Just when I was in the process of going low about my research topic, something wonderful happened in the last two days.
1. I got good reviews for my PACT paper and there is a very good chance that it can be accepted.
2. Today two guys came to give talks. It was one of the busiest days SPCL. One of those guys were Edgar Solomonik, a kind of genius from UC Berkeley. At the age of 24, he is almost finishing his PhD and hoping to join as a Post Doc in our group. I should tell that if he joins our group, it will be really a feather in the cap. But his talk was so fast and so dense, we made a lot of jokes about his "super intellect" and how we will quit PhD and become "pizza delivery boys" after he joins our group to save ourselves from depression.
The other guy was Satoshi Matsuoka, a professor and big name in the area of supercomputing in Japan. His talk was mainly on the future of big data and it covered many other topics. It was one of the longest talks ever and he really gives LONG answers if someone asks questions. But I like his flawless english pronunciation.
But the big thing is, both of these guys were interested in the generation of performance models. Edgar mentioned in the talk that "I know Torsten that you are looking at machine learning methods to generate performance models automatically..." I then exchanged glances with Torsten because I was the one working on it.
After the talk by Dr. Matsuoka, Torsten told me to have a conversation with him. And this conversation really made my day. They were working on a VERY similar topic on the generation of performance models and he was REALLY interested in my tool, PEMOGEN. He told there is not a useful tool on this topic and a good tool is necessary. He also invited me to work for some months in Tokyo (which will be one thing off my bucket list , Visiting Japan). We exchanged email IDs and when I told Torsten this, he told me that this will be a great collaboration.
After this conversation we were gossiping in the la when Timo also mentioned that the research on performance modelling is really cool and one will be really famous if it comes out.
Today I found one use cas eof my research and trust me, this is a great motivation for my PhD. Looking forward for the final PACT answer.
1. I got good reviews for my PACT paper and there is a very good chance that it can be accepted.
2. Today two guys came to give talks. It was one of the busiest days SPCL. One of those guys were Edgar Solomonik, a kind of genius from UC Berkeley. At the age of 24, he is almost finishing his PhD and hoping to join as a Post Doc in our group. I should tell that if he joins our group, it will be really a feather in the cap. But his talk was so fast and so dense, we made a lot of jokes about his "super intellect" and how we will quit PhD and become "pizza delivery boys" after he joins our group to save ourselves from depression.
The other guy was Satoshi Matsuoka, a professor and big name in the area of supercomputing in Japan. His talk was mainly on the future of big data and it covered many other topics. It was one of the longest talks ever and he really gives LONG answers if someone asks questions. But I like his flawless english pronunciation.
But the big thing is, both of these guys were interested in the generation of performance models. Edgar mentioned in the talk that "I know Torsten that you are looking at machine learning methods to generate performance models automatically..." I then exchanged glances with Torsten because I was the one working on it.
After the talk by Dr. Matsuoka, Torsten told me to have a conversation with him. And this conversation really made my day. They were working on a VERY similar topic on the generation of performance models and he was REALLY interested in my tool, PEMOGEN. He told there is not a useful tool on this topic and a good tool is necessary. He also invited me to work for some months in Tokyo (which will be one thing off my bucket list , Visiting Japan). We exchanged email IDs and when I told Torsten this, he told me that this will be a great collaboration.
After this conversation we were gossiping in the la when Timo also mentioned that the research on performance modelling is really cool and one will be really famous if it comes out.
Today I found one use cas eof my research and trust me, this is a great motivation for my PhD. Looking forward for the final PACT answer.
No comments:
Post a Comment