seancribbs

5 ms·

1.
by seancribbs 11y ago
You mean AP, not CA. CA doesn't exist.
2.
There Is No Now – Problems with Simultaneity in Distributed Systems(queue.acm.org)
215 points by seancribbs 12y ago|71 comments
3.
by seancribbs 12y ago
The Solr version in the Release Candidate is actually 4.7.0.
4.
by seancribbs 13y ago
mnesia is indeed a frequent choice by newbies for queryable storage in Erlang, but I've found few production systems use it for similar reasons to the ones you describe. Nowadays you're more likely to see external datastores (Post
5.
by seancribbs 13y ago
#!/usr/bin/env escript main(_) -> io:format("hello, world").
6.
Verified Vector Clocks: An Experience Report, Part 3(christophermeiklejohn.com)
2 points by seancribbs 13y ago|0 comments
7.
by seancribbs 13y ago
Related: Datomic can be configured to use both Zookeeper and Riak, and the implementation is really clever: immutable data in Riak and strongly-consistent "pointers" in Zookeeper. See http://www.infoq.com/presentat
8.
by seancribbs 13y ago
Since they're not likely to approve my comment on their blog, here's what I said: "Way to misrepresent[1] vector clock usage in Riak! LWW deliberately ignores the vector clock. No one would use that in production without a st
9.
by seancribbs 13y ago
Elixir has some good things about it (hygienic macros, protocols, and so on) but the syntax is not enough by itself to make us switch. "Ugh, Erlang syntax sucks" is a refrain heard around the Internet, but it has not hindered us f
10.
by seancribbs 14y ago
This is true, but Riak CS has special configuration requirements for the backing Riak cluster that might make it not the best choice for combining with your typical Riak applications.
11.
by seancribbs 14y ago
That's still an oversimplification. POST only implies that the request is NOT idempotent, and that the request body is submitted to the origin server for "processing". Consider for example, resources that search for other resources via info
12.
by seancribbs 14y ago
Go to http://basho.com/resources/downloads/ , static links are available if you navigate down the tree to the individual files. e.g. http://s3.amazonaws.com/downloads.basho.com/riak/1.2/1.2.0/o...
13.
Protobuffs in Riak 1.2(basho.com)
44 points by seancribbs 14y ago|1 comments
14.
by seancribbs 14y ago
@rb2k_ Here's the relevant docs: http://rdoc.info/gems/riak-client/Riak/RObject#indexes-insta... http://rdoc.info/gems/riak-client/Riak/Bucket#get_index-inst...
15.
by seancribbs 14y ago
Apparently it's too hard to download yourself? wget http://downloads.basho.com.s3-website-us-east-1.amazonaws.com/riak/1.1/1.1.2/riak-1.1.2-1.fc15.x86_64.rpm rpm -Uvh riak-1.1.2-1.fc15.x86_64.rpm
16.
Riak 1.1 Released(lists.basho.com)
108 points by seancribbs 15y ago|26 comments
17.
by seancribbs 15y ago
To what point? Already large portions of Riak (the performance-critical ones) are in C, and have been for a long time.
18.
by seancribbs 15y ago
There will be a minor gem release (pre/beta/something) next week that supports Riak 1.0 features, then in the next major release those will also bubble up to the Ripple document layer. Sorry for the delay.
19.
by seancribbs 15y ago
Yes, the problem was not necessarily using a Bloom filter, but that it was too small. However, 1.0 is smarter about which vnodes it sends list-keys requests to and thus obviates the need for the Bloom filter (at least for that operation).
20.
by seancribbs 15y ago
That is still under debate AFAIK. The latter form would allow you to compose multiple index lookups as well, which is why I'm for it (although the query planner might not support it yet).
21.
by seancribbs 15y ago
The design of secondary indexes is much simpler, more tightly integrated, and more transparent than Riak Search. It will feel more like part of the same datastore, rather than something you have to deal with on the side.
22.
by seancribbs 15y ago
On the contrary, QuickCheck is brilliant at testing stateful systems, as long as you can model the actions that occur and how they modify state (statem in eqc helps here).
23.
by seancribbs 16y ago
Blog post about the release: http://blog.basho.com/2011/01/05/riak-0.14-released/
24.
by seancribbs 16y ago
Riak has been open-source and freely available since August 2009.
25.
by seancribbs 16y ago
Riak means "ripple" in Indonesian. This is both the name of a pre-release version of the software and a reference to how storage operations in the k/v store work.
26.
InfoQ: Justin Sheehy on Riak(infoq.com)
30 points by seancribbs 16y ago|3 comments
27.
by seancribbs 16y ago
We have quite a few customers in production. Most well-known one is Mochi Media, who have multiple clusters backing their online games APIs.
28.
by seancribbs 17y ago
His argument actually seems to point to dynamically-typed functional programming, which is not surprising, considering his work with Scheme.
29.
by seancribbs 17y ago
We uploaded the generated files to S3 at the end of each individual job. We could have used AMQP's 'ack' for each message, but it turned out not to be necessary.
30.
by seancribbs 17y ago
This is the reality of many projects. We got the full dataset around 48 hours before the output was needed. See my comment above too.

More ›