If you are staring out small one logical/physical cluster is probably
the best and only approach.
Long term this is very case by case dependent but I generally believe
Cluster per Application is the best approach. Although I consider it
Cluster per QOS
For our use cases I find that two
You can consider adding -pr. When iterating through all your hosts
like this. -pr means primary range, and will do less duplicated work.
On Mon, Aug 27, 2012 at 8:05 PM, Aaron Turner synfina...@gmail.com wrote:
I use cron. On one box I just do:
for n in node1 node2 node3 node4 ; do
Setting the metadata will set the validation. If you insert to a
column that is supposed to only INT values Cassandra will reject non
INT data on insert time.
Also comparator can not be changed, you only get once chance to set
the column sorting.
On Tue, Aug 28, 2012 at 3:34 PM, A J
If you move from 7.X to 0.8X or 1.0X you have to rebuild sstables as
soon as possible. If you have large bloomfilters you can hit a bug
where the bloom filters will not work properly.
On Thu, Aug 30, 2012 at 9:44 AM, Илья Шипицин chipits...@gmail.com wrote:
we are running somewhat queue-like
You might want to change the name. There is a node.js driver for
cassandra with the same name. I am not sure which one of your got to
the name first.
On Thu, Sep 6, 2012 at 8:00 PM, aaron morton aa...@thelastpickle.com wrote:
Thanks Tomek,
Feel free to add it to
Try to get Cassandra running the TPH-C benchmarks and beat oracle :)
On Fri, Sep 7, 2012 at 10:01 AM, Hiller, Dean dean.hil...@nrel.gov wrote:
So we wrote 1,000,000 rows into cassandra and ran a simple S-SQL(Scalable
SQL) query of
PARTITIONS n(:partition) SELECT n FROM TABLE as n WHERE
Generally tuning the garbage collector is a waste of time. Just follow
someone else's recommendation and use that.
The problem with tuning is that workloads change then you have to tune
again and again. New garbage collectors come out and you have to tune again
and again. Someone at your company
Haha Ok.
It is not a total waste, but practically your time is better spent in other
places. The problem is just about everything is a moving target, schema,
request rate, hardware. Generally tuning nudges a couple variables in one
direction or the other and you see some decent returns. But each
If you are using ext3 there is a hard limit on number if files in a
directory of 32K. EXT4 as a much higher limit (cant remember exactly
IIRC). So true that having many files is not a problem for the file
system though your VFS cache could be less efficient since you would
have a higher
Hector also offers support for 'Virtual Keyspaces' which you might
want to look at.
On Thu, Sep 27, 2012 at 1:10 PM, Aaron Turner synfina...@gmail.com wrote:
On Thu, Sep 27, 2012 at 3:11 PM, Hiller, Dean dean.hil...@nrel.gov wrote:
We have 1000's of different building devices and we stream
I could use some help as I do not have extensive
experience with these two combinations.
Contact me if you have any other ideas as well.
Edward
On Tue, Jun 26, 2012 at 5:25 PM, Edward Capriolo edlinuxg...@gmail.com wrote:
Hello all,
It has not been very long since the first book was published
I have not looked at this JMX object in a while, however the
compaction manager can support multiple threads. Also it moves from
0-filesize each time it has to compact a set of files.
That is more useful for showing current progress rather then lifetime history.
On Fri, Oct 5, 2012 at 7:27 PM,
Read this:
http://wiki.apache.org/cassandra/FAQ#range_ghosts
Then say this to yourself:
http://cn1.kaboodle.com/img/b/0/0/196/4/C1xHoQAAAZZL9w/ghostbusters-logo-i-aint-afraid-of-no-ghost-pinback-button-1.25-pin-badge.jpg?v=1320511953000
On Sun, Oct 7, 2012 at 4:15 AM, Satoshi Yamada
Java abstracts you from all these problems. One thing to look out for
is JVM options do not work across all JVMs. For example if you try to
enable
https://wikis.oracle.com/display/HotSpotInternals/CompressedOops on a
32bit machine the JVM fails to start.
On Tue, Oct 9, 2012 at 1:45 PM, Brian
Yes. That would be a good jira if it is not already listed. If node is
a seed node autobootstrap and replicate_token settings should trigger
a fatal non-start because your giving c* conflicting directions.
Edward
On Fri, Oct 19, 2012 at 8:49 AM, Thomas van Neerijnen
t...@bossastudios.com wrote:
We have been using cassandra and java7 for months. No problems. A key
concept of java is portable binaries. There are sometimes wrinkles with
upgrades. If you hit one undo the upgrade and restart.
On Tuesday, October 23, 2012, Eric Evans eev...@acunu.com wrote:
On Tue, Oct 16, 2012 at 7:54 PM,
Yes another benchmark with 100,000,000 rows on EC2 machines probably
less powerful then my laptop. The benchmark might as well have run 4
vmware instances on the same desktop.
On Thu, Oct 25, 2012 at 7:40 AM, Brian O'Neill b...@alumni.brown.edu wrote:
People probably saw...
...@gmail.com wrote:
Are you using openJDK or Oracle JDK? I know java7 should be based on openJDK
since 7, but still not sure.
On 25 October 2012 05:42, Edward Capriolo edlinuxg...@gmail.com wrote:
We have been using cassandra and java7 for months. No problems. A key
concept of java is portable
Hello all,
Currently we implement wide rows for most of our entities. For example:
user {
event1=x
event2=y
event3=z
...
}
Normally the entires are bounded to be less then 256 columns and most
columns are small in size say 30 bytes. Because the blind write nature
of Cassandra it is possible
can look at the Netflix client as it makes the co-ordinator node
same as the node which holds that data. This will reduce one hop.
On Thu, Oct 25, 2012 at 9:04 AM, Edward Capriolo edlinuxg...@gmail.com
wrote:
Hello all,
Currently we implement wide rows for most of our entities. For example
If you are using sized teired set minCompactionThreshold to 0 and
maxCompactionThreshold to 0. You can probably also use this
https://issues.apache.org/jira/browse/CASSANDRA-2130
But if you do not compact the number of sstables gets high and then
read performance can suffer.
On Sat, Oct 27,
Using 1.2.0-beta1. I am noticing that there is no longer a single way
to get all the schema. It seems like non-compact storage can be seen
with show schema, but other tables are not visible. Is this by design,
bug, or operator error?
http://pastebin.com/PdSDsdTz
If we create a column family:
CREATE TABLE videos (
videoid uuid,
videoname varchar,
username varchar,
description varchar,
tags varchar,
upload_date timestamp,
PRIMARY KEY (videoid,videoname)
);
The CLI views this column like so:
create column family videos
with column_type =
I see. It is fairly misleading because it is a query that does not
work at scale. This syntax is only helpful if you have less then a few
thousand rows in Cassandra.
On Mon, Nov 5, 2012 at 12:24 PM, Sylvain Lebresne sylv...@datastax.com wrote:
On Mon, Nov 5, 2012 at 4:12 PM, Edward Capriolo
There are no built-in trigger. Someone has written an aspect oriented
piece to do triggers outside of the project.
http://brianoneill.blogspot.com/2012/03/cassandra-triggers-for-indexing-and.html
On Mon, Nov 5, 2012 at 12:30 PM, davuk...@veleri.hr wrote:
Hello!
I was wondering if someone
sylv...@datastax.com wrote:
On Mon, Nov 5, 2012 at 6:55 PM, Edward Capriolo edlinuxg...@gmail.com
wrote:
I see. It is fairly misleading because it is a query that does not
work at scale. This syntax is only helpful if you have less then a few
thousand rows in Cassandra.
Just for the sake
it is better to have one keyspace unless you need to replicate the
keyspaces differently. The main reason for this is that changing
keyspaces requires an RPC operation. Having 10 keyspaces would mean
having 10 connection pools.
On Thu, Nov 8, 2012 at 4:59 PM, sankalp kohli kohlisank...@gmail.com
a keyspace aware connection pool.
Edward
On Thu, Nov 8, 2012 at 5:36 PM, sankalp kohli kohlisank...@gmail.com wrote:
Which connection pool are you talking about?
On Thu, Nov 8, 2012 at 2:19 PM, Edward Capriolo edlinuxg...@gmail.com
wrote:
it is better to have one keyspace unless you need
MessageService has to other nodes. Then there will be incoming connections
via thrift from clients. How are they affected by multiple keyspaces?
On Thu, Nov 8, 2012 at 3:14 PM, Edward Capriolo edlinuxg...@gmail.com
wrote:
Any connection pool. Imagine if you have 10 column families in 10
is from the thrift part. I use hector. In hector, I can
create multiple keyspace objects for each keyspace and use them when I want
to talk to that keyspace. Why will it need to do a round trip to the server
for each switch.
On Thu, Nov 8, 2012 at 3:28 PM, Edward Capriolo edlinuxg...@gmail.com
HBase is different is this regard. A table is comprised of multiple
column families, and they can be scanned at once. However, last time I
checked, scanning a table with two column families is still two
seeks across three different column families.
A similar thing can be accomplished in cassandra
No it does not exist. Rob and I might start a donation page and give
the money to whoever is willing to code it. If someone would write a
tool that would split an sstable into 4 smaller sstables (even an
offline command line tool) I would paypal them a hundo.
On Sat, Nov 10, 2012 at 1:10 PM,
just a note for all. The default partitioner is no longer
randompartitioner. It is now murmur, and the token range starts in negative
numbers. So you don't chose tokens Luke your father taught you anymore.
On Friday, November 9, 2012, Sylvain Lebresne sylv...@datastax.com wrote:
The Cassandra
If you supply metadata cassandra can use it for several things.
1) It validates data on insertion
2) Helps display the information in human readable formats in tools
like the CLI and
sstabletojson
3) If you add a built-in secondary index the type information is
needed, strings sort differently
If you shutdown c* and remove an sstable (and it associated data,
index, bloom filter , and etc) files it is safe. I would delete any
saved caches as well.
It is safe in the sense that Cassandra will start up with no issues,
but you could be missing some data.
On Sun, Nov 11, 2012 at 11:09 PM,
Because you did a major compaction that table is larger then all the
rest. So it will never go away until you have 3 other tables about
that size or you run major compaction again.
You should vote on the ticket:
https://issues.apache.org/jira/browse/CASSANDRA-4766
On Mon, Nov 12, 2012 at 11:51
Yes the row cache could be incorrect so on startup cassandra verify they
saved row cache by re reading. It takes a long time so do not save a big
row cache.
On Tuesday, November 13, 2012, Manu Zhang owenzhang1...@gmail.com wrote:
I have a rowcache provieded by SerializingCacheProvider.
The data
I think the code base does not benefit from having too many different read
code paths. Logically what your suggesting is reasonable, but you have to
consider the case of one being slow to respond.
Then what?
On Tuesday, November 13, 2012, Manu Zhang owenzhang1...@gmail.com wrote:
If consistency
is not big.
On Wed, Nov 14, 2012 at 10:38 AM, Edward Capriolo edlinuxg...@gmail.com
wrote:
Yes the row cache could be incorrect so on startup cassandra verify they
saved row cache by re reading. It takes a long time so do not save a big row
cache.
On Tuesday, November 13, 2012, Manu Zhang
There are several reasons. First there is no absolute offset. The
rows are sorted by the data. If someone inserts new data between your
query and this query the rows have changed.
Unless you doing select queries inside a transaction with repeatable
read and your database supports this the query
. (if the key is
Long, could be more than 1M rows)
Thanks.
-Wei
From: Edward Capriolo edlinuxg...@gmail.com
To: user@cassandra.apache.org
Sent: Tuesday, November 13, 2012 11:13 PM
Subject: Re: unable to read saved rowcache from disk
http://wiki.apache.org
On Thursday, November 15, 2012, Dwight Smith dwight.sm...@genesyslab.com
wrote:
I have a 4 node cluster, version 1.1.2, replication factor of 4,
read/write consistency of 3, level compaction. Several questions.
1) Should nodetool repair be run regularly to assure it has
completed before
We should build an eclipse plugin named Eclipsandra or something.
On Thu, Nov 15, 2012 at 9:45 PM, Wz1975 wz1...@yahoo.com wrote:
Cqlsh is probably the closest you will get. Or pay big bucks to hire someone
to develop one for you:)
Thanks.
-Wei
Sent from my Samsung smartphone on ATT
This was my first question after I git the inserts working. Hive has udfs
like array contains. It also has lateral view syntax that is similar to
transposed.
On Monday, November 19, 2012, Timmy Turner timm.t...@gmail.com wrote:
Is there no option to query for the contents of a collection?
even if you made the calls through cql you would have the same issue since
cql uses thrift. 1.2:0 is supposed to be nicer with concurrent
modifications.
On Monday, November 19, 2012, Everton Lima peitin.inu...@gmail.com wrote:
I was using cassandra direct because it has more performace than
http://www.acunu.com/2/post/2011/12/cql-benchmarking.html
Last I checked, thrift still had an edge over cql due to string
serialization and de serialization. Might be even more dramatic for
later columns. Not that client speed matters much overall in
cassandra's speed, but CQL client does more.
On Tue, Nov 20, 2012 at 5:23 PM, aaron morton aa...@thelastpickle.com wrote:
My understanding of the compaction process was that since data files keep
continuously merging we should not have data files with very old last
modified timestamps
It is perfectly OK to have very old SSTables.
But
I am just taking a stab at this one. UUID's interact with system time and
maybe your real time os is doing something funky there. The other option,
which seems more likely, is that your unit tests are not cleaning up their
data directory and there is some corrupt data in there.
On Tue, Nov 27,
Hector does not require an outdated version of thift, you are likely using
an outdated version of hector.
Here is the long and short of it: If the thrift thrift API changes then
hector can have compatibility issues. This happens from time to time. The
main methods like get() and insert() have
The difference between Replication factor =1 and replication factor 1 is
significant. Also it sounds like your cluster is 2 node so going from RF=1
to RF=2 means double the load on both nodes.
You may want to experiment with the very dangerous column family attribute:
- replicate_on_write:
You can do something like this:
Divide your nodes up into 4 datacenters art1,art2,art3,core
[default@unknown] create keyspace art1 placement_strategy =
'org.apache.cassandra.locator.NetworkTopologyStrategy' and
strategy_options=[{art1:2,core:2}];
[default@unknown] create keyspace art2
'replicate_on_write:
false' fixes the performance issue in our tests.
How dangerous is it? What exactly could go wrong?
On 12-11-27 01:44 PM, Edward Capriolo wrote:
The difference between Replication factor =1 and replication factor 1 is
significant. Also it sounds like your cluster is 2 node so
performance by simply writing to two
separate clusters rather than using single cluster with replicate=2. Which
is kind of stupid :) I think something's fishy with counters and
replication.
Edward Capriolo wrote
I mispoke really. It is not dangerous you just have to understand what it
means
By the way the other issues you are seeing with replicate on write at false
could be because you did not repair. You should do that when changing rf.
On Tuesday, November 27, 2012, Edward Capriolo edlinuxg...@gmail.com
wrote:
Cassandra's counters read on increment. Additionally
in
parallel rather than rely on Cassandra replication.
And yes, Rainbird was the inspiration for what we are trying to do here :)
Edward Capriolo wrote
Cassandra's counters read on increment. Additionally they are distributed
so that can be multiple reads on increment. If they are not fast enough
it couldn't be
done.
When I run the command I get the error
syntax error at position 21: missing EOF at 'placement_strategy'
that is probably because I still need to set the correct properties in
the conf files
On November 27, 2012 at 5:41 PM Edward Capriolo edlinuxg...@gmail.com
wrote
@Bill
Are you saying that now cassandra is less schema less ? :)
Compact storage is the schemaless of old.
On Tuesday, November 27, 2012, Bill de hÓra b...@dehora.net wrote:
I'm not sure I always
understand what people mean by schema less
exactly and I'm curious.
For 'schema less', given
, 2012 at 3:21 PM, Edward Capriolo edlinuxg...@gmail.com
wrote:
I mispoke really. It is not dangerous you just have to understand what it
means. this jira discusses it.
https://issues.apache.org/jira/browse/CASSANDRA-3868
Per Sylvain on the referenced ticket :
I don't disagree about
with Cassandra replication (possibly as
simple as me misconfiguring something) -- it shouldn't be three times
faster
to write to two separate nodes in parallel as compared to writing to
2-node
Cassandra cluster with replication=2.
Edward Capriolo wrote
Say you are doing 100 inserts rf1 on two
Astyanax is a hector fork. You can see many of the hector' authors comments
still in the astyanax code. There is some nice stuff in there but (IMHO) I
do not see the fork as necessary. It has split up the community a bit, as
there are now 3 high level Java clients.
I would advice follow Josh's
Since the cluster name is only cosmetic people do not often change it. I
would not do this in a production cluster for sure.
On Thu, Nov 29, 2012 at 2:56 PM, Wei Zhu wz1...@yahoo.com wrote:
Hi,
I am trying to rename a cluster by following the instruction on Wiki:
Cassandra says ClusterName
Row cache has to store the entire row. It is a very bad option for wide
rows.
On Sunday, December 2, 2012, Mike mthero...@yahoo.com wrote:
Hello,
We recently hit an issue within our Cassandra based application. We
have a relatively new Column Family with some very wide rows (10's of
Rob,
Have you played with this I have many CFs, some big some small some using
large caches some using small ones, some that take many requests, some that
take a few.
Over time I have cooked up a strategy for how to share the cache love, even
thought it may not be the best solution to the
http://wiki.apache.org/cassandra/LargeDataSetConsiderations
On Thu, Dec 6, 2012 at 9:53 AM, Poziombka, Wade L
wade.l.poziom...@intel.com wrote:
“Having so much data on each node is a potential bad day.”
** **
Is this discussed somewhere on the Cassandra documentation (limits,
Good point . hadoop sprays its blocks around randomly. Thus if replication
factor nodes are down some blocks are not found. The larger the cluster the
higher chance nodes are down.
To deal with this increase rf once the cluster gets to be very large.
On Wednesday, December 5, 2012, Eric Parusel
Assuming you need to work with quorum in a non-vnode scenario. That means
that if 2 nodes in a row in the ring are down some number of quorum
operations will fail with UnavailableException (TimeoutException right
after the failures). This is because the for a given range of tokens quorum
will be
Until the secondary indexes do not read before write is in a release and
stabilized you should follow Ed ENuff s blog and do your indexing yourself
with composites.
On Thursday, December 13, 2012, aaron morton aa...@thelastpickle.com
wrote:
The IndexClause for the get_indexed_slices takes a
It should be good stuff. Brian eats this stuff for lunch.
On Wednesday, December 12, 2012, Brian O'Neill b...@alumni.brown.edu
wrote:
FWIW --
I'm presenting tomorrow for the Datastax C*ollege Credit Webinar Series:
This issue has to be looked from a micro and macro level. On the microlevel
the best way is workload specific. On the macro level this mostly boils
down to data and memory size.
Companions are going to churn cache, this is unavoidable. Imho solid state
makes the micro optimization meanless in the
Here is a good start.
http://www.anuff.com/2011/02/indexing-in-cassandra.html
On Thu, Dec 13, 2012 at 11:35 AM, Alain RODRIGUEZ arodr...@gmail.comwrote:
Hi Edward, can you share the link to this blog ?
Alain
2012/12/13 Edward Capriolo edlinuxg...@gmail.com
Ed ENuff s
Is there a way to turn this on and off through configuration? I am not
necessarily sure I would want this feature. Also it is confusing if these
writes show up in JMX and look like user generated write operations.
On Mon, Dec 17, 2012 at 10:01 AM, Mike mthero...@yahoo.com wrote:
Thank you
CQL2 and CQL3 indexes are not compatible. I guess CQL2 is able to detect
that the table was defined in CQL3 probably should not allow it. Backwards
comparability is something the storage engines and interfaces have to
account for. At least they should prevent you from hurting yourself.
But do not
In the TCP mib for SNMP (Simple Network Management Protocol) this
information is available
http://www.simpleweb.org/ietf/mibs/mibSynHiLite.php?category=IETFmodule=TCP-MIB
On Wed, Dec 19, 2012 at 12:22 AM, Michael Kjellman
mkjell...@barracuda.comwrote:
netstat + cron is your friend at this
The cli using microsecond precision your client might be using something
else and the insert with lower timestamps are dropped.
On Friday, December 21, 2012, Qiaobing Xie qiaobing@gmail.com wrote:
Hi,
I am developing a thrift client that inserts and removes columns from a
column-family
You could store the order as the first part of a composite string say first
picture as A and second as B. To insert one between call it AA. If you
shuffle alot the strings could get really long.
Might be better to store the order in a separate column.
Neither solution mentioned deals with
This what versions are supported is kinda up to you for example earlier
versions of jdk now have bugs. I have a version of java 1.6.0_23 I believe
that will not even start with the latest cassandra releases. Likewise
people suggest not running the newest ones 1.7.0 because they have not
tested it.
Unfortunately one of the first command everyone needs to use to use to work
with cassandra changes very often.
You can use
cqlsh help create_keyspace;
But some times even the documentation is not in line.
Using this permutation of goodness:
cqlsh 2.3.0 | Cassandra 1.2.0-beta2-SNAPSHOT | CQL
There is a crazy, very bad, don't do it way to do this. You can set RF=1
and hack the LocalPartitioner (because the local partitioner has been
made not to do this)
Then the node you connect to and write is the node the data will get stored
on.
Its like memcache do it yourself style sharding.
By the way 10% faster does not necessarily mean 10% more requests.
https://issues.apache.org/jira/browse/CASSANDRA-2975
https://issues.apache.org/jira/browse/CASSANDRA-3772
Also if you follow the tickets My tests show that Murmur3Partitioner
actually is worse than MD5 with high cardinality
Just a shot in the dark, but I would try setting -Xss higher then the
default. It's probably like 180, but I cant even start at that level,
bumped it up to 256 for JDK 7.
On Thu, Jan 3, 2013 at 12:02 PM, Michael Kjellman
mkjell...@barracuda.comwrote:
:) yes, I'm crazy
The assertion appears to
been fixed in 1.1.7 ??
From: Edward Capriolo edlinuxg...@gmail.com
Reply-To: user@cassandra.apache.org user@cassandra.apache.org
Date: Thursday, January 3, 2013 11:57 AM
To: user@cassandra.apache.org user@cassandra.apache.org
Subject: Re: Error after 1.2.0 upgrade
There is a bug
Yes. They were really just introduced and if you are ready to hitch your
wagon to every new feature you put yourself in considerable risk. With any
piece of software not just Cassandra.
On Fri, Jan 4, 2013 at 11:59 AM, Alain RODRIGUEZ arodr...@gmail.com wrote:
But I don't really get the point
There is some point where you simply need more machines.
On Mon, Jan 7, 2013 at 5:02 PM, Michael Kjellman mkjell...@barracuda.comwrote:
Right, I guess I'm saying that you should try loading your data with
leveled compaction and see how your compaction load is.
Your work load sounds like
at 7:27 AM, DE VITO Dominique
dominique.dev...@thalesgroup.com wrote:
Hi,
Edward Capriolo described in his Cassandra book a faster way [1] to start
new nodes if the cluster size doubles, from N to 2 *N.
It's about splitting in 2 parts each token range taken in charge, after
the split
to
do it this way anymore
I guess it's true in v1.2.
Is it true also in v1.1 ?
Thanks.
Dominique
*De :* Edward Capriolo [mailto:edlinuxg...@gmail.com]
*Envoyé :* mardi 8 janvier 2013 16:01
*À :* user@cassandra.apache.org
*Objet :* Re: about validity of recipe A node join using
I ask myself this every day. CQL3 is new way to do things, including wide
rows with collections. There is no upgrade path. You adopt CQL3's sparse
tables as soon as you start creating column families from CQL. There is not
much backwards compatibility. CQL3 can query compact tables, but you may
By no upgrade path I mean to say if I have a table with compact storage I
can not upgrade it to sparse storage. If i have an existing COMPACT table
and I want to add a Map to it, I can not. This is what I mean by no upgrade
path.
Column families that mix static and dynamic columns are pretty
, that do not
bother me anyway.
4 are these sparse columns also taking memtable space?
This questions would give me serious pause to use sparse tables
On Wednesday, January 9, 2013, Edward Capriolo edlinuxg...@gmail.com
wrote:
By no upgrade path I mean to say if I have a table with compact storage
I think 1.6.0_24 is too low and 1.7.0 is too high. Try a more recent 1.6.
I just had problems with 1.6.0_23 see here:
https://issues.apache.org/jira/browse/CASSANDRA-4944
On Thu, Jan 10, 2013 at 10:29 AM, Sloot, Hans-Peter
hans-peter.sl...@atos.net wrote:
I have 4 vm's with 1024M memory.
1
You have to change the column family cache info from keys_only to otherwise
the cache will not br on for this cf.
On Wednesday, January 16, 2013, Brian Tarbox tar...@cabotresearch.com
wrote:
We have quite wide rows and do a lot of concentrated processing on each
row...so I thought I'd try the
I think at this point cassandra startup scripts should reject versions
since cassandra won't even star with many jvms at this point.
On Tuesday, January 15, 2013, Michael Kjellman mkjell...@barracuda.com
wrote:
Do yourself a favor and get a copy of the Oracle 7 JDK (now with more
security
If you have 40ms NTP drift something is VERY VERY wrong. You should have a
local NTP server on the same subnet, do not try to use one on the moon.
On Thu, Jan 17, 2013 at 4:42 AM, Sylvain Lebresne sylv...@datastax.comwrote:
So what I want is, Cassandra provide some information for client, to
Wow you managed to do a load test through the cassandra-cli. There should
be a merit badge for that.
You should use the built in stress tool or YCSB.
The CLI has to do much more string conversion then a normal client would
and it is not built for performance. You will definitely get better
You can not be /mostly/ consistent readlike you can not be half-pregnant
or half transactional. You either are or you are not.
If you do not have enough nodes for a QUORUM the read fails. Thus you never
get stale reads you only get failed reads.
The dynamic snitch makes reads sticky at READ.ONE.
This was described in good detail here:
http://thelastpickle.com/2011/04/28/Forces-of-Write-and-Read/
On Tue, Jan 22, 2013 at 9:41 AM, Brian Tarbox tar...@cabotresearch.comwrote:
Thank you! Since this is a very non-standard way to display data it
might be worth a better explanation in the
By default Cassandra uses 1/3rd heap size for memtable storage. If you make
sure memtables smaller they should flush faster and you commit logs should
not grow large.
Large commit logs are not a problem, some use cases that write to some
Column Families more then other can make the commit log
1. The commit log is only read on startup.
W: If writes are unflushed then the commit logs need to be replayed
2: shrink the memtable settings.
but you dont want to do this.
3. Commit log size is not directly related to sstable size.
E.g. if you write the same row a billion times the commit log
Make sure the timestamp on your delete is then timestamp of the data.
On Sat, Jan 26, 2013 at 1:33 PM, Kasun Weranga kas...@wso2.com wrote:
Hi all,
When I delete some rowkeys programmatically I can see two rowkeys remains
in the column family. I think it is due to tombstones. Is there a way
One technique is on the client side you build a tool that takes the even
and produces N mutations. In c* writes are cheap so essentially, re-write
everything on all changes.
On Sun, Jan 27, 2013 at 4:03 PM, Fredrik Stigbäck
fredrik.l.stigb...@sitevision.se wrote:
Hi.
Since denormalized data
LOVE
the performance of our ACL checks.
Ps. 30,000 writes in cassandra is not cheap when done from one server ;)
but in general parallized writes is very fast for like 500.
Later,
Dean
From: Edward Capriolo edlinuxg...@gmail.commailto:edlinuxg...@gmail.com
Reply-To: user
1 - 100 of 742 matches
Mail list logo
