Various improvements #22
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
With the current subnet whitelist implementation, each lookup requires a linear scan of the entire whitelist. If the whitelist contains millions of entries, this can slow down connections significantly. In other projects, CIDR lookups are often implemented using prefix trees. This makes lookups very fast but the implementation is significantly more complex. An alternative, much simpler solution is to store the whitelisted subnets in a hash set. To perform a query, one can then apply all 33 possible CIDR masks to the IP address and check whether any of the resulting CIDR blocks exists in the hash set. With this implementation, the time complexity of single lookups is still in O(1). An additional speedup is achieved by maintaining a hash set of the CIDR masks occurring in the whitelist, instead of always testing all 33 values. Also, since a strict separation between IP address whitelists and subnet whitelists are no longer needed, the two features are merged.
Currently, the link-connect debug information is squelched if the IP address is whitelisted or does not have a Sender Score. Change this and display the line for all IPv4 connections.
Replace the old formula d - (d / 100) * s by d * (100 - s) / 100 (where d denotes the slow factor and s denotes the score). The new formula does not only match to the expected behavior "interpret the inverse score as a percentage and take that fraction of the slow factor" more closely, it is also less prone to unwanted rounding effects. For example, if the score is 100, the new formula is guaranteed to yield a delay of 0, while the original formula might not.
|
These changes are based on the "constant time whitelist queries" patch which is included here for convenience. If all changes look good, it might be easier to just merge this PR and close #21. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.