARW’20 (The Eighth International Conference on Computing and Networking Workshops)にてオンラインで発表を行い， Workshop Best Paper of LHAM: International Workshop on Large-scale HPC Application Modernization を受賞しました．
COVID-19の影響によりオンライン開催となり，開催予定地であった沖縄での発表はできませんでしたが，大変有意義な経験になりました．
学会で発表することができたことも，賞を頂くことができたことも，共著者の方々と研究室の皆さんの協力があったからこそ成し遂げられたものでした．
日頃から研究活動を支えてくださっている研究室の皆さんへ改めて感謝を申し上げます．
Improving the Accuracy in SpMV Implementation Selection with Machine Learning
Reo Furuhata*, Minglu Zhao*, Mulya Agung†, Ryusuke Egawa‡, and Hiroyuki Takizawa†
*Graduate School of Information Science, Tohoku University
†Cyberscience Center, Tohoku University
‡Department of Information and Communication Engineering, Tokyo Denki University
Workshop Best Paper of LHAM Award @CANDARW’20
Good evening, I am Furuhata, a M1 student.
I had a presentation in CANDARW’20 (The Eighth International Conference on Computing and Networking Workshops), from November 25, 2020 (JST) to November 27, 2020 (JST), taking place virtually this year. And I received the Workshop Best Paper of LHAM (International Workshop on Large-scale HPC Application Modernization). Even though I couldn’t go to Okinawa due to COVID-19, it still was a very meaningful experience for me.
It won’t be possible to receive such an award without all of co-authors and labmates’ help. I would like to express my gratitude again to all the members of the laboratory for their support of my research activities.
Improving the Accuracy in SpMV Implementation Selection with Machine Learning
Reo Furuhata*, Minglu Zhao*, Mulya Agung†, Ryusuke Egawa‡, Hiroyuki Takizawa†
*Graduate School of Information Sciences, Tohoku University
†Cyberscience Center, Tohoku University
‡Department of Information and Communication Engineering, Tokyo Denki University
Reo Furuhata*, Minglu Zhao*, Mulya Agung†, Ryusuke Egawa‡, Hiroyuki Takizawa†
*Graduate School of Information Sciences, Tohoku University
†Cyberscience Center, Tohoku University
‡Department of Information and Communication Engineering, Tokyo Denki University
SC20においてアグン(Agung)さんが発表
SC20 UrgentHPC Workshopにおいてアグンさんが発表を行いました．
発表の概要は以下のSC20のサイトより閲覧可能です．
Agung gave a talk at SC20.
Agung made a presentation at the SC20 UrgentHPC Workshop.
The summary of his lightning talk can be accessed from the SC20 website:
SC20において江川客員教授が発表
Prof. Egawa gave a talk at SC20.
Professor Egawa made a presentation at SC20.
The contents of the presentation are as follows.
Ryusuke Egawa, Souya Fujimoto, Tsuyoshi Yamashita, Daisuke Sasaki, Yoko Isobe, Youichi Shimomura, and Hiroyuki Takizawa “Exploiting the Potentials of the Second Generation SX-Aurora TSUBASA” SC20, November 12
Fall Diploma Ceremony
On September 25, the Graduate School of Information Sciences held its fall diploma ceremony. At the ceremony, M2 student Mike Zielewski was awarded a master’s degree.
Message from the graduate:
> I am sincerely thankful to my professors for their guidance and to my labmates for their support during my studies. As I continue my research, I hope to be able to collaborate more with my fellow students.
秋の学位記伝達式
9月25日に情報科学研究科の学位記伝達式が執り行われ，M2のマイク(Mike)さんが修士号を授与されました．
卒業生からのメッセージ：
教授のご指導，それから研究室の仲間の助けには心から感謝しています． 私はこれからも研究を続けるので，仲間の学生とより一層協力していきたいです．
NUG XXXIIにおいて滝沢教授が講演しました
9月24日にオンライン開催されたNEC User Group Meetingで、滝沢教授が第2世代SX-Aurora TSUBASAの性能評価について発表しました。
タイトルはEarly evaluation of the 2nd generation SX-Aurora TSUBASAです。
この第2世代SX-Aurora TSUBASAは、新しく10月1日からサイバーサイエンスセンターで稼働する「スーパーコンピュータAOBA」でも使われています。
https://www.ss.cc.tohoku.ac.jp/
性能評価結果は、第2世代SX-Aurora TSUBASAがメモリ律速の計算科学計算を効率的に実行できることを示しています。
スーパーコンピュータAOBAの稼働開始が楽しみです。
Prof. Takizawa gave a talk at NUG XXXII.
Prof. Takziawa gave a talk about performance evaluation of the second-generation SX-AUrora TSUBASA at NEC User Group Meeting on September 24.
The title is “Early Evaluation of the 2nd-generation SX-Aurora TSUBASA.”
The second-generation SX-Aurora TSUBASA is used in Supercomputer AOBA at the Cyberscience Center, which will start operation on October 1.
https://www.ss.cc.tohoku.ac.jp/
The performance evaluation results demonstrate that the second-generation SX-Aurora TSUBASA can efficiently execute memory-bound scientific computations.
We are looking forward to the operation start of Supercomputer AOBA.