Re: [DNSOP] Clarification question: compression pointers always to names earlier in the packet?

[Shane Kerr <[e-mail]>]

John,

On 24/10/2018 15.38, John Dickinson wrote:
> On 24 Oct 2018, at 10:01, Viktor Dukhovni wrote:
> 
>> My reading of RFC 1035 is that DNS name "compression"
>> via "pointers" is restricted to name strictly earlier
>> in the DNS message:
>>
>>    4.1.4. Message compression
>>
>>    In order to reduce the size of messages, the domain system utilizes a
>>    compression scheme which eliminates the repetition of domain names 
>> in a
>>    message.  In this scheme, an entire domain name or a list of labels at
>>    the end of a domain name is replaced with a pointer to a prior 
>> occurance
>>                                                             
>> ---------------
>>    of the same name.
>>
> 
> Not strictly to do with loops but we noticed that not all nameservers 
> use the same compression algorithm. See section 9.1 and appendix B of 
> https://datatracker.ietf.org/doc/draft-ietf-dnsop-dns-capture-format

It's not too surprising. I am pretty sure that producing optimal name 
compression (by size) is the same as one of packing problems, which are 
either NP-complete or NP-hard depending on the particulars:

https://en.wikipedia.org/wiki/Packing_problems

I think that this is only true if you allow re-ordering of RRset as well 
as RR within a set, but still.

If you don't allow re-ordering then I think the overall complexity of 
compressing a message is the same as sorting, O(NlgN), which while a lot 
better is still not great, because you need to search through all prior 
labels to find the longest match when compressing each name.

I think it's due to these fundamentally expensive costs that different 
name servers use the various heuristics you mention in the C-DNS draft 
to try to get a good balance of speed & size.

Cheers,

--
Shane