rbranson

44 ms·

451.
by rbranson 15y ago
... because the only use for file I/O in applications is structured tables of data? What about audio/video, documents, graphs, log-structured data, object caches, unstructured flat files, etc? Don't get me wrong, SQLite is awesome, but this
452.
by rbranson 15y ago
Exactly, it's all posturing, but when it comes down to it, they'll cut a last-minute deal to keep things afloat.
453.
by rbranson 15y ago
Yeah those Erlang cooties. ewwwwwwww.
454.
by rbranson 15y ago
It leverages merge_index from Riak Search, which stores the index data on the same partition as the object. Queries are only performed against a subset of partitions that the "query planner" thinks will contain the object.
455.
by rbranson 15y ago
This presentation is particularly interesting because I think it really touches on why secondary indices should be implemented at the datastore level and are impractical to simply graft onto a distributed K/V store.
456.
Querying Riak Just Got Easier: Introducing Secondary Indices(slideshare.net)
76 points by rbranson 15y ago|15 comments
457.
by rbranson 15y ago
This makes sense, the landscape was very different. Cassandra wasn't something I would have used back then either.
458.
by rbranson 15y ago
1,000s, 10,000s, 100,000s, millions. MongoDB columns are designed for serialization of rich documents, not store unbounded ranges of data values. Even without tens of thousands of concurrent users, you'll eventually run into a deeply critic
459.
by rbranson 15y ago
Curious to hear why you thought it would take longer to get going with Cassandra. A single-node Cassandra cluster can be up with a 1-line command. Was it the availability of clients or interface abstractions? I understand that tutorial &#38
460.
by rbranson 15y ago
As long as it all the data for all metrics fits in RAM, it will work great in MongoDB. Of course, as soon as it doesn't, you're totally hosed. To expand on the parent, a column store like HBase or Cassandra is perfect as each row can repres
461.
by rbranson 15y ago
Debt isn't the problem, it's that it's not paid down when there is a surplus. We chose to cut taxes or create new entitlements instead of paying down the debt. Postscript: The "take from me by threat of force as taxes" makes you sound like
462.
by rbranson 15y ago
Unfunded liabilities over long periods of time should be indexed for inflation: they shrink. this is why it's totally absurd to throw around the unfunded liabilities figures like they're meaningful. In 100 years, at 3% inflation, $1,026,647
463.
Ruby Feature: Compress a Sequence of Ends(redmine.ruby-lang.org)
6 points by rbranson 15y ago|3 comments
464.
by rbranson 15y ago
I don't think Apple is directly behind these innovations, but they finance it and make things feasible at industrial scale. Apple is basically doing what Wal-Mart did for the retail supply chain: they aren't sitting around waiting for their
465.
ElasticSearch 0.17.0 Released(elasticsearch.org)
16 points by rbranson 15y ago|1 comments
466.
by rbranson 15y ago
Is there any evidence to support this? The value I see generated by small businesses and entrepreneurs has more to do with market incentives, not innovation. I'm seeing them mostly create wealth by offering different mixes of price, quality
467.
by rbranson 15y ago
It's also a popular myth that innovation only comes from small businesses and entrepreneurs. A huge amount of true innovation actually comes out of well-funded universities, businesses, and government projects.
468.
by rbranson 15y ago
"Yo Ruby, I'm really happy for you, but Erlang has one of the best VMs of all time."
469.
by rbranson 15y ago
Erlang is 100% bug free. Use that instead.
470.
Lucene 3.3 / Solr 3.3 Released(lucene.apache.org)
5 points by rbranson 15y ago|0 comments
471.
by rbranson 15y ago
A composite index will work wonders for this situation. EXPLAIN is your friend.
472.
by rbranson 15y ago
Uhm, MySQL indexes efficiently sort in both directions. You can use a composite index to achieve an ordered set with a set of while conditions, even with millions of rows. As long as the app is only fetching like 10 rows, it should be light
473.
Strange Loop 2011 conference schedule published(thestrangeloop.com)
24 points by rbranson 15y ago|1 comments
474.
by rbranson 15y ago
You are correct that it's broken into extents, but as the extents become sparse over time without compaction, I/O gets scattered further and further around the disk. The WAL is as much a durability features as it is about fast recovery, fas
475.
by rbranson 15y ago
A random operation results in random I/O, another unavoidable problem. What I speak of is the way in which modern, MVCC-based databases deal with concurrency and optimally pattern writes in a way that makes reads lock-free. In addition, the
476.
by rbranson 15y ago
Reddit also just uses it as a dumb K/V store and don't take advantage of the column-store model. In theory, all of the comments for a story could be stored as columns in a column family. While this would reduce the write throughput to somet
477.
by rbranson 15y ago
You lose aggregates and joins, arguably the most powerful parts of an ad-hoc query system (business intelligence kind of stuff). Without those features, you must move a great deal of these concerns into the application layer (or whatever ma
478.
by rbranson 15y ago
Let's also consider that MongoDB's entire I/O architecture and data structure layout is built around a global lock and in-place updates, which makes implementing non-blocking compaction nearly impossible without either a MAJOR overhaul of t
479.
by rbranson 15y ago
This was an EBS issue -- restoring EBS volumes into a useful from snapshots was basically impossible as well.
480.
by rbranson 15y ago
"Mongo eliminates the need for a separate object caching layer." -10gen

More ›