Latest Database Releases: Key Features of Oracle, MySQL, PostgreSQL, OceanBase & More
This newsletter compiles the most important updates from June 2021 across RDBMS, NoSQL, NewSQL, time‑series, big‑data, domestic and cloud databases, highlighting new versions, major features, performance improvements and security enhancements for products such as Oracle 21c, MySQL 8.0.24/8.0.25, PostgreSQL 14 Beta, TiDB 5.0, OceanBase V3.1 and many others.
DB‑Engines Ranking Overview
Based on June 2021 data, the DB‑Engines ranking combines five factors—search engine keyword volume, Google Trends, job postings on Indeed, LinkedIn profile mentions, and StackOverflow activity—to rank database products.
RDBMS Updates
Oracle 21c
Immutable blockchain table: provides append‑only, tamper‑detectable rows with built‑in change detection.
Native JSON data type: speeds up JSON scans 10× and updates 4×, supports indexing and parallel SQL.
In‑database AutoML: automated model building and comparison, with new anomaly detection, regression and deep‑learning algorithms.
Embedded JavaScript (GraalVM): allows JavaScript code to run inside the database and call SQL directly.
Persistent memory support: stores data and redo logs in local PMEM for lower I/O latency.
High‑performance graph model: optimised memory usage and native Java‑based graph algorithms.
Automated column‑store memory management: auto‑compresses and re‑balances columnar data.
Sharding automation: Sharding Advisor evaluates schema and workload to suggest optimal designs.
MySQL 8.0.24 & 8.0.25
Enterprise Audit now supports log trimming.
Improved error messages for client‑initiated connection closures.
Added port information to client‑side failure messages.
Keyring functionality moved from plugin to server component.
Upgraded bundled curl to 7.74.0.
Performance Schema adds new metrics such as memory/sql/dd::infrastructure and memory/sql/dd::object.
Added system variable to configure caching_sha2_password hash iterations.
New spatial functions: ST_LineInterpolatePoint(), ST_LineInterpolatePoints(), ST_PointAtDistance(), ST_Collect().
InnoDB AUTOEXTEND_SIZE increased from 64 MiB to 4 GiB. clone_donor_timeout_after_network_failure timeout extended to up to 30 minutes.
MGR allowlist can be updated without restarting the group. --skip-slave-start now starts a replica without requiring a login to the source host.
PostgreSQL 14 Beta 1
Performance: high‑concurrency connection handling, improved btree index scaling, and faster GiST/SP‑GiST index creation.
Parallel execution enhancements for sequential scans, PL/pgSQL, and materialised view refreshes.
Extended sharding support via postgres_fdw, including async external table scans and bulk insert.
Partition‑pruning improvements and concurrent ALTER PARTITION without locking.
Window function speed‑ups with incremental sort.
Extended statistics for better selectivity estimation, including multi‑expression stats.
Large‑IN list optimisation using hash table probing.
TOAST now supports LZ4 compression.
Libpq driver adds pipeline mode, session‑level target session attributes, and standby detection.
Security: new roles pg_read_all_data and pg_write_all_data, default SCRAM‑SHA‑256 authentication, and client‑cert verification options.
NoSQL Update
RocksDB 6.19.x & 6.20.x
BackupEngine can open a backup as a read‑only DB.
BlobDB integration with BackupEngine, allowing shared blob files across backups.
Optional output parameter for BackupEngine::CreateNewBackup returns the new backup ID.
New GetBackupInfo / GetLatestBackupInfo APIs for single‑backup queries.
Long‑term support for feature‑filter blocks (compatible with SST ≥ 6.15.0).
BlobDB now supports compression filters via FilterBlobByKey().
BackupEngine can include file names and sizes in backup metadata.
NewSQL Updates
TiDB 5.0
Introduced MPP architecture with TiFlash nodes, achieving 2–3× performance over Greenplum and Spark.
Clustered index support improves OLTP throughput (e.g., TPC‑C up 39%).
Asynchronous commit reduces write latency by ~42% in Sysbench tests.
Stabilised optimizer and limited resource consumption for smoother operation.
Raft Joint Consensus ensures high availability during region changes.
Enhanced EXPLAIN and invisible indexes aid DBA debugging.
Backup/restore integration with Amazon S3 and Google Cloud GCS.
Improved import/export performance for large datasets (e.g., 40% faster TiB‑scale loads).
SequoiaDB 3.4.3 & 5.0.2 – Four Product Lines
SQL engine adds instance‑group functionality and preferedinstance parameter for read routing.
MongoDB‑compatible findAndModify and bulkWrite support.
Data source feature enables cross‑cluster access.
Performance: optimised MySQL index queries and multi‑statement inserts.
Concurrency replay and partition‑key sorting improvements.
Four product lines (DP data platform, TP transaction DB, CM content DB, DOC document DB) launched on a lake‑warehouse architecture.
Time‑Series Database
MatrixDB 4.0
Mars engine: columnar storage, data encoding/compression, push‑down aggregation, and index‑free scans.
Multi‑node automated deployment in under five minutes.
Cold‑hot tiered storage UDF for easy partition management and data migration.
New time_bucket_gapfill function with locf and interpolate fill‑methods.
Big Data Ecosystem
Elastic major releases (7.11.x, 7.12.x, 7.13.x, upcoming 8.x)
Key innovations include runtime fields that replace script‑based calculations, the frozen tier for cost‑effective snapshot storage, and an upgrade to JDK 16 with Vector API support for future SIMD acceleration.
Greenplum 6.16
Fixed partition‑key type mismatches, planner issues, and memory‑leak bugs.
Improved gpload handling of mixed‑case column names.
Updated GPCC and GPSS components to latest releases.
Domestic Databases
DM 8.4 – adds DSC+DataWatch, user‑proxy authentication, async standby, dblink gateway, large field support, Linux OS log collection, and many performance and security enhancements.
openGauss 2.0.0/2.0.1 – simplified installation, expansion tools, gray‑scale upgrade, WDR diagnostics, delayed standby, logical replication, Xlog archiving, and intelligent index recommendation.
QianBase 1.6.6SP2 – binlog reader, table/index comparison, multi‑savepoint, query‑cache improvements, backup/recovery, rolling upgrade, and HA support.
ArkDB – MySQL 8.0.23 compatibility, rebuilt physical replication model, faster startup, dead‑lock‑free MVCC index, compressed tablespace, custom page size, and extensive stability testing.
Cloud Database Updates
Alibaba Cloud: Open‑source PolarDB for PostgreSQL, instant column addition in PolarDB MySQL, PolarDB‑X 5.4.10 with LOCALITY syntax and global secondary index recommendations.
Tencent Cloud: RDS MySQL multi‑endpoint read instances, validate_password plugin, database proxy for read/write separation, PostgreSQL security groups and multi‑factor authentication, MongoDB 4.2 with sharded transactions and cross‑AZ deployment, Redis‑plus auto‑elastic scaling, and DBbrain autonomous monitoring features.
JD Cloud: MySQL/MariaDB/Percona small‑version upgrades, TiDB parameter tuning and incremental migration, JCHDB latest releases, PostgreSQL 13 with tagging, JDW data warehouse upgrades, DTS data subscription for PostgreSQL and MySQL‑to‑TiDB sync, JMR component upgrades, and RadonDB MySQL 8.0 compatibility.
Newsletter Purpose
The dbaplus Newsletter aims to provide the community with a concise, unbiased overview of the latest developments across the database landscape, covering RDBMS, NoSQL, NewSQL, time‑series, big‑data and cloud offerings without commercial promotion.
Signed-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
dbaplus Community
Enterprise-level professional community for Database, BigData, and AIOps. Daily original articles, weekly online tech talks, monthly offline salons, and quarterly XCOPS&DAMS conferences—delivered by industry experts.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.
