HackerTrans
TopNewTrendsCommentsPastAskShowJobs

awmarthur

no profile record

Submissions

SwissMap: A Golang Port of SwissTable

github.com
2 points·by awmarthur·3 jaar geleden·0 comments

SwissMap: A smaller, faster Golang Hash Table

dolthub.com
2 points·by awmarthur·3 jaar geleden·0 comments

Acid Transactions in Dolt Database

dolthub.com
3 points·by awmarthur·4 jaar geleden·0 comments

Hashing Golang Types with AES Instructions

github.com
11 points·by awmarthur·4 jaar geleden·1 comments

Hacking Go's runtime with generics

dolthub.com
69 points·by awmarthur·4 jaar geleden·109 comments

Benchmarking Dolt with TPC-C

dolthub.com
1 points·by awmarthur·4 jaar geleden·0 comments

Dolt Is a Merkle-Tree Database

dolthub.com
3 points·by awmarthur·4 jaar geleden·0 comments

SQLite Diff vs. Dolt Diff

dolthub.com
3 points·by awmarthur·4 jaar geleden·0 comments

Building a Merkle DAG KeyValue Store

dolthub.com
4 points·by awmarthur·4 jaar geleden·0 comments

Generics can make your Go code Faster

dolthub.com
2 points·by awmarthur·4 jaar geleden·0 comments

comments

awmarthur
·4 jaar geleden·discuss
(OP) I'm mostly interesting in a fast hash for writing a generic, concurrent hash map.

The reference to Tries came from the original issue in the Golang repo about possible uses for a standard library hash function: https://github.com/golang/go/issues/21195

And you're right, the Trie implementation linked there was indeed a Hash-Array Mapped Tried: https://pkg.go.dev/github.com/lleo/go-hamt-key#Key
awmarthur
·4 jaar geleden·discuss
related and useful: "Ribbon filter: Practically smaller than Bloom and Xor" (https://arxiv.org/abs/2103.02515)

This paper does a good job of showing the space/time tradeoffs of various approximate sets.
awmarthur
·4 jaar geleden·discuss
That sounds somewhat similar to Dolt's storage index structure: Prolly Trees https://www.dolthub.com/blog/2020-04-01-how-dolt-stores-tabl...