Ideas, releases, practical guides, and perspectives from the people building with MySQL.
I have written a new post on the MySQL blog about the hypergraph optimizer, which is now available in MySQL 9.7 Community Edition.
The post gives a high-level technical overview of what is different from the classic join optimizer, why it can produce better plans for some multi-table queries, and where it is most useful to try. It also includes early benchmark results and some caveats, since the feature is still evolving and remains off by default.
On April 21, 2026, our third public discussion continued the conversation around transparency, participation, and the future of MySQL. Building on the momentum from earlier sessions, the discussion focused on progress and improvements to increase community transparency and practical ways for community members to get involved. At the center of the discussion was the MySQL […]
MySQL 9.7 Community Edition now includes the hypergraph optimizer as an alternative to the classic join optimizer, making this capability available across all MySQL editions. This is not a cosmetic change. The hypergraph optimizer uses a new join-planning framework aimed at queries where plan shape can make a real difference, particularly for multi-table joins, workloads […]
When vertical scaling, read replicas, and MEMORY tables weren't enough, Readyset's SQL-level caching broke through the ceiling, delivering over 109,000 queries per second on MySQL.
Key Results
109K
Queries Per Second
0
Lines of Code Changed
<1ms
Cached Query Latency
8x
Read Throughput Increase
About Lemit
Lemit (Lemit Tecnologia da Informacao) is a Brazilian data enrichment and credit intelligence company based in Sao Paulo. Their platform enables businesses to make faster, more accurate credit decisions by enriching, qualifying, and cleansing customer data at scale, powering individual lookups, batch integrations, and real-time API queries across millions of records with 100% all-flash storage.
"Using Readyset with our existing …
[Read more]When people ask me how to tune MySQL, they usually mean, “Which variables should I change in my.cnf?” (the MySQL configuration file) That certainly is a reasonable place to start, because most day-to-day performance tuning really is driven by MySQL itself: memory allocation, redo flushing, connection handling, temporary tables, and InnoDB I/O behavior. But the […]
As an update to our recent blog post, “Where can you find MySQL next? (Updated events list for March–May 2026)” we’d like to share the latest confirmed MySQL events and webinars, including newly added ones plus refreshed details for sessions we already announced. Since plans and agendas evolve quickly, this post is intended to provide […]
As part of our MySQL community engagement plan, which includes delivering new features into MySQL Community Edition, increasing collaboration and transparency, and expanding and growing the MySQL Ecosystem, we have created a new rhythm of public MySQL Community Discussions—starting with our first session on February 25 and continuing with our second session on March 23, […]
You ship a query on Monday. It runs in 2 ms. Six months later, the same query — same SQL, same indexes — takes 4 seconds. Nothing changed. Except one thing did: the table grew.
Welcome to the everyday problem Big O was invented to describe. This post is a practical, MySQL-flavoured tour of time complexity. We’ll keep the math light, the pictures clear, and we’ll finish with the exact algorithms your database uses to keep your queries fast (or slow).
What is Big O, really?
Forget seconds. Big O is not a unit of time. It’s a way of describing how the cost of an algorithm grows as the input grows. Specifically: if you double the data, does the work double? Stay the same? Quadruple? That shape is the algorithm’s complexity.
We call the input size n. A few shapes come up over
and over in real systems:
Production data is invaluable for day-to-day operations—support, troubleshooting, analytics, and development. But when that data contains sensitive fields such as SSNs, emails, phone numbers, or other identifiers, broad read access can quickly become unnecessary exposure. Just as importantly, many organizations operate under regulatory and contractual requirements that expect strong controls around access to sensitive data—often including data masking as […]
Production data is invaluable for day-to-day operations—support, troubleshooting, analytics, and development. But when that data contains sensitive fields such as SSNs, emails, phone numbers, or other identifiers, broad read access can quickly become unnecessary exposure. Just as importantly, many organizations operate under regulatory and contractual requirements that expect strong controls around access to sensitive data—often including data masking as […]