We have documented streaming replication, and we have also mentioned how you can use streaming replication, and we also once did a webinar with recording on it, and we wrote a very in-depth blog post on how to use streaming replication to manage large transactions with Galera Cluster.

However, we had a question recently about streaming replication around the usage of wsrep_trx_fragment_unit and wsrep_trx_fragment_size (also, well …

Just last week we had a Galera Cluster for MySQL streaming replication feature deep dive webinar, by the person that implemented it – Seppo Jaakola. In fact, if you haven’t already read the entire story, we encourage you to read: Galera Streaming Replication: Too Big A Bite Can Choke Your Cluster. We also have the 60 minute video recording for reference.

A huge number of Galera Cluster users have always complained about having issues with large transactions (where writes hang but reads are OK), and we are certain that the correct usage of streaming replication is what will ease that burden. What constitutes a large transaction? Typically when you make big number of …

A significant improvement has come to Galera Cluster 4 for MySQL, in the form of streaming replication. When you are dealing with large (greater than 2GB in size) or long running transactions, previous releases of Galera Cluster would reject the transaction; in the current release transactions can be split into fragmented chunks and those rows can be replicated in chunks to the other nodes even before commit. This is the feature that will remove you from all cluster stall issues due to large transaction processing!

While disabled by default, you can use it and configure it dynamically, and there is also the ability to monitor streaming replication progress. Come to this webinar to learn from the primary author, Seppo Jaakola, on why he implemented the feature, how you can use it, finding the optimal fragment size based on workload, and the design trade-offs for the initial release in where the focus was more on functionality over …

Percona XtraDB Cluster 8.0 comes with an upgraded Galera 4.0 library, which provides a new feature – streaming replication. Let’s review what it is and when it might be helpful.

Previous versions of Percona XtraDB Cluster with Galera 3.x had a limitation in how big transactions are handled.

Let’s review the performance under sysbench-tpcc workload when in parallel we update a big update on a table that is even non-related to the tables in the primary workload.

Without Streaming Replication

Let’s run two workloads.

1. sysbench-tpcc workload with 1 sec resolution
2. In parallel run UPDATE oltp.sbtest SET k=k+1 LIMIT 1000000

Running update:

mysql> update sbtest1 set k=k+1 limit 1000000;
Query OK, 1000000 rows affected (34.48 sec)
Rows matched: 1000000  Changed: …

I was testing the latest Percona XtraDB Cluster 8.0 (PXC) release which has the Galera 4 plugin, and I would like to share my experiences and thoughts on the Streaming Replication feature so far.

What Is Streaming Replication, in One Sentence?

In Galera 4, the large transaction could split into smaller fragments, and even before it got committed these fragments have been replicated to the other nodes and have already started the certification and apply process.

The manual describes all the pros and cons, but let’s see how it works. I have created a table with 10M rows and I am going to run some large updates on that.

First I have run the updates without Streaming Replication, and because it is disabled by …