Dont see much benefit using a keydb tls enabled cluster for latencies

[sangeethdba]

trafficstars

Describe the bug I dont see any benifit using keydb tls cluster over redis 6.x tls cluster in terms of latency improvements that keydb claiming .intact p99 latencies of redis 6.x cluster looks promising than keydb

** Log Files **

redis 6.x tls enabled cluster:

./memtier_benchmark --hide-histogram --tls --cert /opt/isv/redis/certfiles/newqpssc_certs/com-redis-cluster-qpssc.cer --key /opt/isv/redis/certfiles/newqpssc_certs/private.key -s cmqsrd00 -p 9001 --threads=32 --cacert=/opt/isv/redis/certfiles/newqpssc_certs/rootca.crt --cluster-mode
< Writing results to stdout [RUN #1] Preparing benchmark client... [RUN #1] Launching threads now... [RUN #1 100%, 134 secs] 0 threads: 16000000 ops, 109980 (avg: 118931) ops/sec, 7.80MB/sec (avg: 8.43MB/sec), 41.50 (avg: 38.78) msec latencyyy

32 Threads 50 Connections per thread 10000 Requests per client

ALL STATS

Type Ops/sec Hits/sec Misses/sec MOVED/sec ASK/sec Avg. Latency p50 Latency p99 Latency p100 Latency KB/sec

Sets 11184.06 --- --- 0.00 0.00 41.38825 7.55100 733.18300 933.88700 861.62 Gets 111717.65 111717.65 0.00 0.00 0.00 32.66920 5.37500 659.45500 917.50300 8061.01 Waits 0.00 --- --- --- --- --- --- --- --- --- Totals 122901.71 111717.65 0.00 0.00 0.00 33.46199 5.50300 667.64700 917.50300 8922.63

Key DB tls enabled cluster:

./memtier_benchmark --hide-histogram --tls --cert /opt/isv/redis/certfiles/newqpssc_certs/com-redis-cluster-qpssc.cer --key /opt/isv/redis/certfiles/newqpssc_certs/private.key -s cmqsrd00 -p 9003 --threads=32 --cacert=/opt/isv/redis/certfiles/newqpssc_certs/rootca.crt --cluster-mode < Writing results to stdout [RUN #1] Preparing benchmark client... [RUN #1] Launching threads now... [RUN #1 100%, 121 secs] 0 threads: 16000000 ops, 139138 (avg: 131499) ops/sec, 9.86MB/sec (avg: 9.32MB/sec), 32.15 (avg: 34.84) msec latency

32 Threads 50 Connections per thread 10000 Requests per client

ALL STATS

Type Ops/sec Hits/sec Misses/sec MOVED/sec ASK/sec Avg. Latency p50 Latency p99 Latency p100 Latency KB/sec

Sets 12327.04 --- --- 0.00 0.00 31.80614 4.76700 729.08700 942.07900 949.69 Gets 123134.97 123134.97 0.00 0.00 0.00 30.61775 4.60700 716.79900 937.98300 8884.81 Waits 0.00 --- --- --- --- --- --- --- --- --- Totals 135462.01 123134.97 0.00 0.00 0.00 30.72582 4.60700 720.89500 937.98300 9834.50 $

keydb config:

cluster-enabled yes cluster-config-file /opt/isv/redis/data/9002/nodes.conf dir /opt/isv/redis/data/9002 pidfile /opt/isv/redis/run/node_9002.pid logfile /opt/isv/redis/logs/node_9002.log cluster-node-timeout 5000 #daemonize no appendonly yes

MAKE INTERFACE UNSECURE !!

protected-mode yes bind XX>XX>XX>XX timeout 0 tcp-keepalive 0 tcp-backlog 511 loglevel notice syslog-enabled yes syslog-ident redis_9002 syslog-facility USER databases 16

Snapshotting

save 900 1 save 300 10 save 60 10000

Security

#requirepass C0mpl3x#1

Limits

maxclients 10000 maxmemory-policy noeviction

Append Only Mode

appendonly no appendfilename "appendonly.aof" appendfsync everysec no-appendfsync-on-rewrite no auto-aof-rewrite-percentage 100 auto-aof-rewrite-min-size 64mb

Slow Log

slowlog-log-slower-than 10000 slowlog-max-len 128

Advanced

hash-max-ziplist-entries 512 hash-max-ziplist-value 64 list-max-ziplist-entries 512 list-max-ziplist-value 64 set-max-intset-entries 512 zset-max-ziplist-entries 128 zset-max-ziplist-value 64 activerehashing yes client-output-buffer-limit normal 0 0 0 client-output-buffer-limit slave 256mb 64mb 60 client-output-buffer-limit pubsub 32mb 8mb 60 hz 10 aof-rewrite-incremental-fsync yes

#TLS config port 0 tls-port 9002 tls-cluster yes tls-replication yes

tls-cert-file /opt/isv/redis/certfiles/newqpssc_certs/com-redis-cluster-qpssc.com.cer tls-key-file /opt/isv/redis/certfiles/newqpssc_certs/private.key tls-ca-cert-file /opt/isv/redis/certfiles/newqpssc_certs/rootca.crt server-thread-affinity true server-threads 7

Please copy everything after this line.

To Reproduce

[sangeethdba]

@JohnSully

[sangeethdba]

@JohnSully any help is appreciated

[JohnSully]

Hey @sangeethdba looking now.

[sangeethdba]

@JohnSully any help

[msotheeswaran-sc]

Is there anyone still experiencing this issue, will be helpful to understand how to prioritize this.

[msotheeswaran-sc]

Closing as there has been no response in 30 days