Simple but significant performance improvements.

[brettwooldridge]

No description provided.

[codecov-io]

Codecov Report

Merging #330 into master will increase coverage by 0.24%.
The diff coverage is 94%.

@@             Coverage Diff              @@
##             master     #330      +/-   ##
============================================
+ Coverage     79.66%   79.91%   +0.24%     
  Complexity      137      137              
============================================
  Files            11       11              
  Lines           723      732       +9     
  Branches         78       76       -2     
============================================
+ Hits            576      585       +9     
+ Misses          106      104       -2     
- Partials         41       43       +2

Impacted Files	Coverage Δ	Complexity Δ
src/main/java/org/influxdb/impl/InfluxDBImpl.java	83.49% <100%> (+0.8%)	43 <1> (ø)	⬇️
src/main/java/org/influxdb/dto/Point.java	83.22% <92.85%> (-0.11%)	24 <3> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update df18f64...901bf4f. Read the comment docs.

[majst01]

Can you show how big the performance improvements are ?

[brettwooldridge]

Locally measured.

Current:

Single Point Write for 10000 writes of Points took:240.6 ms
5Mio points:7.300 s

New:

Single Point Write for 10000 writes of Points took:192.6 ms
5Mio points:5.306 s

[brettwooldridge]

More difficult to measure, though measurable, and not insignificant, is the reduction of garbage.

[majst01]

Great thanks, i will merge it in the next iteration.
Thanks for your contribution.

[brettwooldridge]

@majst01 Thanks! We may have more contributions in the coming weeks.

[brettwooldridge]

Latest commit gets the maxWritePointsPerformance() test down to 5Mio points: 4.781 s compared to 5Mio points:7.300 s on master. Also generates +800MB less garbage (yes, that test generates several gigabytes of garbage).

[brettwooldridge]

@majst01 A few questions. I see that v2.5 was released six months ago, and the general cadence seems to be 3-6mo between releases. When is v2.6 slated for release? And what is the reason this commit is scheduled for 2.7? That would imply a wait of 6-9 months before we see this commit in a release version.

[majst01]

Oh no, 2.6 should be released in the next few days if not already release (it depends on support from som influxdata guys).
After 2.6 is out i would like to release 2.7 very shortly afterwards, only with a few important fixes.

[majst01]

2.6 is finally released.

[brettwooldridge]

@majst01 Thanks!

[majst01]

Hi Brett,

one last nit, can you please add a proper changelog entry, after that i will merge it.
You should open a new 2.7 section therefore