Insights

Managed Databases

UpCloud Managed Database is a database-as-a-service offering a carefree option for running well-known open-source databases in the cloud. It is deployable via the users’ UpCloud Control Panel with configuration options including the choice of database management systems, clustered or single-node systems, and a number of resource plans. Managed Databases are available in every UpCloud data centre location.

Database plans

Managed Databases are offered with 2 and 3-node clustered solutions as well as single-node unclustered options. The nodes can be configured with the required resources by selecting from the available plans. Each plan includes a set number of CPU cores, system memory, storage and backups. The Managed Database plans are priced according to the number of nodes and the resources allocated to each node.

1 node

MemoryCPU coresStorageBackupsPrice
2 GB125 GB24 hours$0.042/h$30/mo
4 GB250 GB24 hours$0.083/h$60/mo
4 GB2100 GB24 hours$0.104/h$75/mo

2 nodes

MemoryCPU coresStorageBackupsPrice
4 GB250 GB7 days$0.139/h$100/mo
4 GB2100 GB7 days$0.208/h$150/mo
8 GB450 GB7 days$0.243/h$175/mo
8 GB4100 GB7 days$0.278/h$200/mo
16 GB6100 GB7 days$0.556/h$400/mo
16 GB6250 GB7 days$0.778/h$560/mo
32 GB8100 GB7 days$1.042/h$750/mo
32 GB8250 GB7 days$1.111/h$800/mo
32 GB8500 GB7 days$1.222/h$880/mo
64 GB161 TB7 days$2.222/h$1600/mo
64 GB161.5 TB7 days$2.500/h$1800/mo
128 GB202 TB7 days$3.889/h$2800/mo

3 nodes

MemoryCPU coresStorageBackupsPrice
4 GB2100 GB14 days$0.243/h$175/mo
4 GB2200 GB14 days$0.313/h$225/mo
8 GB4100 GB14 days$0.389/h$280/mo
8 GB4200 GB14 days$0.451/h$325/mo
16 GB6200 GB14 days$0.833/h$600/mo
16 GB6500 GB14 days$0.903/h$650/mo
32 GB8200 GB14 days$1.556/h$1120/mo
32 GB8500 GB14 days$1.694/h$1220/mo
32 GB81 TB14 days$1.972/h$1420/mo
64 GB161 TB14 days$3.111/h$2240/mo
64 GB162 TB14 days$3.667/h$2640/mo
128 GB202 TB14 days$5.528/h$3980/mo

Note that the database management system and host operating system use a small portion of the allocated storage capacity, typically 1 – 2 GB.

Available databases

Currently, available database systems include the following.

MySQL

MySQL is the world’s most popular enterprise-grade open-source relational database management system. Managed Databases utilise MySQL Community Edition which is supported by a huge and active community of open source developers.

Customisable database properties

The following MySQL database properties can be set by the user via the UpCloud API.

admin_passwordinnodb_lock_wait_timeoutnet_read_timeout
admin_usernameinnodb_log_buffer_sizenet_write_timeout
automatic_utility_network_ip_filterinnodb_online_alter_log_max_sizepublic_access
backup_hourinnodb_print_all_deadlocksslow_query_log
backup_minuteinnodb_rollback_on_timeoutsort_buffer_size
binlog_retention_periodinteractive_timeoutsql_mode
connect_timeoutinternal_tmp_mem_storage_enginesql_require_primary_key
default_time_zoneip_filtertmp_table_size
group_concat_max_lenlong_query_timeversion
information_schema_stats_expirymax_allowed_packetwait_timeout
innodb_ft_min_token_sizemax_heap_table_size
innodb_ft_server_stopword_tablemigration

PostgreSQL

PostgreSQL, also known as Postgres, is a free and open-source relational database management system emphasizing extensibility and SQL compliance.

Customisable database properties

The following PostgreSLQ database properties can be set by the user via the UpCloud API.

admin_passwordip_filterpg_partman_bgw.role
admin_usernamejitpg_read_replica
automatic_utility_network_ip_filterlog_autovacuum_min_durationpg_service_to_fork_from
autovacuum_analyze_scale_factorlog_error_verbositypg_stat_statements.track
autovacuum_analyze_thresholdlog_line_prefixpgbouncer
autovacuum_freeze_max_agelog_min_duration_statementpglookout
autovacuum_max_workersmax_files_per_processpublic_access
autovacuum_naptimemax_locks_per_transactionshared_buffers_percentage
autovacuum_vacuum_cost_delaymax_logical_replication_workerssynchronous_replication
autovacuum_vacuum_cost_limitmax_parallel_workerstemp_file_limit
autovacuum_vacuum_scale_factormax_parallel_workers_per_gathertimescaledb
autovacuum_vacuum_thresholdmax_pred_locks_per_transactiontimezone
backup_hourmax_prepared_transactionstrack_activity_query_size
backup_minutemax_replication_slotstrack_commit_timestamp
bgwriter_delaymax_slot_wal_keep_sizetrack_functions
bgwriter_flush_aftermax_stack_depthtrack_io_timing
bgwriter_lru_maxpagesmax_standby_archive_delayvariant
bgwriter_lru_multipliermax_standby_streaming_delayversion
deadlock_timeoutmax_wal_senderswal_sender_timeout
default_toast_compressionmax_worker_processeswal_writer_delay
enable_ipv6migrationwork_mem
idle_in_transaction_session_timeoutpg_partman_bgw.interval

Supported extensions

Managed Database using PostgreSQL supports the following extensions.

address_standardizerhstorepgroutingpostgres_fdw
address_standardizer_data_usintaggpgrowlockspromscale
aiven_extrasintarraypgstattuplerum
bloomisnplcoffeesslinfo
btree_ginltreepllstablefunc
btree_gistpg_buffercacheplperltimescale_analytics
chkpasspg_cronplpgsqltimescaledb
citextpg_partmanplv8tsearch2
cubepg_prometheuspostgistsm_system_rows
dblinkpg_repackpostgis_legacyunaccent
dict_intpg_similaritypostgis_rasterunit
earthdistancepg_stat_statementspostgis_sfcgaluuid-ossp
fuzzystrmatchpg_trgmpostgis_tiger_geocoder
hllpgcryptopostgis_topology

Database features

High availability

In clustered database deployments, nodes are created on physically separated backend hosts to ensure redundancy in case of hardware failure.

In normal operation, the DNS entry of the database points to a primary node. One of the replicas will be promoted automatically in case of a primary failure.

Scalability

Users are able to scale their Managed Database services both vertically and horizontally by increasing or decreasing the number of nodes in the cluster or the resources allocated to each node.

Changes to the Managed Database configuration are handled seamlessly in the background with minimal interruption to the services.

Connectivity

Managed Databases can be reached via the user’s private UpCloud Utility network or optionally through the public network connection. Private connection is enabled by default. This makes the database accessible for all the user’s Cloud Servers within the same data centre as the Managed Database.

At the configuration settings, users can set a preferred hostname. The hostname is used to automatically create DNS entries using a unique subdomain name. Each Managed Database can have up to four DNS entries, one for private and another for public access, as well as read-only access for each.

Private: example-mctapmghasah.db.upclouddatabases.com Private read-only: replica-example-mctapmghasah.db.upclouddatabases.com
Public: public.example-mctapmghasah.db.upclouddatabases.com Public read-only: public-replica-example-mctapmghasah.db.upclouddatabases.com

Maximum connections

To ensure a consistent experience, concurrent database connections are limited based on the node resources.

MemoryMySQLPostgreSQL
2 GB15025
4 GB400100
8 GB800200
16 GB1600400
32 GB3200800

Data metering

Both inbound and outbound network transfer is unmetered by the Managed Database. The database plans do not include transfer quota nor will the Managed Databases benefit from the user’s Network Transfer Pool. However, any network transfer via the public network is metered normally user’s Cloud Servers utilising the Network Transfer Pool.

Data transfer via the Utility network is always free of charge.

IP access control

Users can set connection permissions to specific IP addresses by selecting from the list of Cloud Servers on the user’s account or manually entering IP addresses.

By default, all Utility network IP addresses are allowed. Setting a permitted IP address will exclude connection from all other addresses unless “allow all” is enabled for either public or Utility networks.

Maintenance

Users can set their preferred maintenance windows at their UpCloud Control Panel.

Normal version upgrades are performed in place. Users will be provided with the ability to opt-in for any major upgrades.

Backups

Each Managed Database plan includes automated backups. Backup retention is set to either 1, 7 or 14 days for 1, 2, and 3 node clusters respectively.

The databases are backed up automatically with full backups daily and the write-ahead log (WAL) is recorded continuously. Full backups are version-specific binary backups, which when combined with WAL allow consistent recovery to a point in time (PITR).

All backups are encrypted and stored off-site independently from the user’s Managed Database without using the cluster’s storage capacity.

The 1-day backups allow the user to go back to any given minute within the last 24 hours. 2-node plans offer 7 days of backups and 3-node plans include 14 days.

Back to top