MySQL Ecosystem – complementary talks at the conference?

Its times like this, I want to hear from the greater community – the ones that are reading say, Planet MySQL or Planet MariaDB.

MySQL to me, and many others is an ecosystem. We’ve had for the longest time, complementary technology talks, like for memcached (which have been popular, filled rooms). NoSQL is becoming quite popular, and there are complementary technologies sitting around. To get an idea, if terms like the following turn you on: Hadoop, Redis, Pig, NDB (yes, MySQL Cluster is largely NoSQL before NoSQL became popular), Tokyo Tyrant, StormCloud (formerly Waffle Grid).

Now, do you want to see these kinds of talks at the MySQL Conference & Expo 2010?

Check out the schedule grid. Its pretty healthy already ;)

Also, how interested are you in talks about PostgreSQL and MySQL in similar environments? What about replicating between PostgreSQL and Drizzle?

So a simple yes/no, would help. I should get this into a poll, clearly… maybe next time.


  1. I am more interested in Redis, Tokyo Tyrant, MongoDB etc than I am in Hadoop, Cassandra, or other “big” things. And actually I'm more interested in complementary-but-not-directly-competing technologies, such as these, than I am in Postgres or Drizzle, which overlap much more.

  2. Regardless of what I am interested in, talks about data servers other than MySQL will make the conference much better.

    Things that I am interested in:
    * talk that repeats the MySQL/Cassandra benchmark which shows 300ms to 350ms reads/writes for MySQL and 100X better performance for Cassandra. I will bring my PC XT with IDE hard disk to make it more likely that results can be repeated.
    * talk from Twitter with details on MySQL -> Cassandra migration
    * talk from MongoDB explaining why it is OK to not be crash safe. We can make it a joint talk with MyISAM.
    * talk from CouchDB explaining how it will overcome single-node performance limits (limited/no concurrency on writes, low peak QPS). Are people willing to pay the price in extra scale out to overcome this? Or is use of this limited to cases where other CouchDB features are needed and high QPS is not?
    * talk from someone really, really good explaining what NoSQL is. What are the criteria by which these systems can be judged? (crash safe, scale-out, single-node performance, hash indexes versus indexes that support range scans, how do you fake secondary indexes, backup, monitoring, …)

    I am interested in talks on the big things, but clearly, I am biased. It would be great to use MySQL row-based replication for a demo of MySQL -> !MySQL replication

    I am also interested in talk showing the use of MySQL/InnoDB for OLTP and something else for data warehouse queries with batch extraction (ugh) or replication between them. What is the something else? LucidDB? InfiniDB? InfoBright? PostgreSQL? Hive?

  3. My only fear is when these alternative talks come in they're all simple introductions, because that's all the time slot allows.

    I fear this the most with things like Cassandra, NDB and Hadoop. The presenter can't assume that everyone in the audience has done enough background reading before getting into the real 'meat of it'.

  4. Brian Aker says:


    How about :)


  5. colincharles says:

    We have a Belorussian translation of this, provided by Bohdan Zograf now –