Percona Backup for MongoDB (PBM) is presented as a superior alternative to common MongoDB backup approaches like volume snapshots and mongodump/mongorestore. Volume snapshots risk cross-shard inconsistency in replica sets and sharded clusters, while mongodump is too slow for large datasets and lacks native point-in-time recovery. PBM addresses these gaps by running distributed agents across all cluster nodes, coordinating cross-shard consistent backups, streaming oplogs continuously for granular RPO, and supporting physical backups (WiredTiger file copies) for dramatically faster restore times — capabilities previously limited to MongoDB Enterprise.
Nguồn: https://www.percona.com/blog/why-percona-backup-for-mongodb-is-the-right-choice-for-production-backups. 8sync News chỉ tóm tắt và dẫn link; bản quyền nội dung thuộc tác giả và nguồn gốc.
Bài viết hướng dẫn chi tiết cách xây dựng nền tảng thương mại điện tử bằng Node.js (backend) kết hợp với các framework frontend như Angular, React hoặc Vue.js. Nội dung bao gồm thiết lập backend, cài đặt packages, cấu hình server, tạo models và routes với MongoDB, kiểm thử API, cũng như phát triển tính năng giỏ hàng và bài tập thực hành.
Lập trình viên Node.js nên đọc bài này để học cách xây dựng cơ sở hạ tầng backend cho ứng dụng e-commerce hiệu quả, từ thiết lập cơ sở dữ liệu MongoDB đến triển khai tính năng tiện ích cart với các framework frontend phổ biến.
DumboDB v0.2.2 introduces Log Filters for the dumboLog command, enabling expressive querying of a document database's commit history. Three filter types are supported: collection-level filtering (analogous to Git file filtering), _id-based filtering to track specific document changes, and $match-based filtering that applies MongoDB's query operators against commit diffs. A new DumboDB-specific $changed operator lets you find commits where a field changed regardless of value. The release also adds an all argument to walk all branch HEADs and pagination support via next/from arguments. Examples use a simulated e-commerce dataset with products, customers, and orders collections.
DumboDB 0.3.0 fixes numerous index bugs that made the MongoDB-compatible, version-controlled database largely unusable. On the MongoDB parity side, deleteMany and updateMany failed to remove index keys, the query planner always chose full column scans over index scans, and .hint() with $natural caused panics. On the version-control side, merging branches caused index corruption with duplicate unique entries, no structural sharing occurred during merges, and conflict resolution returned all null values. The fixes involved writing 77 parity tests run against both MongoDB and DumboDB, porting Dolt's structural sharing logic, and introducing a new dumboConflicts response that provides actionable conflict details including conflict type, human-readable messages, and full three-way merge documents. Query performance improvements are planned for a future release.
MongoDB Atlas celebrates its 10th anniversary, tracing its evolution from a managed cloud database launched in 2016 to an AI-ready data platform. Key milestones include multi-cloud cluster support across AWS, Azure, and Google Cloud, ACID transactions, Atlas Search, Vector Search, and the acquisition of Voyage AI for improved embedding and reranking. Today Atlas serves over 250,000 new builders monthly, processes more than three trillion queries per day, and hosts over 726,000 vector indexes across nearly 70,000 customers. The post highlights how Atlas is now positioned as the data layer for agentic AI applications, enabling semantic search, RAG, and real-time context retrieval without requiring separate infrastructure.
MongoDB has introduced Native Reranking as a public preview feature in Atlas, powered by Voyage AI. It runs directly within the MongoDB aggregation pipeline and can improve retrieval quality by up to 30%. Analysts highlight that embedding reranking natively into the database reduces developer overhead by eliminating separate orchestration layers, retry logic, and failure handling. For CIOs, it simplifies AI stack governance and can lower inference costs by trimming irrelevant context before it reaches expensive LLMs. However, trade-offs include potential vendor lock-in and limited value for enterprises whose data spans multiple repositories. The move reflects a broader industry trend of database and data warehouse vendors consolidating AI, transactional, and analytical capabilities to reduce the integration complexity from multi-vendor AI stacks.
Part 3 of a series on building an AI-powered operations assistant using Spring AI and MongoDB Atlas. This installment introduces stateful workflow checkpointing to handle multi-step incident investigations that span session boundaries. Key additions include: a checkpoint document stored in MongoDB tracking workflow state, current step, and status (RUNNING, WAITING_APPROVAL, COMPLETED, FAILED); a ServiceStatusTool annotated with @Tool that lets the model fetch real-time service metrics; ThreadLocal-based context propagation for passing conversation IDs into tool methods; ToolExecution audit documents for full traceability; and pause/resume API endpoints that rehydrate investigation state. MongoDB serves as a unified persistence layer for runbooks, conversation memory, long-term knowledge, and workflow checkpoints. The architecture allows horizontal scaling since workflow state lives outside any single instance.
The Linux 7.2 kernel's memory management (MM) subsystem has received several notable improvements now merged to Git. The headline change is an improvement to MGLRU's reclaim loop and dirty writeback handling, yielding up to 30% throughput gains for MongoDB with NVMe storage and up to 100% gains with slower I/O. Other improvements include faster batch freeing of order-0 pages (addressing vmalloc regressions), DAMON enhancements for pausing/resuming without state loss and monitoring general data attributes, KASAN speedups by removing ineffective stack/page-table tagging, and tightened mmap_miss hit accounting that reduces synchronous mmap readahead I/O for sparse random and strided memory access workloads.
Zernio, API mạng xã hội hợp nhất cho 15 nền tảng, đã thay thế MongoDB bằng Tinybird để xử lý 6 triệu bài đăng mỗi ngày. Nhờ Tinybird, đội 7 người triển khai thành công trong vòng chưa đầy một tuần, đạt 700.000 yêu cầu API/ngày với độ trễ 250ms (p95) và ingest 16,7 triệu dòng dữ liệu mỗi ngày, đồng thời đẩy nhanh tốc độ phát triển tính năng mới từ tuần xuống giờ.
Lập trình viên cần đọc bài này để hiểu cách chuyển đổi từ MongoDB sang Tinybird giúp tối ưu hóa quy trình phát triển, giảm thời gian triển khai tính năng mới từ hàng tuần xuống chỉ vài giờ trong một đội nhỏ chỉ 7 người.