Re: short length hash key generation for riak keys?

2013-09-30 Thread Alex Rice
Ah thanks Brian, that's good to know! On Mon, Sep 30, 2013 at 12:40 PM, Brian Sparrow wrote: > In addition to Jeremiah's suggestions I would suggest using single character > bucket names. This will save you loads of memory in the long run as each key > in bitcask has its bucket prepended onto its

Re: short length hash key generation for riak keys?

2013-09-30 Thread Alex Rice
That is awesome! thanks Jeremiah - I really need to take the time to browse through all your stuff on Gitub :) On Mon, Sep 30, 2013 at 12:28 PM, Jeremiah Peschka wrote: > Howdy Alex! > > It's not ludicrous at all since changes in Riak 2.0 should reduce that > overhead to about 13 bytes per key,

Re: short length hash key generation for riak keys?

2013-09-30 Thread Jeremiah Peschka
Howdy Alex! It's not ludicrous at all since changes in Riak 2.0 should reduce that overhead to about 13 bytes per key, IIRC. In these situations, you could always use an external identity generation process. Here's where I plug Rustflakes [1] which can generate unsigned 64-bit, decimal, or BigInt

Re: short length hash key generation for riak keys?

2013-09-30 Thread Alex Rice
Although with 40 bytes per key overhead, which I just read about, perhaps this is somewhat of a ridiculous optimization for me to be attempting to shorten the key size ? :) On Mon, Sep 30, 2013 at 12:20 PM, Alex Rice wrote: > Hi all, coming from a traditional database background of course my > fi

short length hash key generation for riak keys?

2013-09-30 Thread Alex Rice
Hi all, coming from a traditional database background of course my first tendency was wanting to auto_increment a primary key somewhere. Not possible with Riak, obviously! :) I was wondering if anyone has a good link or white paper about short length hash key creation. I want to keep my keys short

Re: Riak Control display memory usage

2013-09-30 Thread Christopher Meiklejohn
Hi Nicolas, As we are still investigating another bug with the memory reporting [1], I would wait for the next packaged release of Riak. - Chris [1] https://github.com/basho/riak_control/issues/118 -- Christopher Meiklejohn Software Engineer Basho Technologies, Inc. On Monday, September

Riak Control display memory usage

2013-09-30 Thread nicolasc
Dear all, I am experiencing the bug describe here: https://github.com/basho/riak_control/pull/140 However, I am running version 1.4.2-1, directly installed from the Debian package, using a basic config. All other features of Riak Control seem to be working properly. Any advice for troubleshoo

Re: Riak Control display memory usage

2013-09-30 Thread Christopher Meiklejohn
This bug has been fixed on the master and 1.4 branches, but has not made it into one of our packages yet. My apologies for the trouble. - Chris -- Christopher Meiklejohn Software Engineer Basho Technologies, Inc. On Monday, September 30, 2013 at 11:36 AM, nicolasc wrote: > Dear all, > >

Riak Control display memory usage

2013-09-30 Thread nicolasc
Dear all, I am experiencing the bug describe here: https://github.com/basho/riak_control/pull/140 However, I am running version 1.4.2-1, directly installed from the Debian package, using a basic config. All other features of Riak Control seem to be working properly. Any advice for troubleshoo

Re: Riak Java client isKeyPresent function?

2013-09-30 Thread Sam Elliott
Guido, You can do a HEAD request to just get the object's metadata. In the PB api, there's a boolean `head` that you can set to do the same. It doesn't return the object's value(s) for either interface. Docs: PB: http://docs.basho.com/riak/latest/dev/references/protocol-buffers/fetch-object/#

Riak Java client isKeyPresent function?

2013-09-30 Thread Guido Medina
Hi, Is there a way to quick check if a key is present without fetching it using the Riak Java client? It would be nice to have one for quick checks without fetching the key: /interface Bucket {// // // // public boolean isKeyPresent(String key);// // // //}/ Of course, that wo