0

I am aware that ip_to_hostname(STRSXP) in iptools in R does help users to find domain names of IP addresses.

However, with my data sized over 300,000,000 entries of DNS IP addresses, ip_to_hostname is not a feasible option due to the hefty load of processing time that ultimately aborts R client.

I was wondering if anyone had experience with this and figured out a solution.

AvidCoder
  • 11
  • 3
  • how large is the files size of 300,000,000 IPs? With that many entries I am wondering if you will need a to pay for that many lookups in a single run – Daniel_j_iii Aug 06 '20 at 16:01
  • Well, I was able to record them in a facility around 4k users over ten (10) days. The entire size of the PCAP file which I converted to CSV is around 120GB, but of course, there are duplicates in the entries (mainly google dns). I do not have 300,000,000 distinct IPs, but 300,000,000 entries of DNS IP addresses (Source + Destination). EDIT: I decreased my test dataset into just 8000 entries, yet no chance; iptools' `ip_to_hostnname` freezes my system although I have i7-9900K with 32GB of RAM at the moment. – AvidCoder Aug 06 '20 at 17:08
  • I recently wrote my own implementation in [`ipaddress::ip_to_hostname()`](https://davidchall.github.io/ipaddress/reference/ip_to_hostname.html). I can't promise it'll do any better, because the implementations are very similar. But it's worth a shot! – David Hall Aug 06 '20 at 17:56
  • I guess the biggest problem for myself is coding on a Windows system... I do not really want to mess with bash, so I think I will head to my lab and test it on my Linux system. I will give you an update! Thank you! – AvidCoder Aug 06 '20 at 18:27

0 Answers0