Thanks Bowen,
* "How do you split?"
challenging to answer short, but let me try: physical host has cores
from idx 0 - 11 (6 physical and 6 virtual in pairs - they are in
pairs as 0,6 belongs together, then 1,7 and then 2,8 and so on)
What we do is that in the virt-install command we use --cpu
host-passthrough --cpuset={{virtinst_cpu_set}} --vcpus=6
where {{virtinst_cpu_set}} is
- 0,6,1,7,2,8 - for CassandraVM
- 3,9,4,10,5,11 - for the other VM
(we split the physical host into 2 VMs)
* "do you expose physical disks to the VM or use disk image files"
no images, physical host has 2 spinning disks and 1 SSD drive
CassandraVM gets assigned explicitly 1 of the spinning disks and she
also gets assigned a partition of the SSD (which is used for commit
logs only so that is separated from the data)
* "A 50-70% utilization of a 1 Gbps network interface on average
doesn't sound good at all."
Yes, this is weird... Especially because e.g. if we bring down a
node, the other 2 nodes (we go with RF=2) are producing ~600Mb hints
files / minute
And assuming hint files is basicall the saved "network traffic"
until node is down this would still just give 10Mb/sec ...
OK, these are just the replicated updates, there is also read and of
course App layer is also reading but even with that in mind it does
not add up... So we will try to do further analysis here
Thanks for the article also regarding the Counter tables!
Actually we already know for a while there are "interesting" things
going around the Counter tables it is surprising how difficult to find
info regarding this topic...
I personally tried to look around here several times and always just
getting the same and the same information in posts...
Moving away from counters would not be bad especially because of the
difficulties around DELETEing (we also feel it) them however I do not
see any obvious migration strategy here...
But maybe let me ask this in a separate question. Might make more
sense... :-)
Thanks again - and thanks to others as well
It looks mastering the "nodetool tpstats" and the Cassandra thread pools
would worth some time... :-)
Attila Wind
http://www.linkedin.com/in/attilaw <http://www.linkedin.com/in/attilaw>
Mobile: +49 176 43556932
06.03.2021 13:03 keltezéssel, Bowen Song írta:
Hi Attila,
Addressing your data modelling issue is definitely important, and this
alone may be enough to solve all the issues you have with Cassandra.
* "Since these are VMs, is there any chance they are competing for
resources on the same physical host?"
We are splitting the physical hardware into 2 VMs - and resources
(cpu cores, disks, ram) all assigned in a dedicated fashion to the
VMs without intersection
How do you split? Number of cores in all VMs sums to the total
physical CPU cores is not enough, because context switches and
possible thread contentions will waste CPU cycles. Since you have also
said 8-12% CPU time is spent in sys mode, I think it warrants an
investigation.
Also, do you expose physical disks to the VM or use disk image files?
Disk image files can be slow, especially for high IOPS random reads.
Personally, I won't recommend running a database on a VM other than
for dev/testing/etc. purposes. If possible, you should try to add a
node running on a bare metal server of the similar spec as the VM, and
see if there's any noticeable performance differences between this
bare metal node and the VM nodes.
* The bandwidth limit is 1Gbit/sec (so 120Mb/sec) BUT it is the
limit of the physical host - so our 2 VMs competing here. Possible
that Cassandra VM has ~50-70% of it...
A 50-70% utilization of a 1 Gbps network interface on average doesn't
sound good at all. That over 60MB/s network traffic constantly. Can
you investigate why is this happening? Do you really read/write that
much? Or is it something else?
* "nodetool tpstats"
whooa I never used it, we definitely need some learning here to
even understand the output... :-) But I copy that here to the
bottom ... maybe clearly shows something to someone who can read it...
I noticed that you are using counters in Cassandra. I have to say that
I haven't had a good experience with Cassandra counters. An article
<https://ably.com/blog/cassandra-counter-columns-nice-in-theory-hazardous-in-practice>
which I read recently may convince you to get rid of it. I also don't
think counter is something the Cassandra developers are focused on,
because things like CASSANDRA-6506
<https://issues.apache.org/jira/browse/CASSANDRA-6506> have been
sitting there for many years.
Use your database software for their strengths, not their weaknesses.
You have Cassandra, but you don't have to use every feature in
Cassandra. Sometimes another technology may be more suitable for
something that Cassandra can do but isn't very good at.
Cheers,
Bowen
On 05/03/2021 18:37, Attila Wind wrote:
Thanks for the answers @Sean and @Bowen !!!
First of all, this article described very similar thing we experience
- let me share
https://www.senticore.com/overcoming-cassandra-write-performance-problems/
we are studying that now
Furthermore
* yes, we have some level of unbalanced data which needs to be
improved - this is on our backlog so should be done
* and yes we do see clearly that this unbalanced data is slowing
down everything in Cassandra (there is proof of it in our
Prometheus+Grafana based monitoring)
* we will do this optimization now definitely (luckily we have plan
already)
@Sean:
* "Since these are VMs, is there any chance they are competing for
resources on the same physical host?"
We are splitting the physical hardware into 2 VMs - and resources
(cpu cores, disks, ram) all assigned in a dedicated fashion to
the VMs without intersection
BUT!!
You are right... There is one thing we are sharing: network
bandwidth... and actually that one does not come up in the
"iowait" part for sure. We will further analyze into this
direction definitely because from the monitoring as far as I see
yeppp, we might hit the wall here
* consistency level: we are using LOCAL_ONE
* "Does the app use prepared statements that are only prepared once
per app invocation?"
Yes and yes :-)
* "Any LWT/”if exists” in your code?"
No. We go with RF=2 so we even can not use this (as LWT goes with
QUORUM and in our case this would mean we could not tolerate
losing a node... not good... so no)
@Bowen:
* The bandwidth limit is 1Gbit/sec (so 120Mb/sec) BUT it is the
limit of the physical host - so our 2 VMs competing here.
Possible that Cassandra VM has ~50-70% of it...
* The CPU's "system" value shows 8-12%
* "nodetool tpstats"
whooa I never used it, we definitely need some learning here to
even understand the output... :-) But I copy that here to the
bottom ... maybe clearly shows something to someone who can read
it...
so, "nodetool tpstats" from one of the nodes
Pool Name Active Pending Completed Blocked All time blocked
ReadStage 0 0 6248406
0 0
CompactionExecutor 0 0 168525
0 0
MutationStage 0 0 25116817
0 0
MemtableReclaimMemory 0 0 17636
0 0
PendingRangeCalculator 0 0 7
0 0
GossipStage 0 0 324388
0 0
SecondaryIndexManagement 0 0 0
0 0
HintsDispatcher 1 0 75
0 0
Repair-Task 0 0 1
0 0
RequestResponseStage 0 0 31186150
0 0
Native-Transport-Requests 0 0 22827219
0 0
CounterMutationStage 0 0 12560992
0 0
MemtablePostFlush 0 0 19259
0 0
PerDiskMemtableFlushWriter_0 0 0 17636
0 0
ValidationExecutor 0 0 48
0 0
Sampler 0 0 0
0 0
ViewBuildExecutor 0 0 0
0 0
MemtableFlushWriter 0 0 17636
0 0
InternalResponseStage 0 0 44658
0 0
AntiEntropyStage 0 0 161
0 0
CacheCleanupExecutor 0 0 0
0 0
Message type Dropped Latency waiting in
queue (micros)
50% 95% 99% Max
READ_RSP 18 1629.72 8409.01
155469.30 386857.37
RANGE_REQ 0 0.00 0.00
0.00 0.00
PING_REQ 0 0.00 0.00
0.00 0.00
_SAMPLE 0 0.00 0.00
0.00 0.00
VALIDATION_RSP 0 0.00 0.00
0.00 0.00
SCHEMA_PULL_RSP 0 0.00 0.00
0.00 0.00
SYNC_RSP 0 0.00 0.00
0.00 0.00
SCHEMA_VERSION_REQ 0 0.00 0.00
0.00 0.00
HINT_RSP 0 943.13 3379.39
5839.59 52066.35
BATCH_REMOVE_RSP 0 0.00 0.00
0.00 0.00
PAXOS_COMMIT_REQ 0 0.00 0.00
0.00 0.00
SNAPSHOT_RSP 0 0.00 0.00
0.00 0.00
COUNTER_MUTATION_REQ 94 1358.10 5839.59
14530.76 464228.84
GOSSIP_DIGEST_SYN 0 1358.10 5839.59
25109.16 25109.16
PAXOS_PREPARE_REQ 0 0.00 0.00
0.00 0.00
PREPARE_MSG 0 0.00 0.00
0.00 0.00
PAXOS_COMMIT_RSP 0 0.00 0.00
0.00 0.00
HINT_REQ 0 0.00 0.00
0.00 0.00
BATCH_REMOVE_REQ 0 0.00 0.00
0.00 0.00
STATUS_RSP 0 0.00 0.00
0.00 0.00
READ_REPAIR_RSP 0 0.00 0.00
0.00 0.00
GOSSIP_DIGEST_ACK2 0 1131.75 5839.59
7007.51 7007.51
CLEANUP_MSG 0 0.00 0.00
0.00 0.00
REQUEST_RSP 0 0.00 0.00
0.00 0.00
TRUNCATE_RSP 0 0.00 0.00
0.00 0.00
REPLICATION_DONE_RSP 0 0.00 0.00
0.00 0.00
SNAPSHOT_REQ 0 0.00 0.00
0.00 0.00
ECHO_REQ 0 0.00 0.00
0.00 0.00
PREPARE_CONSISTENT_REQ 0 0.00 0.00
0.00 0.00
FAILURE_RSP 9 0.00 0.00
0.00 0.00
BATCH_STORE_RSP 0 0.00 0.00
0.00 0.00
SCHEMA_PUSH_RSP 0 0.00 0.00
0.00 0.00
MUTATION_RSP 17 1131.75 4866.32
8409.01 464228.84
FINALIZE_PROPOSE_MSG 0 0.00 0.00
0.00 0.00
ECHO_RSP 0 0.00 0.00
0.00 0.00
INTERNAL_RSP 0 0.00 0.00
0.00 0.00
FAILED_SESSION_MSG 0 0.00 0.00
0.00 0.00
_TRACE 0 0.00 0.00
0.00 0.00
SCHEMA_VERSION_RSP 0 0.00 0.00
0.00 0.00
FINALIZE_COMMIT_MSG 0 0.00 0.00
0.00 0.00
SNAPSHOT_MSG 0 0.00 0.00
0.00 0.00
PREPARE_CONSISTENT_RSP 0 0.00 0.00
0.00 0.00
PAXOS_PROPOSE_REQ 0 0.00 0.00
0.00 0.00
PAXOS_PREPARE_RSP 0 0.00 0.00
0.00 0.00
MUTATION_REQ 265 1358.10 5839.59
223875.79 802187.44
READ_REQ 45 1629.72 5839.59
36157.19 386857.37
PING_RSP 0 0.00 0.00
0.00 0.00
RANGE_RSP 0 0.00 0.00
0.00 0.00
VALIDATION_REQ 0 0.00 0.00
0.00 0.00
SYNC_REQ 0 0.00 0.00
0.00 0.00
_TEST_1 0 0.00 0.00
0.00 0.00
GOSSIP_SHUTDOWN 0 0.00 0.00
0.00 0.00
TRUNCATE_REQ 0 0.00 0.00
0.00 0.00
_TEST_2 0 0.00 0.00
0.00 0.00
GOSSIP_DIGEST_ACK 0 1629.72 5839.59
43388.63 43388.63
SCHEMA_PUSH_REQ 0 0.00 0.00
0.00 0.00
FINALIZE_PROMISE_MSG 0 0.00 0.00
0.00 0.00
BATCH_STORE_REQ 0 0.00 0.00
0.00 0.00
COUNTER_MUTATION_RSP 96 1358.10 4866.32
8409.01 464228.84
REPAIR_RSP 0 0.00 0.00
0.00 0.00
STATUS_REQ 0 0.00 0.00
0.00 0.00
SCHEMA_PULL_REQ 0 0.00 0.00
0.00 0.00
READ_REPAIR_REQ 0 0.00 0.00
0.00 0.00
ASYMMETRIC_SYNC_REQ 0 0.00 0.00
0.00 0.00
REPLICATION_DONE_REQ 0 0.00 0.00
0.00 0.00
PAXOS_PROPOSE_RSP 0 0.00 0.00
0.00 0.00
Attila Wind
http://www.linkedin.com/in/attilaw <http://www.linkedin.com/in/attilaw>
Mobile: +49 176 43556932
05.03.2021 17:45 keltezéssel, Bowen Song írta:
Based on my personal experience, the combination of slow read
queries and low CPU usage is often an indicator of bad table schema
design (e.g.: large partitions) or bad query (e.g. without partition
key). Check the Cassandra logs first, is there any long
stop-the-world GC? tombstone warning? anything else that's out of
ordinary? Check the output from "nodetool tpstats", is there any
pending or blocked tasks? Which thread pool(s) are they in? Is there
a high number of dropped messages? If you can't find anything useful
from the Cassandra server logs and "nodetool tpstats", try to get a
few slow queries from your application's log, and run them manually
in the cqlsh. Are the results very large? How long do they take?
Regarding some of your observations:
/> CPU load is around 20-25% - so we have lots of spare capacity/
Is it very few threads each uses nearly 100% of a CPU core? If so,
what are those threads? (I find the ttop command from the sjk tool
<https://github.com/aragozin/jvm-tools> very helpful)
/> network load is around 50% of the full available bandwidth/
This sounds alarming to me. May I ask what's the full available
bandwidth? Do you have a lots of CPU time spent in sys (vs user) mode?
On 05/03/2021 14:48, Attila Wind wrote:
Hi guys,
I have a DevOps related question - hope someone here could give
some ideas/pointers...
We are running a 3 nodes Cassandra cluster
Recently we realized we do have performance issues. And based on
investigation we took it seems our bottleneck is the Cassandra
cluster. The application layer is waiting a lot for Cassandra ops.
So queries are running slow on Cassandra side however due to our
monitoring it looks the Cassandra servers still have lots of free
resources...
The Cassandra machines are virtual machines (we do own the physical
hosts too) built with kvm - with 6 CPU cores (3 physical) and 32GB
RAM dedicated to it.
We are using Ubuntu Linux 18.04 distro - everywhere the same
version (the physical and virtual host)
We are running Cassandra 4.0-alpha4
What we see is
* CPU load is around 20-25% - so we have lots of spare capacity
* iowait is around 2-5% - so disk bandwidth should be fine
* network load is around 50% of the full available bandwidth
* loadavg is max around 4 - 4.5 but typically around 3 (because
of the cpu count 6 should represent 100% load)
and still, query performance is slow ... and we do not understand
what could hold Cassandra back to fully utilize the server resources...
We are clearly missing something!
Anyone any idea / tip?
thanks!
--
Attila Wind
http://www.linkedin.com/in/attilaw <http://www.linkedin.com/in/attilaw>
Mobile: +49 176 43556932