The JusDB Database Engineering Blog

A critical bug in MongoDB 8.0 (SERVER-97368) affects TTL index behavior — causing documents to persist beyond expiration or delete prematurely. Here's how to detect and mitigate it.

Compare Redis persistence modes: RDB snapshots for fast restarts, AOF for sub-second durability, and the combined RDB+AOF approach. Includes data loss and I/O tradeoff matrix.

Aerospike 7.0 delivers 1 million writes per second with sub-millisecond latency using its Hybrid Memory Storage architecture. Learn when Aerospike beats Redis and Cassandra, and how to configure it for production.

MongoDB Atlas M30 costs 5.6x more than equivalent EC2 instances, but self-hosted TCO includes DBA labor that most comparisons ignore. Here's the honest math and when each option wins.

Choose the right Cassandra compaction strategy: STCS for write-heavy workloads, LCS for read-heavy point lookups, and TWCS for time-series data with TTL-based expiry.

ScyllaDB 6.0's shard-per-core architecture delivers 3-8x Cassandra throughput on identical hardware. Learn the performance differences, migration path from Cassandra, and when ScyllaDB tablets beat vnodes.

Model DynamoDB data with single-table design. Covers composite sort keys, GSI design for multiple access patterns, and on-demand vs provisioned capacity.

Manage Elasticsearch indexes with ILM policies, correct shard sizing, and force merge for historical indices. Covers hot/warm/delete phases and cluster health monitoring.

Optimize MongoDB aggregation pipelines by ordering stages correctly, using indexes in $match and $sort, projecting early, and tuning $lookup with pipeline filters.

Use Redis Streams for event sourcing and message queues. Covers XADD, XREADGROUP, consumer groups, XAUTOCLAIM for stale messages, and Python producer/consumer examples.

Design Cassandra tables around query patterns, not entities. Covers partition key cardinality, hot partition prevention, time bucketing, and secondary indexes vs materialized views.

DynamoDB and MongoDB are the two most common NoSQL choices, but cost structures, query flexibility, and operational overhead differ dramatically depending on your access patterns.