Quote from Gary:
batch_mutate makes no atomicity guarantees. It s intended to help avoiding
many round-trips.
It can fail half-way through leaving you with a partially completed batch.
On Mon, Jun 7, 2010 at 9:39 AM, David Boxenhorn wrote:
> Is batch mutate atomic? If not, can we make it so?
>
Is batch mutate atomic? If not, can we make it so?
On Mon, Jun 7, 2010 at 4:11 AM, Tatu Saloranta wrote:
> Yeah, or maybe just "clustering", since there is no branching structure.
> It's quite commonly useful even on regular b-tree style storage (BDB
> et al), as it can reduce per-entry overhead
After a long time (hours) of running, we cannot use nodetool to retrieve
information of cassandra.
[cassan...@nd3-rack0-cloud cassandra]$ ../cassandra/bin/nodetool -h
10.24.1.16 -p 8081 info
Exception in thread "main" java.lang.IllegalArgumentException:
java.lang:type=Memory not found in the conn
I just experienced a weird issue where writing a large amount of text to a
column caused my script to crash and hang without any error output. I'm using
PHP Thrift client and Cassandra 5.1. I did some searching but didn't find
anything related to a maximum column value size. Has anyone run into
Yeah, or maybe just "clustering", since there is no branching structure.
It's quite commonly useful even on regular b-tree style storage (BDB
et al), as it can reduce per-entry overhead quite a bit. And allows
very efficient compression, if entries have lots of redundancy (xml or
json serialized da
I created a web crawler using Cassandra as the datastore and push to a bunch
of Solr shards. It works well.
-Abhi
On Sun, Jun 6, 2010 at 9:17 AM, Sten Roger Sandvik wrote:
>
> 2010/6/6 David Boxenhorn
>
>> Solr looks like exactly what I want! How mature is it?
>>
>>
> It's very mature. You sho
2010/6/6 David Boxenhorn
> Solr looks like exactly what I want! How mature is it?
>
>
It's very mature. You should also look at ElasticSearch. Much better
distribution model.
/srs
Solr looks like exactly what I want! How mature is it?
On Sun, Jun 6, 2010 at 6:32 PM, William Ashley wrote:
> Have you looked at Solr? Chances are it meets for your needs, and it is
> much simpler than Lucandra.
>
>
> On Jun 6, 2010, at 7:44 AM, David Boxenhorn wrote:
>
> > We're thinking of us
Have you looked at Solr? Chances are it meets for your needs, and it is much
simpler than Lucandra.
On Jun 6, 2010, at 7:44 AM, David Boxenhorn wrote:
> We're thinking of using Lucandra. We already use Lucene, but not in a
> production-level environment, and we are concerned about the problem
i am curious how to intersect results from multi-terms with lucandra
2010/6/6 David Boxenhorn
> We're thinking of using Lucandra. We already use Lucene, but not in a
> production-level environment, and we are concerned about the problem of
> distributing Lucene over multiple servers. Lucandra se
We're thinking of using Lucandra. We already use Lucene, but not in a
production-level environment, and we are concerned about the problem of
distributing Lucene over multiple servers. Lucandra seems like an obvious
solution to this problem.
Any comments or advice?
My primary concern at this poin
sounds interesting... btree on top of cassandra ;)
On Sun, Jun 6, 2010 at 12:16 PM, David Boxenhorn wrote:
> I'm still thinking about the problem of how to handle range queries on very
> large sets of data, using Random Partitioning.
>
> Has anyone used tree search to solve this? What do you thi
I'm still thinking about the problem of how to handle range queries on very
large sets of data, using Random Partitioning.
Has anyone used tree search to solve this? What do you think?
More specifically, something like this:
- Store a maximum of 1000 values per supercolumn (or some other fixed
n
Thanks everybody. This your advice will be carefully considered in our
decision making.
On Fri, Jun 4, 2010 at 1:46 AM, Ben Standefer wrote:
> Mike, yep, there are a lot of benchmarks proving it (plus it just makes
> sense)
>
> http://stu.mp/2009/12/disk-io-and-throughput-benchmarks-on-amazons-e
14 matches
Mail list logo
