April 16, 2014

MySQL Sizing questions

I frequently get questions along the lines of “how many transactions per second MySQL can do” or “how many servers I need to handle 100.000 users” or “which hardware would be enough to handle my 40GB” database. There are two things which are very interesting about these questions which make them funny and annoying at […]

Growth of Percona software releases

It was once said that “real artists ship.” In looking over the history of Percona software releases, we are currently shipping more software than ever before. First, let’s look at Percona Server. Let’s look at all major versions: 5.1, 5.5 and 5.6 as well as the total for each year. The estimate for 2013 comes […]

The relationship between Innodb Log checkpointing and dirty Buffer pool pages

This is a time-honored topic, and there’s no shortage of articles on the topic on this blog. I wanted to write a post trying to condense and clarify those posts, as it has taken me a while to really understand this relationship. Some basic facts Most of us know that writing into Innodb updates buffer […]

Distributed set processing performance analysis with ICE 3.5.2pl1 at 20 nodes.

Demonstrating distributed set processing performance Shard-Query + ICE scales very well up to at least 20 nodes This post is a detailed performance analysis of what I’ve coined “distributed set processing”. Please also read this post’s “sister post” which describes the distributed set processing technique. Also, remember that Percona can help you get up and […]

Distributed Set Processing with Shard-Query

Can Shard-Query scale to 20 nodes? Peter asked this question in comments to to my previous Shard-Query benchmark. Actually he asked if it could scale to 50, but testing 20 was all I could due to to EC2 and time limits. I think the results at 20 nodes are very useful to understand the performance: […]

InnoDB Flushing: a lot of memory and slow disk

You may have seen in the last couple of weekly news posts that Baron mentioned we are working on a new adaptive flushing algorithm in InnoDB. In fact, we already have three such algorithms in Percona Server (reflex, estimate, keep_average). Why do we need one more? Okay, first let me start by showing the current […]

Effect from innodb log block size 4096 bytes

In my post MySQL 5.5.8 and Percona Server: being adaptive I mentioned that I used innodb-log-block-size=4096 in Percona Server to get better throughput, but later Dimitri in his article MySQL Performance: Analyzing Percona’s TPCC-like Workload on MySQL 5.5 sounded doubt that it really makes sense. Here us quote from his article: “Question: what is a […]

On Good Instrumentation

In so many cases troubleshooting applications I keep thinking how much more efficient things could be going if only there would be a good instrumentation available. Most of applications out there have very little code to help understand what is going on and if it is there it is frequently looking at some metrics which […]

Why delayed flushing can result in less work

I can think of at least two major reasons why systems delay flushing changes to durable storage: 1. So they can do the work when it’s more convenient. 2. So they can do less work in total. Let’s look at how the second property can be true.

Can Innodb Read-Ahead reduce read performance ?

I ran into pretty interesting behavior today. We needed to dump and reload large database and we had pretty good IO subsystem so we started number of mysqldump processes in parallel. Unlike in other case when we did load in parallel, dump in parallel did not increase IO rate significantly and we could still see […]