[ https://issues.apache.org/jira/browse/CASSANDRA-19104?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17794334#comment-17794334 ]
Leo Toff commented on CASSANDRA-19104: -------------------------------------- [~bschoeni] Yeah, some questionable marketing tricks forced the industry to clearly define binary prefixes KiB/MiB/GiB. What is 500GB? Is it 500,000,000,000 bytes (500 * 10^9) or 536,870,912,000 bytes (500 * 2^30)? I think it could be either while 500 * 10^9 is ~7% less which is a significant cost reduction "strategy" for the manufacturer. At the same time, 500GiB is definitively 500 * 2^30 bytes. > Standardize tablestats formatting and data units > ------------------------------------------------ > > Key: CASSANDRA-19104 > URL: https://issues.apache.org/jira/browse/CASSANDRA-19104 > Project: Cassandra > Issue Type: Bug > Components: Tool/nodetool > Reporter: Brad Schoening > Assignee: Leo Toff > Priority: Normal > > Tablestats reports output in plaintext, JSON or YAML. The human readable > output currently has a mix of KiB, bytes with inconsistent spacing > Simplify and default output to 'human readable'. Machine readable output is > available as an option and the current mixed output formatting is neither > friendly for human or machine reading and can be replaced. > !image-2023-11-27-13-49-14-247.png! > *Not a goal now (consider a follow up Jira):* > Fix inconsistencies with KiB/MiB/GiB and KB/MB/GB formatting: > * gcstats - uses MB > * getcompactionthroughput - uses MB/s > * getstreamthroughput - uses MB/s > * info - uses MiB/GiB -- This message was sent by Atlassian Jira (v8.20.10#820010) --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org For additional commands, e-mail: commits-h...@cassandra.apache.org