David G. Andersen AMA (SOSP Day 3)

Dave has broad research interests in computer systems in the networked environment.

What do you think systems community should be working on but isn't getting enough attention?

With the stuttering of Moore's law as we get into nanoscales, there is more need to extract performance from systems through integrated co-design of hardware and software. We need integrated work through the entire stack to make systems faster and more reliable.

How do you pick research ideas/projects?

I stumble on interesting questions. There is a lot of cross-fertilization going on between different areas I am working on. If everything else fails, once a year or so, I take a notebook, go for a walk, and write down my ideas. My two sabbaticals were also very fruitful for finding research ideas and projects. At Google, I worked with AI/ML people, which opened new horizons for me.

What do you think about the future prospects of blockchain systems? Will there be a killer application, like ever?

There are very few legal applications that check all the boxes for blockchain fit, such as completely decentralize management, sybil protection mechanism, etc. Public ledgers alone would be great if certain companies/governments would do that in a machine parseable way. The problem comes when you put all features together.

Most blockchains are useless. They are slow, much worse than centralized or semi-centralized. I don't need a participant in Russia to certify US stock exchange transactions. It is much better to give a bit o trust to institutions and get 1000s of transactions per second in exchange.

To be honest, there is so much bloody hype around blockchains... [Blockchain] is actually a bad enchilada. 

What advice you give to early career researchers.

It gets better. The first couple of years are hell. Teaching was very hard the first year. It took a lot of time. But you will get better at that in a couple years. Similarly for grants, and writing papers with students. 

What are pros and cons of academia vs industry?

In academia, you can do anything you want. When I am bored/unsatisfied, I think of something new. The only catch is you have to convince people to fund your research, but you can do it. I have been doing deep learning for example. I started as a networking person to go to high-performance data structures, because they are cool! What I hate in academia is chasing the funding. It is miserable, it is a huge waste of everyone's time. It is a crappy system.  Also we are underpaid compared to industry: my industry salary is 3x my academic salary, but I cannot complain about the academic salary either.

In industry, you learn a lot from your stuff getting used by internal/external customers.