jaksprats

6 ms·

1.
by jaksprats 9y ago
Two phase commit (in spanner) is similar to a lock, so the short answer is YES you need locks for distributed TRANSACTIONS But you dont need transactions to do distributed computation. This is where CRDTs come in, they allow you to modify d
2.
Birth of the NearCloud: Serverless and CRDTs Edge Is the New Next Thing(highscalability.com)
5 points by jaksprats 9y ago|3 comments
3.
by jaksprats 10y ago
This is a very good question but its also a leading question :) Datanet provides a level of abstraction such that business logic does not need to be framed in commutative operations rather you just do normal data manipulation on JSON object
4.
Datanet a New CRDT Database That Let's You Do Bad Bad Things to Distributed Data(highscalability.com)
9 points by jaksprats 10y ago|2 comments
5.
SegundoNet: Secondary Cell Network Using Google Fiber as Backhaul(slidesha.re)
1 points by jaksprats 12y ago|0 comments
6.
Glass' evolution, some suggestions(slideshare.net)
1 points by jaksprats 12y ago|0 comments
7.
AlchemyDB is all grown up(jaksprats.wordpress.com)
1 points by jaksprats 13y ago|0 comments
8.
by jaksprats 14y ago
Link is back up as of 12:30 PDT
9.
Russ’ 10 Ingredient Recipe For Making 1 Million TPS On $5K Hardware(highscalability.com)
6 points by jaksprats 14y ago|2 comments
10.
3 blog kings of data(jaksprats.wordpress.com)
1 points by jaksprats 15y ago|0 comments
11.
AlchemyDB, world’s first integrated GraphDB + RDBMS + KV Store + Document Store(jaksprats.wordpress.com)
8 points by jaksprats 15y ago|0 comments
12.
by jaksprats 15y ago
in the redis community we have also discussed having a daemon sitting next to the server and doing localhost redis communication. The downsides to doing this are it introduces an extra piece to break. W/ luajit2, which is almost as fast as
13.
by jaksprats 16y ago
that will work (as good as can be hoped for). the process can be thought of as a map-reduce. first pass would be figure out where all your SETs are located and how big they are. then intersect all co-located SETs. then (in parallel) take s
14.
by jaksprats 16y ago
I am the author of Redisql and I wrote a blog post today explaining the concepts behind Redisql (a RDBMS on top of the NOSQL datastore redis) http://jaksprats.wordpress.com/2010/09/28/introducing-redisq... The original idea was to have a
15.
by jaksprats 16y ago
what are you using as backend? mysql? postgres? couchdb? redis?