Monday, 2014-09-22

*** jeromatron has joined #magnetodb00:23
*** jeromatron has quit IRC00:27
*** charlesw has joined #magnetodb01:19
*** charlesw has quit IRC01:34
*** charlesw has joined #magnetodb01:44
*** ikhudoshyn has quit IRC01:52
*** aostapenko has quit IRC01:52
*** ikhudoshyn has joined #magnetodb01:53
*** aostapenko has joined #magnetodb01:55
*** charlesw has quit IRC04:15
*** rushiagr_away is now known as rushiagr04:40
*** ajayaa has joined #magnetodb05:24
*** ajayaa has quit IRC05:41
*** ajayaa has joined #magnetodb05:54
*** k4n0 has joined #magnetodb06:20
*** jeromatron has joined #magnetodb07:00
*** jeromatron has quit IRC07:31
*** romainh has joined #magnetodb08:27
*** ajayaa has quit IRC08:32
*** k4n0 has quit IRC08:33
*** rushiagr is now known as rushiagr_away08:44
*** rushiagr_away is now known as rushiagr08:45
*** ajayaa has joined #magnetodb08:48
*** k4n0 has joined #magnetodb08:50
*** rushiagr is now known as rushiagr_away08:56
*** rushiagr_away is now known as rushiagr09:23
*** k4n0 has quit IRC09:44
*** ajayaa has quit IRC09:50
*** ajayaa has joined #magnetodb10:02
*** romainh has quit IRC10:14
*** k4n0 has joined #magnetodb10:29
*** romainh has joined #magnetodb11:06
isviridovHello romainh, how are you today?11:10
romainhHi isviridov, I'm fine thanks :)11:12
romainhI'm interested by magnetodb11:13
isviridovInterested in magnetodb?11:13
romainhyes11:13
romainhI like Cassandra and I think that magnetodb is necessary in the openstack environment11:14
isviridovromainh, good to know.11:15
isviridovAre you DBA, developer?11:15
isviridovHow can I help you?11:15
romainhI'm a developer but I also worked for 4 years as Cassandra DBA11:17
isviridovWOW. Do you have expierence of multi dc replication configuration?11:17
romainhyes11:17
romainhIn my past job, we ran Cassandra over 3 DC11:18
romainhWe started developments with Cassandra 0.6 beta but we started production with 1.011:20
romainhDo you face any issues with multi DC replication?11:21
isviridovOne of the features for kilo-1 in magnetodb is replication between dcs, your eyes and thougths whould be very useful.11:22
romainhI'd be glad to help :)11:25
isviridovrushiagr, great! I believe the specification will be discussed in ML and IRC meetings and I'll try to involve you.11:26
isviridov* rushiagr -> romainh11:27
rushiagrwelcome romainh. Glad to see you :)11:27
isviridovTill then feel free to ask any questions here or in ML11:27
rushiagrsee=meet :D11:27
isviridovrushiagr, good morning, sorry to disturbing ;)11:28
romainhhi rushiagr, nice to meet you too ;)11:28
rushiagrisviridov: no issues. I was already discussing Cassandra with my teammate, and was pondering over its limitations11:28
rushiagr:)11:28
romainhisviridov: ok. I have to dive into magnetodb's source code11:30
rushiagrromainh: If you have some time, I have some questions regarding Cassandra. Newbie questions really, so shouldn't be a big deal for you..11:30
isviridovrushiagr, romainh would love to follow your conversation :)11:31
rushiagrisviridov: I think you too can help me. I'm way too behind w.r.t Cassandra basics :)11:32
rushiagrok, so here are my questions. I am reading and finding out that in Cassandra, secondary index can only be of type 'hash', and not 'range'11:32
rushiagrThey say create another column family (table), where row names are the values from the first 'table', which you want to index on. But if I am using randompartitioner, this won't help either!11:34
rushiagr#cassandra11:34
rushiagrwoops :)11:34
rushiagrso, does that mean I cannot have secondary indexes based on a range?11:36
rushiagr=(range key index)11:36
romainhrushiagr: hmm first of all I'm not a big fan of secondary index. Do you have many many data?11:36
romainh*Do you plan to* have many many data?11:37
rushiagrromainh: let's say I will have a million or so 'items' (supercolumn)11:38
rushiagrI think that comes to 'many', but not 'many many' :)11:38
rushiagrbut yes, let's say my data is going to huge11:39
rushiagrI'm fine with storing data in another columnfamily, and use that columnfamily as 'index'11:40
rushiagrbut is it even possible to have 'randompartitioner' for one index, and 'ordered' one for another?11:40
rushiagrrandompartitioner for one *columnfamily* and ordered one for another11:41
romainhno the partioner is defined by cluster11:43
romainhrushiagr: What are your needs? I'm not sure to understand why you need this11:46
rushiagrromainh: I was trying to map openstack APIs with a cassandra, or a magnetodb data model12:00
rushiagre.g. say glance APIs12:01
*** vivekd has joined #magnetodb12:01
rushiagr'glance list' has sorting, filters, pagination, and what not12:02
rushiagror let's even forget about that. Amazon has local secondary indexes, based on range, and not hash12:05
*** openstackgerrit has joined #magnetodb12:15
*** isviridov is now known as isviridov_away12:34
*** rushiagr is now known as rushiagr_away12:35
*** ajayaa has quit IRC12:45
*** vivekd has quit IRC12:52
*** charlesw has joined #magnetodb13:39
openstackgerritDmitriy Ukhlov proposed a change to stackforge/magnetodb: Improve json schema validation for requests left  https://review.openstack.org/12244513:59
*** k4n0 has quit IRC14:19
openstackgerritAndrei V. Ostapenko proposed a change to stackforge/magnetodb: (WIP) "ALL_OLD" return_values for put_item  https://review.openstack.org/12191114:25
openstackgerritDmitriy Ukhlov proposed a change to stackforge/magnetodb: Improve json schema validation for requests left  https://review.openstack.org/12244514:58
*** idegtiarov is now known as idegtiarov_afk15:08
openstackgerritAndrei V. Ostapenko proposed a change to stackforge/magnetodb: (WIP) "ALL_OLD" return_values for put_item  https://review.openstack.org/12191115:26
*** jeromatron has joined #magnetodb16:24
*** jeromatron has quit IRC16:53
*** jeromatron has joined #magnetodb17:00
*** jeromatron has quit IRC17:02
*** jeromatron has joined #magnetodb17:02
*** jeromatron has quit IRC17:15
*** jeromatron has joined #magnetodb17:16
*** romainh has left #magnetodb17:18
*** jeromatron has quit IRC17:19
*** jeromatron has joined #magnetodb17:21
*** jeromatron has quit IRC17:41
*** jeromatron has joined #magnetodb17:44
openstackgerritDmitriy Ukhlov proposed a change to stackforge/magnetodb: Improve json schema validation for requests left  https://review.openstack.org/12244518:00
*** jeromatron has quit IRC18:03
*** jeromatron has joined #magnetodb18:08
*** jeromatron has quit IRC18:39
*** jeromatron has joined #magnetodb18:42
*** jeromatron has quit IRC18:46
*** jeromatron has joined #magnetodb18:53
*** openstack has joined #magnetodb18:56
*** jeromatron has quit IRC19:16
*** jeromatron has joined #magnetodb19:17
*** jeromatron has quit IRC19:19
*** jeromatron has joined #magnetodb19:25
*** rushiagr_away is now known as rushiagr19:28
*** jeromatron has quit IRC19:30
*** jeromatron has joined #magnetodb19:38
*** rushiagr is now known as rushiagr_away20:05
*** jeromatron has quit IRC20:48
*** jeromatron has joined #magnetodb21:04
*** jeromatron has quit IRC21:12
*** jeromatron has joined #magnetodb21:20
*** jeromatron has quit IRC21:37
*** jeromatron has joined #magnetodb21:44
*** keith_newstadt has joined #magnetodb22:29
*** charlesw has quit IRC22:36
*** jeromatron has quit IRC23:52
*** jeromatron has joined #magnetodb23:58

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!