Documente Academic
Documente Profesional
Documente Cultură
Bloom Filters
References
A. Broder and M. Mitzenmacher, “Network
Network applications of Bloom
filters: A survey,” Internet Mathematics, vol. 1 no. 4, pp. 485-509,
2004.
1
2/16/2017
2
2/16/2017
N
Numerous applications
li ti iin networking
t ki since
i 2000
2/16/2017 Bloom Filters (Simon S. Lam) 3
3
2/16/2017
4
2/16/2017
Insert X1 and X2
Check Y1 and Y2
5
2/16/2017
6
2/16/2017
7
2/16/2017
1 kn k
f ' = (1 − p ') = (1 − (1 − ) )
k
m
f = (1 − p ) k = (1 − e − kn / m ) k
Two competing forces as k increases
o Larger k ->> (1 − p ')) k is smaller for a fixed p’
p
8
2/16/2017
Number of
9
2/16/2017
10
2/16/2017
11
2/16/2017
12
2/16/2017
Optimal k is
1 1
k 'opt = ln( p ') = ln(1 / 2)
n ln(1 − 1 / m ) n ln(1 − 1 / m )
13
2/16/2017
14
2/16/2017
m/n
2/16/2017 Bloom Filters (Simon S. Lam) 15
15
2/16/2017
16
2/16/2017
17
2/16/2017
nk 1
P[c(i ) ≥ j ] ≤ j
j m
j
enk
≤ (a very loose upper bound)
jm
18
2/16/2017
19
2/16/2017
20
2/16/2017
Conclusions
Wherever a list or set is used, and space is at
a premium,
prem um, a Bloom filter
f lter may be used if
f the
effect of false positives can be mitigated
o No false negative
With a counting Bloom filter, false negatives
are possible, albeit highly unlikely
21
2/16/2017
The End
22