Miles to go …

I presented on Creating Quick and Powerful Web Applications with MySQL, GlassFish, and NetBeans. The key messages conveyed during the preso are:

• GlassFish is an open source community and delivers production-quality Java EE compliant Application Server.
• GlassFish v2 is the Java EE 5 Reference Implementation and GlassFish v3 for Java EE 6. Read complete difference here.
• Java Persistence API makes it really easy to create database-backed Web applications. It even creates MySQL-specific queries, when possible.
• The web-based administration console and CLI are powerful GlassFish management tools that meets the need of any IT administrator.
• NetBeans provides comprehensive and seamlessly integrated tooling for GlassFish. The goal is to make the Eclipse tooling at par with NetBeans.

The slides are available here.

And then notes from some of the sessions I attended:

State of the Dolphin

• 12+ million users, 70k downloads/day, 1100 MySQL Partners
• Multiple platforms: LAMP, Windows, Mac, OpenSolaris, Solaris, RedHat, Suse, Ubuntu
• Multiple Languages: php, Perl, Python, Ruby, Java, C, C++, C#
• MySQL 5.1: 3 million downloads in 100 days
• MySQL 5.4 announced: InnoDB Scalalbility, Sub-query optimizations, 59% faster than 5.1, 40% improvement in read/write test, 71% throughput increase
• InnoDB: Fast index creation (add/drop indexes w/o copying the data), Data compression (shrink tables, to significantly reduce storage and i/o)
• Embedded InnoDB (announced today): Proven high-performance and reliability and functionality of InnoDB, low-level but powerful non-SQL API for app programmers, operational characteristics needed for stand-alone apps where there is no DBA
• Dr DBA was awarded “Acquirer of the Year: Oracle”
• MySQL Cluster 7.0: 99.999% availability, 4.3x higher throughput, 140k+ TPM and 4x less power and consumption than 6.3
• MySQL Query Analyzer: Continuous query monitoring, find and fix problem SQL code, historical and real-time analysis, drill down into execution statistics

InnoDB: Innovative Technologies for Performance and Data Protection

• Dr Heikki Tuuri, was professor at Helsinki, founded Innobase, got acquired by Oracle
• Performance and Data Integrity are basic features
• Architected and written by one person
• Full transaction support, Unlimited row-level locking, multi-version read-consistency, automatic deadlock detection
• Innovative: adaptive hash indexes, insert buffer (performance benefits), doublewrite buffer, InnoDB plugin
• Oracle/Innobase + Sun/MySQL

Rethinking MySQL, Enter Drizzle

• Goals
• Pluggable/Infrastructure Aware
• Community Developed
• Multicore/Concurrency (load up 10,000 connections in db)
• Focus on Web applications/enable others
• Modernize codebase for manageability (currently C/C++, can we reuse STL and other libraries)
• Philosophies
• Have open and well-documented interfaces
• Have transparent goals and processes, that are communicated publicly
• Have fun and encourage collaboration
• Remove barriers to contribution and participation for everyone
• Enable contributors to build a business around Drizzle
• Drizzle announced at OSCON last year
• Translated into 30+ languages since then
• 7% of developers are from Sun
• 100+ contributors (>500 on the mailing list), even Postgres and Firebird developers \
• Cirrus available now, Aloha next
• Drizzle Developer Day 2009 scheduled this Friday
• No patches are contributed back to MySQL Enterprise
• Will be ready for production deployment Jun 2010
• References
• launchpad.net/drizzle
• IRC #drizzle
• Email

High Performance Rails and MySQL

• David Berube: Apress books on “Author Practical Ruby Gems”, “Practical Rails Plugins”, “Practical Reporting with Ruby on Rails”
• Finding performance issues in Rails
• Rails development log
• eabe_db_tools: Ajax popup- displays query count, query each time for a each query on a page. Will be available on github next week.
• mysql_slow_log
• Is it a database problem: Firebug, YSlow, Ping, tracert, etc.
• Let the database do the heavy lifting instead of Ruby: for example, don’t sort in Ruby
• Deep eager loading: don’t load that is not required
• Use built-int Rails grouping and aggregate functions
• Caching: simple ootb caching, Cache Fu, MySQL triggers for DB function caching, Rails triggers for other caching

Did you know 1.3 billion emails were sent as part of Obama’s election campaign – and all powered by MySQL ? Hear the details from Blue State Digital engineers who created the solution and maintained it:

And you can always read the complete case study.

Some pictures from earlier today …

And then the evolving picture album is available at:

Come meet us at the GlassFish booth in the Exhibit Floor. Or you can stop by at room #205 for the Whisper Suite for a more personal and 1-1 conversation.

Technorati: conf mysqlconf mysql santaclara glassfish netbeans

1. GlassFish and NetBeans at MySQL Users Conference 2009
2. MySQL Users Conference 2009 Day 3 – Cloud Shootout
3. MySQL Users Conference 2009 Day 1 in Photos
4. Offshore monitoring of windfarms using GlassFish – MySQL Users Conference 2009 Day 3
5. TOTD #121: JDBC resource for MySQL and Oracle sample database in GlassFish v3