1brc submission by godofwharf

[godofwharf]

Check List:

• You have run ./mvnw verify and the project builds successfully
• Tests pass (./test.sh <username> shows no differences between expected and actual outputs)
• All formatting changes by the build are committed
• Your launch script is named calculate_average_<username>.sh (make sure to match casing of your GH user name) and is executable
• Output matches that of calculate_average_baseline.sh
• For new entries, or after substantial changes: When implementing custom hash structures, please point to where you deal with hash collisions (line number)

• Execution environment: 8 vcores AMD EPYC 7000 (Rome) / 16 GiB RAM / SMT disabled
• Execution time: 7.9 s
• Execution time of reference implementation: 3m 11s

[godofwharf]

@gunnarmorling Please review the same. My implementation doesn't use Unsafe and doesn't use Graal VM. It goes about processing the file in the following manner

• Maps the input file into memory
• Divides the input file into multiple splits and multiple pages
• Processes the multiple splits in parallel using a custom thread pool one page after another
• Uses Java's Vector API for identifying positions of new-line characters within each page
• Uses thread local state maps to perform local aggregation within each split
• Uses a ConcurrentHashMap to create a global aggregate map
• Sorts the station names by importing map entries into a tree map

[gunnarmorling]

00:11.363.