[
https://issues.apache.org/jira/browse/FLINK-7465?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16141856#comment-16141856
]
Jacob Park commented on FLINK-7465:
-----------------------------------
[~sunzhenya] When you say you want to process the retract record, are you
meaning that you want to approximately classify records as distinct or
duplicate?
For a finite multi-set, classification can be done by a traditional Bloom
Filter. For an infinite multi-set, there is no well-accepted algorithm of
choice; there exists variants such as Stable Bloom Filters or Randomized Load
Balanced Biased Sampling based Bloom Filters, but they have false negative and
false positive characteristics of which users must make a conscious decision on.
As all Bloom Filter variants degenerate when too many elements are inserted
into it (for finite multi-sets, the size helps size the Bloom Filter for a
desired false positive rate, yet this technique does not work for infinite
multi-sets), if Bloom Filter variants are used for approximate classification,
we should classify records with a tuple of (isDuplicate,
falsePositiveProbability, falseNegativeProbability) such that users incorporate
the probabilities for approximate SLAs to fail fast or to take other actions.
> Add build-in BloomFilterCount on TableAPI&SQL
> ---------------------------------------------
>
> Key: FLINK-7465
> URL: https://issues.apache.org/jira/browse/FLINK-7465
> Project: Flink
> Issue Type: Sub-task
> Components: Table API & SQL
> Reporter: sunjincheng
> Assignee: sunjincheng
> Attachments: bloomfilter.png
>
>
> In this JIRA. use BloomFilter to implement counting functions.
> BloomFilter Algorithm description:
> An empty Bloom filter is a bit array of m bits, all set to 0. There must also
> be k different hash functions defined, each of which maps or hashes some set
> element to one of the m array positions, generating a uniform random
> distribution. Typically, k is a constant, much smaller than m, which is
> proportional to the number of elements to be added; the precise choice of k
> and the constant of proportionality of m are determined by the intended false
> positive rate of the filter.
> To add an element, feed it to each of the k hash functions to get k array
> positions. Set the bits at all these positions to 1.
> To query for an element (test whether it is in the set), feed it to each of
> the k hash functions to get k array positions. If any of the bits at these
> positions is 0, the element is definitely not in the set – if it were, then
> all the bits would have been set to 1 when it was inserted. If all are 1,
> then either the element is in the set, or the bits have by chance been set to
> 1 during the insertion of other elements, resulting in a false positive.
> An example of a Bloom filter, representing the set {x, y, z}. The colored
> arrows show the positions in the bit array that each set element is mapped
> to. The element w is not in the set {x, y, z}, because it hashes to one
> bit-array position containing 0. For this figure, m = 18 and k = 3. The
> sketch as follows:
> !bloomfilter.png!
> Reference:
> 1. https://en.wikipedia.org/wiki/Bloom_filter
> 2.
> https://github.com/apache/hive/blob/master/storage-api/src/java/org/apache/hive/common/util/BloomFilter.java
> Hi [~fhueske] [~twalthr] I appreciated if you can give me some advice. :-)
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
