As part of work on “High Performance MySQL, 3rd edition”, Baron asked me to compare different MySQL version in some simple benchmark, but on decent hardware.

So why not.

I took our Cisco UCS C250 and ran simple sysbench oltp read-write all data fits into memory workload.

Versions in question:

• MySQL 4.1
• MySQL 5.0
• MySQL 5.1 (with built-in InnoDB)
• MySQL 5.1 with InnoDB-plugin
• MySQL 5.5
• MySQL 5.6

All versions are vanilla MySQL, not Percona Server.

The results are there:

There is …

I recently ran into an issue where our request throughput was showing very erratic and spikey behavior despite very smooth response times from the application servers. Using Splunk, we analyzed every log that we had: nginx, haproxy, apache, and the application logs themselves and we were seeing similarly spikey throughput. Because those tools all log upon request completion, there was no way to determine from the logs themselves whether it was one tier of the stack in particular that was delaying request arrival, or if it the spikes were endemic to the traffic we were receiving.

So, we decided to perform some analysis of the raw tcp data on the edge server using a couple of tools. First, was tcpdump, which is a tool that should be in every SysAdmin’s arsenal.

First, grab all the traffic on the interface and write it to a pcap formatted file:

# tcpdump -c 200000 -w output.pcap -i any

This command will capture …

I was looking into the history about why SYSDATE() and NOW() behave differently in MySQL, and it looks like in 4.0 and 4.1 they used to be the same. But as of MySQL 5.0, SYSDATE() was changed to emulate Oracle’s behavior, that is, it returns the time as of the function execution, not as of the statement start. There are a number of bug reports related to this: 15101, 12480, 12481, and 12562.

Sitting on a shelf in my garage is a reminder that Steve Jobs' product launches didn't always go the way he hoped for. It's my NeXT station and my NeXT laser printer. The printer has a perfectly straight paper path; it can print on cardboard. As for the computer, I learned a few years ago in NERT class that if the house were ever on fire, the firemen would probably want to be alerted about its magnesium case.


Continue reading "The NeXT"

The MySQL Utilities project continues to evolve with key new features for
replication and export. The latest release, 1.0.3, is no exception.

MySQL Utilities is included in the MySQL Workbench product which can be
downloaded from HTTP://dev.mysql.com/downloads/workbench/5.2.html

If you want the latest developments for MySQL Utilities, you can create a
bazaar branch using the following command:

bzr branch lp:~mysql/mysql-utilities/trunk

New Utility - mysqlrplshow

You can now view a list of the slaves attached to your master with
mysqlrplshow. The utility displays a graph of the master and its slaves y
default but you can also get a list of the slaves in GRID, CSV, TAB, or
VERTICAL format as follows.

• GRID - Displays output formatted like that of the mysql monitor in a grid or table …

Marketo, Inc. is the revenue performance management company, transforming how marketing and sales teams of all sizes work - and work together - to accelerate predictable revenue. Marketo faced the same data management problems as many 24x7 SaaS and online businesses that are build on MySQL: quick recovery from failures, seamless maintenance, and efficient hardware utilization. In this video

SkySQL today announced rapid growth and success in its first year of operations, fueled by new customers, partnerships and corporate growth. Highlights of the company's first year include:

• Signing more than 160 customers in its 12 months of operation

• Attracting the attention of leading technology and distribution partners

• Receiving 900,000 Euros from Tekes to fund R&D

“The first year can often make or break a young company,” said Ralf Wahlsten, chairman and investor in SkySQL. “Establishing a name for a startup is often a giant hurdle, especially in this economic climate. But it is one that SkySQL has surpassed, as evidenced by the number of customers, partners, and talented employees that have been attracted to the company. SkySQL’s rapid growth validates the value and quality of expertise that can only be delivered by a company with an innate understanding of what customers want from their …

For those people that weren’t able to attend the MySQL Cluster demo or sessions at this year’s Oracle Open World (or even for those that did) and would like copies of the material, links are provided here.

• Building Scalable & Highly Available Database Services with MySQL Cluster – includes a description of the features in the new MySQL Cluster7.2  Development Milestone Release 2
• NoSQL Access to MySQL: The Best of Both Worlds – Covers what people are looking for from NoSQL data stores and how MySQL Cluster can meet these needs (both in terms of database characteristics and the APIs offered to …

Last week I helped 2 different customers with infinite replication loops. I decided to write a blog post about these infinite loop of binary log statements in MySQL Replication. To explain what they are, how to identify them… and how to fix them.

An infinite replication loop is one or more transactions that are replicating over and over in a Multi Master Replication relationship. This happens because those transactions belong to a server-id which is different from the two actual masters.

Let’ s have a look first at the normal flow:

Replication flow: [1]

When you setup MySQL replication, you define the global variable server_id, this value must be unique for each host. In the illustration above, arrows represent the replication flow, so 1 is replicating to 2 and to 3, and 2 is replicating …

Some limitations of partitioning in MySQL 5.1 on one page.

• All columns used in the partitioning expression for a partitioned table must be part of every unique key that the table may have.These rules also apply to existing nonpartitioned tables that you wish to partition.
• Partitioning applies to all data and indexes of a table; you cannot partition only the data and not the indexes, or vice versa, nor can you partition only a portion of the table.
• Prior to MySQL 5.1.43, it was not possible to partition a table by RANGE based on the value of a TIMESTAMP column by using the UNIX_TIMESTAMP() function.
• When partitioning by LIST, you can match variable against only a list of integers values and some time NULL but not with char values. Never tried partition with string, it will not work.
• If you explicitly define any subpartitions using SUBPARTITION option on any partition of a …