Estimates how parallel batches count affects load time of same amount of persistent data.
The lesser score (average time per load) the better.

Results in short: looks like 1 batch per thread is enough. But we can keep x2 for network issues.

servers - server nodes
maxDsOps - streamer's batches per node
sendMsgDelay - simulated network delay or write responses.

==================================================================================================================================================================
Benchmark                                               (cacheWriteMode)  (maxDsOps)  (sendMsgDelay)  (servers)  (walMode)  Mode  Cnt   Score     Error  Units

SERVERS: 2
JmhPersistentStreamerReceiverBenchmark.benchIndividual      PRIMARY_SYNC           4               3          2   LOG_ONLY  avgt    3   8,879 ±  21,115   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual      PRIMARY_SYNC           8               3          2   LOG_ONLY  avgt    3   9,516 ±  20,080   s/op
>>> [Load time decrease stopped at `CPUs * 1`]
JmhPersistentStreamerReceiverBenchmark.benchIndividual      PRIMARY_SYNC          16               3          2   LOG_ONLY  avgt    3   6,328 ±  11,400   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual      PRIMARY_SYNC          32               3          2   LOG_ONLY  avgt    3   8,272 ±  14,475   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual      PRIMARY_SYNC          64               3          2   LOG_ONLY  avgt    3   6,470 ±  10,581   s/op
=== FULL_SYNC: 
>>> [The same load time]
JmhPersistentStreamerReceiverBenchmark.benchIndividual         FULL_SYNC           4               3          2   LOG_ONLY  avgt    3  11,965 ±  23,694   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual         FULL_SYNC           8               3          2   LOG_ONLY  avgt    3  10,712 ±  62,010   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual         FULL_SYNC          16               3          2   LOG_ONLY  avgt    3   9,671 ±   8,370   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual         FULL_SYNC          32               3          2   LOG_ONLY  avgt    3  11,302 ±  31,113   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual         FULL_SYNC          64               3          2   LOG_ONLY  avgt    3  10,407 ±  21,700   s/op

SERVERS: 3
>>> [The same load time]
JmhPersistentStreamerReceiverBenchmark.benchIndividual      PRIMARY_SYNC           4               3          3   LOG_ONLY  avgt    3   8,950 ±  13,192   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual      PRIMARY_SYNC           8               3          3   LOG_ONLY  avgt    3   7,891 ±  14,695   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual      PRIMARY_SYNC          16               3          3   LOG_ONLY  avgt    3   8,788 ±   1,277   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual      PRIMARY_SYNC          32               3          3   LOG_ONLY  avgt    3   9,041 ±  26,050   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual      PRIMARY_SYNC          64               3          3   LOG_ONLY  avgt    3   8,169 ±   6,059   s/op
=== FULL_SYNC:
>>> [The same load except fluctuation with x64]
JmhPersistentStreamerReceiverBenchmark.benchIndividual         FULL_SYNC           4               3          3   LOG_ONLY  avgt    3  14,333 ±   2,293   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual         FULL_SYNC           8               3          3   LOG_ONLY  avgt    3  15,487 ±  12,209   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual         FULL_SYNC          16               3          3   LOG_ONLY  avgt    3  15,182 ±  27,163   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual         FULL_SYNC          32               3          3   LOG_ONLY  avgt    3  14,643 ±  19,024   s/op
JmhPersistentStreamerReceiverBenchmark.benchIndividual         FULL_SYNC          64               3          3   LOG_ONLY  avgt    3  22,955 ± 273,853   s/op
==============================================================================================================================================================

OS: Linux void 5.18.0-4-amd64 #1 SMP PREEMPT_DYNAMIC Debian 5.18.16-1 (2022-08-10) x86_64 GNU/Linux

# JMH 1.13
# VM version: JDK 11.0.2, VM 11.0.2+9
# VM options: -Xms2g -Xmx2g -server -XX:+AlwaysPreTouch

CPU:
vendor_id       : AuthenticAMD
cpu family      : 25
model           : 80
model name      : AMD Ryzen 7 5800U with Radeon Graphics
cpu MHz         : 1600.000
cache size      : 512 KB
siblings        : 16
cpu cores       : 8
cpuid level     : 16

Disk:
`dd if=/dev/zero of=/tmp/test1.img bs=1G count=1 oflag=dsync`: 1073741824 bytes (1,1 GB, 1,0 GiB) copied, 1,25661 s, 854 MB/s
`dd if=/dev/zero of=/tmp/test2.img bs=512 count=1000 oflag=dsync`: 512000 bytes (512 kB, 500 KiB) copied, 2,80265 s, 183 kB/s