Skip to content

HBASE-29040 Fix description of "sampleRate" of PerformanceEvaluation #6558

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Apache9 merged 3 commits into from
Jan 13, 2025
Merged

HBASE-29040 Fix description of "sampleRate" of PerformanceEvaluation #6558

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
cdf7c94
HBASE-29040 Fix description of "sampleRate" of PerformanceEvaluation
junegunn Dec 23, 2024
989162d
Simplify everyN expression
junegunn Dec 23, 2024
0d62987
Merge branch 'master' into HBASE-29040
junegunn Dec 24, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 2 additions & 3 deletions hbase-diagnostics/src/main/java/org/apache/hadoop/hbase/PerformanceEvaluation.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1205,7 +1205,7 @@ private static long nextRandomSeed() {
this.opts = options;
this.status = status;
this.testName = this.getClass().getSimpleName();
everyN = (long) (opts.totalRows / (opts.totalRows * opts.sampleRate));
everyN = (long) (1 / opts.sampleRate);
Copy link
Contributor

@Apache9 Apache9 Dec 29, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm not sure whether (long) (1 / 1.0) will give us the correct result? Could it be 0?

Copy link
Member Author

@junegunn junegunn Dec 29, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

AFAIK, we don't have to worry about it because 1.0 (1 / 20) has a finite binary representation (unlike numbers like 0.1 or 0.2), so (1 / 1.0) will not yield any arithmetic error.

irb(main):001> [0].pack('G').bytes.map { "%08b" % it }.join
=> "0000000000000000000000000000000000000000000000000000000000000000"
irb(main):002> [1].pack('G').bytes.map { "%08b" % it }.join
=> "0011111111110000000000000000000000000000000000000000000000000000"
irb(main):003> [0.1].pack('G').bytes.map { "%08b" % it }.join
=> "0011111110111001100110011001100110011001100110011001100110011010"
irb(main):004> [0.2].pack('G').bytes.map { "%08b" % it }.join
=> "0011111111001001100110011001100110011001100110011001100110011010"

https://en.wikipedia.org/wiki/Double-precision_floating-point_format#Double-precision_examples

if (options.isValueZipf()) {
this.zipf = new RandomDistribution.Zipf(this.rand, 1, options.getValueSize(), 1.2);
}
Expand Down Expand Up @@ -2707,8 +2707,7 @@ protected static void printUsage(final String shortName, final String message) {
+ "Default: depend on oneCon parameter. if oneCon set to true, then connCount=1, "
+ "if not, connCount=thread number");

System.err.println(" sampleRate Execute test on a sample of total "
+ "rows. Only supported by randomRead. Default: 1.0");
System.err.println(" sampleRate Execute test on a sample of total rows. Default: 1.0");
System.err.println(" period Report every 'period' rows: "
+ "Default: opts.perClientRunRows / 10 = " + DEFAULT_OPTS.getPerClientRunRows() / 10);
System.err.println(" cycles How many times to cycle the test. Defaults: 1.");
Expand Down