Mismatch and negative counts when using hll cardinality

[Sunil Badala]

Hello,

I see a huge difference in intersection number when using hll.

The numbers are just too weird and sometimes negative. In comparison number with actual data sets are perfect.

Can any one let me know why it is happening and what are the alternatives available?

Thanks

Sunil Badala

*************************

CONFIDENTIALITY CAUTION

This communication (including any accompanying documents) is intended only for the use of the addressee(s) and contains PRIVATE AND CONFIDENTIAL information. Unauthorized reading, dissemination, distribution or copying of this communication is prohibited. If you have received this communication in error, please notify us immediately by e-mail and promptly destroy the original communication. Please do not copy it or use it for any other purposes, or disclose its contents to any other person. Unless stated to the contrary, any opinions or comments are personal to the writer and do not represent the official view of the Company. In case where it represents our official view, it needs to be revalidated and is subject to the terms of our engagement.

Thank you for your co-operation.

*************************