Hi,

BLASTn should do the job if you are just interested in finding DNA fragments within larger DNA sequences:

https://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE_TYPE=BlastSearch

Once you selected several sequences, constructing a multiple alignment could be helpful:

http://mobyle.pasteur.fr/cgi-bin/portal.py#forms::mafft

If you want to compare a number of sequences to identify a conserved DNA segment, then you need to do a multiple sequence alignment. Algorithms like MUSCLE, CLUSTAL, COBALT etc. will do the job.

Alternatively, if you need to look for a specific DNA segment (motif) in your sequences, then try FIMO (http://meme-suite.org/tools/fimo).

Good luck brother!

You can also use BLAT from UCSC on their annotated genomes

At this glance of this situation, you have to be more specific about the nature of sequences, because as Boas Pucker suggested blastn is the standard strategy, but it relies on a strong dependence of the specific parameters. So, I suggest to know more deeply the nature of your sequences because by blastn with default parameters could generate nothing or merely noise.

Also, the concept of 'sharing' must be previously defined, in accordance to the information that you want to search. So, Do you want to retrieve the most similar sequence in another specie? (i.e identity greater than 90% and coverage greater than 90% respect the query length, maybe use ) or somewhat flexible, (i.e identity greater that 60% and a coverage greater than 50%, BLAST). More information and comparisons: 

http://www.ebi.ac.uk/training/online/course/ensembl-browsing-chordate-genomes/how-search-ensembl/searching-sequence-using-blat-or-blast