Mirror/zeek
1
0
Fork 0
mirror of https://github.com/zeek/zeek.git synced 2025-10-12 19:48:20 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

Switch to double hashing.

Browse source 
For large k, standard hashing imposes an unnecessary overhead. By switchting to
double hashing, we invoke the hash function code at most two times.
This commit is contained in:
Matthias Vallentin 2014-06-05 16:02:25 +02:00
parent 1d50874256
commit 673607f9a7
1 changed files with 2 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

4
src/probabilistic/bloom-filter.bif
View file

@ -44,7 +44,7 @@ function bloomfilter_basic_init%(fp: double, capacity: count,
size_t optimal_k = BasicBloomFilter::K(cells, capacity); size_t optimal_k = BasicBloomFilter::K(cells, capacity);
size_t seed = Hasher::MakeSeed(name->Len() > 0 ? name->Bytes() : 0, size_t seed = Hasher::MakeSeed(name->Len() > 0 ? name->Bytes() : 0,
name->Len()); name->Len());
const Hasher* h = new DefaultHasher(optimal_k, seed); const Hasher* h = new DoubleHasher(optimal_k, seed);
return new BloomFilterVal(new BasicBloomFilter(h, cells)); return new BloomFilterVal(new BasicBloomFilter(h, cells));
%} %}
@ -84,7 +84,7 @@ function bloomfilter_basic_init2%(k: count, cells: count,
size_t seed = Hasher::MakeSeed(name->Len() > 0 ? name->Bytes() : 0, size_t seed = Hasher::MakeSeed(name->Len() > 0 ? name->Bytes() : 0,
name->Len()); name->Len());
const Hasher* h = new DefaultHasher(k, seed); const Hasher* h = new DoubleHasher(k, seed);
return new BloomFilterVal(new BasicBloomFilter(h, cells)); return new BloomFilterVal(new BasicBloomFilter(h, cells));
%} %}