Library
47 results
Clear filters
ElasticsearchIntermediateDynamic mapping made your field text, so term queries miss, and mappings can't be changed in place. Use keyword fields, then reindex behind an alias with zero downtime.
#indexing#schema-design#troubleshooting
RedisIntermediateOne slow command blocks every client. Find it with SLOWLOG and LATENCY DOCTOR, then replace KEYS with SCAN, split big keys, delete with UNLINK and tame fork pauses.
#caching#performance#troubleshooting
ScyllaDBIntermediateA query without the partition key means a full cluster scan. When ALLOW FILTERING is fine, and how query-first tables, materialized views and global indexes fix the rest.
#data-modeling#indexing#nosql
TimescaleDBIntermediateSince TimescaleDB 2.13 continuous aggregates are materialized-only by default. Refresh policies, real-time aggregation and backfilling explained.
#maintenance#performance#troubleshooting
RedisIntermediateRedis and Memcached compared by mechanism: threading, slab vs allocator memory, expiry and eviction, persistence, clustering, meta commands, operations and licensing, with a recommendation.
#caching#internals#nosql
InfluxDBIntermediateEvery distinct tag value creates new series. Unbounded tags such as IDs and URLs cause slow writes, query timeouts and OOMs. Measure cardinality and fix the schema.
#performance#schema-design#troubleshooting
Throttled at 30% usage? Throughput is split across physical partitions, so one hot partition key throttles alone. Diagnose it, then fix keys, RU costs and scaling.
#data-modeling#performance#troubleshooting
DuckDBIntermediateDuckDB can spill to disk, but memory_limit, temp_directory, insertion order and thread count decide whether it does. Settings and query patterns that prevent OOM errors.
#configuration#performance#troubleshooting
MongoDBIntermediateA catalogue of twelve MongoDB modeling patterns, each with a document shape, the query or update that uses it, and what it costs.
#data-modeling#nosql#schema-design
QdrantIntermediateFast until you add a filter? Create payload indexes before loading data so the filterable HNSW can use them, and set up multi-tenant collections the right way.
#indexing#performance#vector-search
MongoDBIntermediateHow to design a MongoDB schema from the workload, the hard limits that constrain it, and seven anti-patterns with a way to detect and fix each.
#data-modeling#nosql#performance
