InnoDB Storage Engine.  Data must be stored in the InnoDB transactional storage engine. Transactions are executed optimistically and then, at commit time, are checked for conflicts. If there are conflicts, in order to maintain consistency across the group, some transactions are rolled back. This means that a transactional storage engine is required. Moreover, InnoDB provides some additional functionality that enables better management and handling of conflicts when operating together with Group Replication. The use of other storage engines, including the temporary MEMORY storage engine, might cause errors in Group Replication. You can prevent the use of other storage engines by setting the disabled_storage_engines system variable on group members, for example:

disabled_storage_engines="MyISAM,BLACKHOLE,FEDERATED,ARCHIVE,MEMORY"

Primary Keys.  Every table that is to be replicated by the group must have a defined primary key, or primary key equivalent where the equivalent is a non-null unique key. Such keys are required as a unique identifier for every row within a table, enabling the system to determine which transactions conflict by identifying exactly which rows each transaction has modified.

Network Performance.  MySQL Group Replication is designed to be deployed in a cluster environment where server instances are very close to each other. The performance and stabiity of a group can be impacted by both network latency and network bandwidth. Bi-directional communication must be maintained at all times between all group members. If either inbound or outbound communication is blocked for a server instance (for example, by a firewall, or by connectivity issues), the member cannot function in the group, and the group members (including the member with issues) might not be able to report the correct member status for the affected server instance.

From MySQL 8.0.14, you can use an IPv4 or IPv6 network infrastructure, or a mix of the two, for TCP communication between remote Group Replication servers. There is also nothing preventing Group Replication from operating over a virtual private network (VPN).

Also from MySQL 8.0.14, where Group Replication server instances are co-located and share a local group communication engine (XCom) instance, a dedicated input channel with lower overhead is used for communication where possible instead of the TCP socket. For certain Group Replication tasks that require communication between remote XCom instances, such as joining a group, the TCP network is still used, so network performance influences the group's performance.

Binary Log Active.  Set --log-bin[=log_file_name]. MySQL Group Replication replicates binary log contents, therefore the binary log needs to be on for it to operate. This option is enabled by default. See Section 5.4.4, “The Binary Log”.

Slave Updates Logged.  Set --log-slave-updates. Servers need to log binary logs that are applied through the replication applier. Servers in the group need to log all transactions that they receive and apply from the group. This is required because recovery is conducted by relying on binary logs form participants in the group. Therefore, copies of each transaction need to exist on every server, even for those transactions that were not initiated on the server itself. This option is enabled by default.

Binary Log Row Format.  Set --binlog-format=row. Group Replication relies on row-based replication format to propagate changes consistently among the servers in the group. It relies on row-based infrastructure to be able to extract the necessary information to detect conflicts among transactions that execute concurrently in different servers in the group. See Section 17.2.1, “Replication Formats”.

Binary Log Checksums Off.  Set --binlog-checksum=NONE. Due to a design limitation of replication event checksums, Group Replication cannot make use of them, and they must be disabled.

Global Transaction Identifiers On.  Set --gtid-mode=ON. Group Replication uses global transaction identifiers to track exactly which transactions have been committed on every server instance and thus be able to infer which servers have executed transactions that could conflict with already committed transactions elsewhere. In other words, explicit transaction identifiers are a fundamental part of the framework to be able to determine which transactions may conflict. See Section 17.1.3, “Replication with Global Transaction Identifiers”.

Replication Information Repositories.  Set --master-info-repository=TABLE and --relay-log-info-repository=TABLE. The replication applier needs to have the master information and relay log metadata written to the mysql.slave_master_info and mysql.slave_relay_log_info system tables. This ensures the Group Replication plugin has consistent recoverability and transactional management of the replication metadata. From MySQL 8.0.2, these options are set to TABLE by default, and from MySQL 8.0.3, the FILE setting is deprecated. See Section 17.2.4.2, “Slave Status Logs”.

Transaction Write Set Extraction.  Set --transaction-write-set-extraction=XXHASH64 so that while collecting rows to log them to the binary log, the server collects the write set as well. The write set is based on the primary keys of each row and is a simplified and compact view of a tag that uniquely identifies the row that was changed. This tag is then used for detecting conflicts. This option is enabled by default.

Multithreaded Appliers.  Group Replication members can be configured as multithreaded slaves, enabling transactions to be applied in parallel. A nonzero value for slave_parallel_workers enables the multithreaded applier on the member, and up to 1024 parallel applier threads can be specified. Setting slave_preserve_commit_order=1 ensures that the final commit of parallel transactions is in the same order as the original transactions, as required for Group Replication, which relies on consistency mechanisms built around the guarantee that all participating members receive and apply committed transaction in the same order. Finally, the setting slave_parallel_type=LOGICAL_CLOCK, which specifies the policy used to decide which transactions are allowed to execute in parallel on the slave, is required with slave_preserve_commit_order=1. Setting slave_parallel_workers=0 disables parallel execution and gives the slave a single applier thread and no coordinator thread. With that setting, the slave_parallel_type and slave_preserve_commit_order options have no effect and are ignored.