Redis 7.8 version-specific expert. Deep knowledge of new data type enhancements, cluster improvements, enhanced observability features, and performance optimizations. WHEN: "Redis 7.8", "Redis 7.8 features", "Redis 7.8 cluster", "Redis 7.8 improvements".
You are a specialist in Redis 7.8, released approximately mid-2025. You have deep knowledge of the features introduced in this version, including data type enhancements, cluster improvements, enhanced observability, and performance optimizations.
Support status: Supported. EOL May 2027.
Improved slot migration performance:
# Faster resharding with pipeline depth
redis-cli --cluster reshard node1:6379 --cluster-pipeline 20
# Migrates 20 keys per batch instead of 1, significantly faster for large slots
# Improved cluster check with detailed diagnostics
redis-cli --cluster check node1:6379 --cluster-search-multiple-owners
# Detects and reports slots claimed by multiple masters (split-brain indicator)
Cluster topology awareness enhancements:
# CLUSTER MYSHARDID -- identify which shard this node belongs to
redis-cli -c CLUSTER MYSHARDID
# Returns: shard identifier for the current node
# Enhanced CLUSTER SHARDS output
redis-cli -c CLUSTER SHARDS
# Now includes: health status, replication lag per replica, slot migration status
Improved failover behavior:
Enhanced INFO sections:
redis-cli INFO latencystats
# Expanded per-command latency percentiles with p50, p95, p99, p99.9
# More command categories tracked
# Histogram-based distribution for more accurate tail latency reporting
redis-cli INFO errorstats
# Detailed error categorization:
# errorstat_ERR: general errors with breakdown
# errorstat_WRONGTYPE: type mismatch errors
# errorstat_MOVED: cluster redirections
# errorstat_ASK: cluster ask redirections
# errorstat_CLUSTERDOWN: cluster unavailable errors
# errorstat_LOADING: server loading errors
LATENCY improvements:
# More granular latency event categories
redis-cli LATENCY LATEST
# New events tracked:
# active-defrag-cycle -- defragmentation latency
# expire-cycle -- enhanced detail
# cluster-cron -- cluster maintenance overhead
# aof-stat -- AOF stat tracking overhead
CLIENT LIST enhancements:
redis-cli CLIENT LIST
# New fields:
# lib-name: client library name (e.g., redis-py, jedis)
# lib-ver: client library version
# tot-net-in: total bytes received from this client
# tot-net-out: total bytes sent to this client
# tot-cmds: total commands processed for this client
Stream enhancements:
# XREAD with consumer group improvements
# Better handling of consumer crashes and message redelivery
# More efficient XAUTOCLAIM for large pending entry lists
# Enhanced XINFO output
redis-cli XINFO STREAM mystream FULL COUNT 10
# Now includes: entries-added, max-deleted-entry-id, recorded-first-entry-id
# Better visibility into stream lifecycle
Sorted set improvements:
# ZRANGESTORE performance optimization
ZRANGESTORE dest src 0 100 BYSCORE LIMIT 0 50
# Reduced memory allocation overhead for large range operations
# Better ZRANGEBYSCORE/ZRANGEBYLEX performance
# Optimized skiplist traversal for range queries with LIMIT
Listpack optimization:
Memory allocation optimization:
Command processing:
I/O threading refinements:
Benchmarks (typical improvement over 7.4):
| Workload | Improvement |
|---|---|
| GET/SET (small keys) | ~5-8% throughput |
| Pipeline (100 commands) | ~10-15% throughput |
| Large key operations (>1MB) | ~15-20% latency reduction |
| SCAN (large keyspace) | ~20% faster |
| Cluster resharding | ~30-50% faster |
ACL improvements:
# More granular key permissions with read/write distinction
ACL SETUSER reader on >pass %R~app:* +get +mget +scan
# %R~pattern: read-only access to matching keys
# %W~pattern: write-only access to matching keys
# Improved ACL LOG detail
redis-cli ACL LOG 10
# Now includes: entry-id for unique identification
# Better context for pub/sub channel denials
# Timestamp precision improved to milliseconds
TLS improvements:
New parameters:
# Enhanced active defragmentation controls
active-defrag-max-scan-fields 1000
# Maximum hash/set/zset fields scanned per defrag cycle step
# Improved replication buffer management
repl-min-slaves-max-lag-threshold 5
# More granular replica lag detection
# Stream memory management
stream-node-max-bytes 4096
stream-node-max-entries 100
# Per-node limits for stream radix tree nodes
Modified defaults:
# hz default increased from 10 to adaptive (dynamic-hz yes remains default)
# Better default I/O thread configuration detection
# Improved auto-tuning for latency-monitor-threshold
This version introduced:
Not available in this version (added later):
Available from previous versions:
Backward compatible: No breaking changes. All 7.4 commands and configurations continue to work.
Recommended steps:
INFO latencystats, SLOWLOG GET 20redis-cli --cluster check after each nodePost-upgrade recommendations:
INFO errorstats for previously invisible error patternsCLIENT SETNAME in application clients to leverage new CLIENT LIST fieldsactive-defrag-max-scan-fields for large-collection workloadsImportant: Redis 8.0 is a major release with potential breaking changes. Review the 8.0 SKILL.md carefully before upgrading. Key considerations:
# Verify I/O threading is optimally configured
redis-cli CONFIG GET io-threads
redis-cli CONFIG GET io-threads-do-reads
# Check new latency stats
redis-cli INFO latencystats
# Verify active defragmentation with new controls
redis-cli CONFIG GET active-defrag-max-scan-fields
redis-cli INFO memory | grep active_defrag
# Test cluster resharding speed improvement
redis-cli --cluster reshard node1:6379 --cluster-pipeline 20