794 karma · joined June 19, 2010
* HyperDex (http://hyperdex.org/)
* HyperGraphDB (http://www.hypergraphdb.org/)
* HyperSQL (http://hsqldb.org/)
* Hypertable (http://hypertable.org/)
https://docs.microsoft.com/en-us/sql/relational-databases/in...
* MemSQL (http://highscalability.com/blog/2016/9/7/code-generation-the...)
* Tableau/TUM HyPer (https://blog.acolyer.org/2016/05/23/efficiently-compiling-ef...)
* CMU Peloton (http://db.cs.cmu.edu/papers/2017/p1-menon.pdf)
* Vitesse (https://www.youtube.com/watch?v=PEmVuYjhQFo)
I think Greenplum was talking about doing this too, but that was about a year ago (http://engineering.pivotal.io/post/orca-profiling/)
Most systems just compile the predicates (Impala, SparkSQL).
A lot of companies are talking about adding this now. The performance gains are significant.
I disagree. I had a bad bowel obstruction once after minor surgery. I was backed up for over a week. My doctor told me to head straight to the local mall and get a Cinnabon. It flushed me out like it was opening the gates of the Erie Canal.
It was a miracle. The doctor later told me that he is 14 for 16 for getting people to start pooping again through Cinnabon.
https://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUE...
They bumped it up to 1000 from 255 a few years ago. It was a major engineering effort to support that. It's unlikely that they will be willing to do it again.
Say what? A lot of work is based on the assumption that P!=NP so you have to be clever in other ways.
You are talking about tenure-track or research professors. A teaching-track professor (typically) does not have PhD students so they do not have to get grants.
https://www.cmu.edu/policies/faculty/appointments-teaching-t...
Ignore this person, they don't know what they are talking about.
For better or worse, the 'elite' schools have money. That means they can hire teaching-track professors. These are professors whose sole job is to teach and advise (i.e., they don't spend all their time doing research). And they do a really good job at it. Much better than overworked professor that is more focused about research and grants.
Edit: People are conflating "teaching" versus "research" professors, so I want to provide an example:
https://en.wikipedia.org/wiki/CS50
This course is taught by David Malan. He has dedicated a lot of his life to this course to make it be awesome. The reason why he can do this is because he is a teaching professor. If you look at his CV, you will see that he does not have any PhD students. His main job is to teach undergrads at Harvard and he can put more of his time to it than a professor that has to also worry about research. He doesn't bring the university as much money because he's not writing research grants, but they can still afford to keep him around. This is the point that I was trying to make.
Cool. Do you support existing reporting tools and libraries? What about standard APIs for data discovery (i.e., INFORMATION_SCHEMA)? What about transactions? What about tools for examining log files for when your checkpoints get corrupted?
Why not? If you already support relational algebra, it seems like a non-brainer to just add SQL. Even if it's only SQL-92 you would be able to support some existing tools/ORMs almost for free.
Except that it's not. CockroachDB uses MVCC. Spanner uses 2PL and F1 uses OCC. Not the same at all.
The guy that invented the term "NewSQL" seems to disagree with you:
https://sigmodrecord.org/publications/sigmodRecord/1606/pdfs...
Are you talking about faculty or administration people?
I always try to reply to everyone who contacts me even if they aren't a student. I even respond to the people that email me theories that rely on P=NP.
CMU is working on this (seriously).
Please do.
No, it was not. FoundationDB's SQL layer was a joke. The system was horribly slow. Every DBA that I talked to that tried it said that it did not deserve the hype that it got.