README.md

BloomFilter

Steps to run sequential (CPU) code

1. gcc -o seq_bloomfilter -std=c99 seq_bloomfilter.c
2. ./seq_bloomfilter inserts.txt lookups.txt
   

Steps to run parallel (GPU) code

1. nvcc -o bloomfilter bloomfilter.cu
2. ./bloomfilter inserts.txt lookups.txt
   
   where
   inserts.txt contains the strings to be inserted
   lookups.txt contains the strings to be looked up
   filenames are just placeholders and can be renamed

Steps to generate random insert and lookup files

• python3 random_input_gen.py -N <num> -P <lookup_percent> -fin <inserts_filename.txt> -flp <lookups_filename.txt>
  
  where
  <num> is the number of random strings to be inserted (default=10000)
  <lookup_percent> is the percentage (0,1) of the generated inserts to be looked up (default=0.2)
  The ratio of lookups to inserts is 10:1 so the remaining lookups are randomly generated
  
  <inserts_filename.txt> is the filename to store the generated inserts (default=inserts.txt)
  <lookups_filename.txt> is the filename to store the generated lookups (default=lookups.txt)
  
  Note: The number of generated strings for insertions and lookups is lesser than the actual value since duplicate strings are filtered out to measure performance of the bloom filter.