Re: map CRDT performance/limits

2014-11-14 Thread Mark Rechler
Thanks for the info Sean and Russell! On Fri, Nov 14, 2014 at 2:42 PM, Russell Brown wrote: > Hi, > Sorry to say that Maps are just riak objects underneath, so the same size > limit applies. Anything over 1mb and you'll start to feel the pain. > > Cheers > > Russell > > On 14 Nov 2014, at 19:34,

Re: map CRDT performance/limits

2014-11-14 Thread Sean Cribbs
Hi Mark, Currently maps are contained in a single key. The same warnings we give for value sizes in regular KV apply; however, the map has additional contained metadata that will give an approximately linear overhead (per-embedded type) to the object. If you can constrain the various domains of y

Re: map CRDT performance/limits

2014-11-14 Thread Russell Brown
Hi, Sorry to say that Maps are just riak objects underneath, so the same size limit applies. Anything over 1mb and you'll start to feel the pain. Cheers Russell On 14 Nov 2014, at 19:34, Mark Rechler wrote: > Hi Everyone, > > I'm curious if anyone has had experience using maps as containers

map CRDT performance/limits

2014-11-14 Thread Mark Rechler
Hi Everyone, I'm curious if anyone has had experience using maps as containers for a large amount of data. Are there any inherent size limits for maps? Could you put sets equivalent to several gigabytes of data into one map? I'd also be curious as to the performance impact of deleting said map. W

Re: Is it possible to start with One node and increase later

2014-11-14 Thread John Daily
The n_val represents the number of copies of the data, regardless of the number of servers in the cluster. If you have n_val=3 and only 1 server, 3 copies of each piece of data will be on that server (and performance will be generally abysmal because every read and write request will effectively

Re: Is it possible to start with One node and increase later

2014-11-14 Thread Santi Kumar
Thanks Luke. I don't want to increase the n_val. All I'm worried is with the default value of 3, with one node would there be any reliability issues. Because there are no other nodes to replicate but according to the configuration 2 other nodes might be expected. Can we continue with single node a

Re: Is it possible to start with One node and increase later

2014-11-14 Thread Luke Bakken
Hi Santi, The default n_val is actually 3. If you start with one node and add nodes there will be no issue. You can increase the n_val but requests for objects may return not_found until read repair or AAE fixes the replicas. -- Luke Bakken Engineer / CSE lbak...@basho.com On Thu, Nov 13, 2014

Is it possible to start with One node and increase later

2014-11-14 Thread Santi Kumar
Hi, I'm new to the Riak. I know the default value of n_val is 2, but is it possible to start with one node, no replication initially and then start adding the nodes and increase the replication factor. What happens if I run only a single node without changing the default values? Does it cause any

Re: bitcask io mode

2014-11-14 Thread Oleksiy Krivoshey
Thanks! What kind of negative impact can the NIF mode cause on Bitcask? On 14 November 2014 08:39, Scott Lystig Fritchie wrote: > Oops, sorry, I overlooked your question. The erlang/nif I/O setting > affect all Bitcask file I/O, see > https://github.com/basho/bitcask/blob/develop/src/bitcask_i