Interview Problem 2023

A small Java program I created for an interview session.

There are improved versions implemented after I reviewed my code later, including inserting JMH logic to measure the performance

Description

The purpose is to translate the following input string into the expected result string.

input	expected
ÿaaaabbbcccccaaüüüüüüüüzzzzzz	1ÿ2a3b4a5c6z8ü
aaaabcccccaa111001111111122222zzzzzz	1b202a314a525c6z81
aaaabcccccaa1110011111111kkkkkkkkkkkk22222zzzzzz	1b202a314a525c6z8112k
Note: not actual inputs from the interviewer

Getting Started

Executing program

• Execute the program to validate the logic and produce a JMH report

Current Performance Report Summary

# JMH version: 1.36
# VM version: JDK 20.0.2, OpenJDK 64-Bit Server VM, 20.0.2+9-78
# Hardware: Intel® Core™ i5-3470, DDR3 16GB
# OS: Debian 12 64-bit

Result "org.chakraphan.Main.benchProcessV1":
  611593.834 ±(99.9%) 12874.466 ops/s [Average]
  (min, avg, max) = (592353.791, 611593.834, 644412.818), stdev = 17187.044
  CI (99.9%): [598719.368, 624468.300] (assumes normal distribution)

Result "org.chakraphan.Main.benchProcessV2":
  648557.930 ±(99.9%) 35300.770 ops/s [Average]
  (min, avg, max) = (588551.702, 648557.930, 697004.412), stdev = 47125.518
  CI (99.9%): [613257.161, 683858.700] (assumes normal distribution)

Benchmark             Mode  Cnt       Score       Error  Units
Main.benchProcessV1  thrpt   25  611593.834 ± 12874.466  ops/s
Main.benchProcessV2  thrpt   25  648557.930 ± 35300.770  ops/s

Comment

• The previous V2/V3 cannot handle the case when some character count in 2 or more digits
• New slightly change approach has been implemented instead and receive not that much improvement in terms of performance
• The performance has been significantly improved from V1 to V2, because of the simpler sorting logic
• V2 and V3 share the similar performance result

Acknowledgments

• Java Microbenchmark Harness (JMH)