You have very large heap,it’s take most of cpu time in GC stage.you should
in maximum set heap on 12GB and enable row cache to your cluster become
faster.
On Friday, 24 August 2018, Mun Dega wrote:
> 120G data
> 28G heap out of 48 on system
> 9 node cluster, RF3
>
>
> On Thu, Aug 23, 2018, 17:1
Thanks! Great tips on clearing tombstones. The TTL vs. business rules challenge
is one we’ve seen in enterprises moving from relational to non-relational
because there is no thought to planning a data retention policy.
Periodic business rules based cleaning via Spark works well if you use it to
No what I meant by infinite partition is not auto sub-partitioning, even at
server-side. Ideally Cassandra should be able to support infinite partition
size and make compaction, repair and streaming of such partitions
manageable:
- compaction: find a way to iterate super efficiently through the wh
Hi,
I'm using cassandra 3.11
When I drop a keyspace it's data is not deleted from data dirs in a cluster.
what additional steps are needed to make cluster nodes to deleted
deleted data from the disk?
Regards,
Vitaliy
-
To unsubs
Hi,
I read that once Facebook chose HBase over Cassandra for it's messenger,
but I never found what are the benefits for HBase over Cassandra,
can someone list, if there are any?
Regards,
Vitaliy
-
To unsubscribe, e-mail: user-u
Hi,
I'm using cassandra-unit for integration tests,
which is using regular cassandra.yaml to create a cassandra instance.
What parameters are recommended to be changed in order to speed up
startup process.
Regards
Vitaliy
-
To
Hi,
i'm running count query for a very small table (less than 1000 000 records).
When the amount of records gets to 800 000 i receive read timeout
error in cqlsh.
I tried to run cqlsh with option --request-timeout=3600, but receive same error,
what should I do in order not to recieve timeout excep
You should change read_request_timeout in cassandra.yaml file.
Default is 5 sec
But it is not recommended to do count in cassandra better if u can avoid it
On Fri, Aug 24, 2018, 4:06 PM Vitaliy Semochkin
wrote:
> Hi,
>
> i'm running count query for a very small table (less than 1000 000
> rec
HI,
While using DataStax driver
session.execute("some insert
query")getExecutionInfo().getAchievedConsistencyLevel()
is already returned as null, despite data is stored. Why could it be?
Is it possible to make DataStax driver throw an exception in case
desired consistency level was not achieved d
It takes a while in cluster for drop to propagte this depends on
amount of data and network traffic between your storage nodes
On Fri, Aug 24, 2018 at 1:54 PM Vitaliy Semochkin wrote:
>
> Hi,
> I'm using cassandra 3.11
> When I drop a keyspace it's data is not deleted from data dirs in a cluster.
Thank you for the fast replay, Pranay!
This is testing environment and using count on it will do no harm.
On Sat, Aug 25, 2018 at 12:11 AM Pranay akula
wrote:
>
> You should change read_request_timeout in cassandra.yaml file.
>
> Default is 5 sec
>
> But it is not recommended to do count in cass
Cassandra creates a snapshot when u drop keyspace. So u should run nodetool
clear snapshot on all nodes to reclaim ur space.
On Fri, Aug 24, 2018, 4:14 PM Vineet G H wrote:
> It takes a while in cluster for drop to propagte this depends on
> amount of data and network traffic between your stor
I've worked with both databases. They're suitable for different use-cases. If
you look at the CAP theorem; HBase is CP while Cassandra is a AP. If we talk
about a specific use-case, it'll be easier to discuss.
Dinesh
On Friday, August 24, 2018, 1:56:31 PM PDT, Vitaliy Semochkin
wrote:
At the time that Facebook chose HBase, Cassandra was drastically less
mature than it is now and I think the original creators had already left.
There were already various Hadoop variants running for data analytics etc,
so lots of operational and engineering experience around it available. So,
prob
Thank you very much for the fast reply, Vineet!
Is there any way to speed up this process or manually trigger
something analogs to vacuum full in PostgreSQL?
On Sat, Aug 25, 2018 at 12:14 AM Vineet G H wrote:
>
> It takes a while in cluster for drop to propagte this depends on
> amount of data an
Thank you very much for fast reply, Dinesh!
I was under impression that with tunable consistency Cassandra can
act as CP (in case it is needed), e.g by setting ALL on both reads
and writes.
Do you agree with this statement?
PS Are there any other benefits of HBase you have found? I'd be glad
to
Thank you very much Pranay, that was exactly what I needed!
On Sat, Aug 25, 2018 at 12:17 AM Pranay akula
wrote:
>
> Cassandra creates a snapshot when u drop keyspace. So u should run nodetool
> clear snapshot on all nodes to reclaim ur space.
>
>
>
> On Fri, Aug 24, 2018, 4:14 PM Vineet G H wro
Tobias,
This is very interesting. Can I inquire a bit more on why you have both C*
and Kudu in the system?
Wouldnt keeping just Kudu work (that was its initial purpose?). Is there
something to do with its production readiness? I ask as we have a similar
concern as well.
Finally, how are your das
Spark can read hdfs directly so locality is important but Spark can't read
Cassandra data directly it can only connect by api. So I think you don't
need to install them a same node
2018년 8월 25일 (토) 오후 3:16, Affan Syed 님이 작성:
> Tobias,
>
> This is very interesting. Can I inquire a bit more on why
Nope, Spark cassandra connector leverages data locality and get tremendous
improvements due to localitty.
- Affan
On Sat, Aug 25, 2018 at 11:25 AM CharSyam wrote:
> Spark can read hdfs directly so locality is important but Spark can't read
> Cassandra data directly it can only connect by api.
20 matches
Mail list logo
