Here is the quick notes from the session Falcon from the beginning by Jim Starkey and Ann Harrison

• Why Falcon
  • Hardware is evolving rapidly, world is changing, so taking advantage
  • Customers need ACID transactions
• Where hardware is going
  • CPUS breed like rabbits (more sockets, cores, threads/core)
  • Memory is bigger, faster and cheaper
  • Disks are bigger and cheaper but not much faster
  • In general boxes are getting cheaper
• Where applications are going
  • batch - dead
  • timesharing - dead
  • departmental computing - dead
  • client server - fading fast
  • application servers for most …

Here is the quick notes from the session Scaling MySQL - Up or Out ? moderated by Kaj Arno as part of the todays keynote.

Here is the list of panelists are ordered by Alexa ranking.

1. Monty Taylor (MySQL)
2. Matt Ingerenthron (Sun)
3. John Allspaw (Flickr)
4. Farhan Mashraqi (Fotolog)
5. Domas Mituzas (Wkipedia)
6. Jeff Rotheschild (Facebook)
7. Paul Tuckfield (YouTube)

…

Ok, this blog is currently hosted by 1and1.com, and I think that needs to change, which is sad, because up until recently, I’ve been pretty happy with the performance. However, I recently had an issue, and some things came to light about my package that I wasn’t aware of.

First of all, the maximum time a process can run is about 6 seconds. Second of all, the maximum number of processes you can have running at once is 12. When I asked if this was a limitation of my *package*, the answer I got was that it was a limitation of all shared hosting accounts.

This blog started throwing 500 errors some time yesterday. I called support late last night and some lady said that she was running a tool that should fix my issue, and to call back in 30 minutes if the problem persists. Well, I’m sick as a dog, and it was late, so I went to bed. This morning, the problem was still there. I called support 3 times today, got cut off twice, and …

Here is the notes from “InnoDB status, architecture and new features” by Heikki Tuuri and Ken Jocobs.

• Introduction
  • Announced Plugin 1.0 for MySQL 5.1
  • fast, reliable, proven
  • fast
    • row-level locking, MVCC -> high concurrency & throughput
    • high performance CPU, memory and I/O architecture
    • efficient indexing (covering)
  • reliable
    • automatic crash recovery
    • integrated referencial integrity and transactions
    • online backup
    • well written, well tested and large user community
  …

Here is the notes from MySQL online backup in 6.0 session by Dr. Charles A. Bell, the Author of MySQL Expert book.

• Introduction
  • SQL Driven, run from any client
  • backup to localhost
  • new security privileges for backup/restore
  • blocking restore
  • non-blocking backup for Innodb, Falcon etc
• Overview
  • protection and restore
  • basic functionality of backup and restore available now
  • db level -> table, views, sps, functions, … are supported
  • 6.0.5 has it already, try it
  • source code is another way to look into it
• Design
  • ent level consistency between engines
  • default driver …

Here is the quick notes from MySQL Cluster by Stewart Smith session that I attended today in the MySQL conference 2008.

• What it is
  • Clustering of in memory databases in a shared-nothing system
  • Designed for HA, 99.9% Uptime (not really) and sub-second failover
  • Supports Hot (online) consistent backup along with compression
  • No locks are used during the backup as NDB uses a global counters
• Think as virtual file system
• Redundancy
  • NoOfReplicas (1,2,3,4), For production use 1 or 2 but don’t use any other as it may have bugs (2 is preferred and well tested mode)
  • 1 means no redundancy (a node fails and cluster fails)
  • 2 means two copies …

Today I just happen to read May 2008 edition of Dr.Dobb’s Journal printed edition and the article “Kernel-mode Databases” written by Andrel and Alexander from McObject really excited me.

The basic principle of any database engine is its performance and scalability and success of any database in most part relays on these two concepts. To achieve this, operating system kernel plays a major role in the form of …

• resource allocation
• thread scheduling
• low-level hardware access
• network
• security

So, looks like eXtremeDB try to overcome from this by releasing kernel-mode database. The kernel mode package actually overrides most of the kernel access calls that database engine expects by replacing with optimized …

After Microsoft released its beta version of “SQL Server Data Services“; I was more interested to know what kind of considerations were taken care in the design and architecture of SSDS in order to get performance, distribution of data and scalability as it is developed on top of SQL Server.

To my surprise, I noticed Microsoft is conducting the live event to answer some of these questions by following up Ryan Dunn blog; Today I also got the confirmation to attend the event and I look forward to get some of my general questions answered.

It will be interesting to see the pricing model of SSDS in the coming days, and might even attract if they make it FREE for initial customers and/or by setting the limit on the data usage by …

It is interesting to read RedmondDeveloper News’s take on Oracle’s attitude to open source this morning, especially this paragraph quoting Monica Kumar, Oracle’s senior director for Linux and open source product marketing:

“”We haven’t seen our customers asking for open source databases,” she told me. “Not many customers are interested in looking into the code and mucking around with it, and making changes to it. All they care about is ‘give me the best support, give me the lowest price of entry’.” For that Kumar pointed to Oracle Express.”

It is difficult to disagree with the second part of Monica’s statement. Cost savings are routinely cited as the biggest driver for open source database adoption, while the lack of robust support is the biggest barrier to open source adoption.

Certainly these were the findings of our …

I'd been doing some stress testing of my mysql application today, and I was hitting some weird cases. Several transactions were deadlocking - this was expected - but the number of records that got inserted into my table was more than the number that I expected after subtracting errors.

My test was fairly simple:

1. Fork 15 processes
2. Insert and update 100 records in each process, running each INSERT/UPDATE pair inside one transaction
3. ROLLBACK on error

Either the INSERT or the UPDATE was expected to fail due to deadlock, and the whole transaction should have rolled back leaving no record in the table.

Before I go on, I should mention that I was using InnoDB, which does support transactions.

What I expected was that the total number of records in the table + the total number of INSERT/UPDATE aborts due to deadlock should be equal to 1500 (15*100). What …