THIS! A well-reasoned and clear explanation of a very difficult topic. This is 
the kind of gold that a user mailing list can provide. Thank you, Alain!

Sean Durity

From: Alain RODRIGUEZ <>
Sent: Tuesday, July 03, 2018 6:37 AM
To: user <>
Subject: [EXTERNAL] Re: JVM Heap erratic

Hello Randy,

It's normal that the memory in the heap is having this pattern. Java uses 
memory available and when needed clean some memory for new needs, that's the 
variation you see. In your case, it's not really regular but this can depend on 
the workload as well.

I'm a C# .NET guy, so I have no idea if this is normal Java behavior.

I feel you. I started operating Cassandra with no clue about the Garbage 
collection and other JVM stuff. When I started tuning it the first time with 
some former colleagues, we ended up removing half of the nodes of the cluster 
and still divided latency per 2. It is an important part of Cassandra to tune 
and often people (including myself) overlook it because it's too complex. I'll 
try to give you a big picture so you can have some analysis of what's going on 
and hopefully do some good to this cluster ("some good" - maybe not remove half 
of the nodes and reduce the latency, this was really a strong improvement on a 
badly tuned GC, but let's see :) ).

The heap is a limited amount of memory used to store Java objects. It's 
composed of 3 sections: The New Generation, The Old Generation, the Permananent 
Generation. New objects go to the New Gen ('HEAP_NEW_SIZE' in CMS, auto in G1GC 
- do not set). From time to time, depending on usage and tuning, surviving 
objects are pushed from the Eden Space where they first land to one of the 2 
survivor space (the other one is empty). Then, depending on the tenuring 
threshold option (in CMS, auto in G1GC too I believe), the data will be passed 
from one survivor to the other one, expiring old data in the process. This 
cleaning process in the New Gen is called the minor garbage collection (Minor 
GC) and is triggered when Eden is full. After the tenuring threshold is reached 
and the object was moved around survivor spaces x times, surviving objects will 
be promoted (or tenured) to the Old Gen. This promotion of living objects is 
referenced as a Major GC.
This is the most expensive GC, and even though it will have to happen from time 
to time in almost all cases, it's interesting to reduce the total duration and 
frequency of Major GC to improve GC statistics overall. We can ignore the 
permanent Gen that is not triggering any important GC activity.

Some more information is available here:<>

In Cassandra, especially, in read-heavy workloads, objects can often expire 
before being promoted if given enough space and time to do so. And this is way 
more performant than promoting objects because we hadn't them surviving long 
enough in the New Gen.

Using CMS with 20 GB is not recommended (out of the box, as a starting point at 
least) because CMS performances are known to degrade quickly with bigger heap 
than 8 GB. 20 GB is a lot. It also depends on the total memory available.

tried 8GB = OOM
tried 12GB = OOM
tried 20GB w/ G1 = OOM (and long GC pauses usually over 2 secs)
tried 20GB w/ CMS = running

OOM are not only related to the space available but also to the impossibility 
to clean the heap efficiently enough before we need the space. Thus tuning some 
more option than just the heap size might help.

CMS (over G1CG)
HEAP: 8 to 16 GB.
NEW_HEAP: 25 to 50 % - nothing to do with CPU core contrary to 
documentation/comments in the file imho
MaxTenuringThreshold: 15 - From 1 all the way up to 15, that's what gave me the 
best results in the past, it reduces major GC and makes the most of New 
Gen/minor GC, that are less impacting, but still "stop the world GC". Default 
is 1, which is often way to short to expire objects...
SurvarorRatio: 2 to 8 - controls survivor spaces size. It will be: 'Survivor 
total space = New Gen Size / (SurvarorRatio + 2)'. Dividing by 2 you have the 
size of each survivor. Here it will depend how fast the Eden space is 
allocated. Increasing the survivor space will disminuish the Eden space (where 
new objects are allocated) and there is a tradeoff here as well and a balance 
to find.

I would try with these settings on a canary node:
HEAP - 16 GB (if read heavy, if not probably between 8 and 12 GB is better).
NEW_HEAP - 50% of the heap (4 - 8GB)
MaxTenuringThreshold: 15
SurvarorRatio: 4,

When testing GC, there is not a better way than using a canary node, pick one 
rack and node(s) you want in this rack to test. This should not impact 
availability or consistency. If you're able to reproduce the workload perfectly 
in the staging cluster, it's perfect but I don't know much companies/people 
able to do this and the use of a canary node should be safe :).
I could probably share some thoughts on what the cluster really needs, rather 
than making guesses and suggesting a somewhat arbitrary tuning, as I did above 
if you would share a gc.log file with us from one of the nodes. Garbage 
Collection tuning is a bit tricky, but a good tuning can divide latency while 
cutting in the number of host. I have seen impressives changes in the past 
There is a lot of details in this log file about where is the biggest pressure, 
the allocation rate, the GC duration distribution for each type of GC, etc. 
With this, I could see where the pressure is and suggest how to work on it.

Be aware that extra GC is also sometimes the consequence (and not a cause) of 
an issue. Due to pending requests, wide partitions, ongoing compactions, 
repairs or an intensive workload, GC can pressure can increase and mask another 
underlying, and root issue. You might want to check that the cluster is healthy 
other than GC, as a lot of distinct internal parts of Cassandra have an impact 
on the GC.

Hope that helps,

Alain Rodriguez - @arodream -<>
France / Spain

The Last Pickle - Apache Cassandra Consulting<>

2018-06-28 23:19 GMT+01:00 Elliott Sims 
Odd.  Your "post-GC" heap level seems a lot lower than your max, which implies 
that you should be OK with ~10GB.  I'm guessing either you're genuinely getting 
a huge surge in needed heap and running out, or it's falling behind and garbage 
is building up.  If the latter, there might be some tweaking you can do.  
Probably worth turning on GC logging and digging through exactly what's 

CMS is kind of hard to tune and can have problems with heap fragmentation since 
it doesn't compact, but if it's working for you I'd say stick with it.

On Thu, Jun 28, 2018 at 3:14 PM, Randy Lynn 
<<>> wrote:
Thanks for the feedback..

Getting tons of OOM lately..

You mentioned overprovisioned heap size... well...
tried 8GB = OOM
tried 12GB = OOM
tried 20GB w/ G1 = OOM (and long GC pauses usually over 2 secs)
tried 20GB w/ CMS = running

we're java 8 update 151.

We've got one table that's got a 400MB partition.. that's the max.. the 99th is 
< 100MB, and 95th < 30MB..
So I'm not sure that I'm overprovisioned, I'm just not quite yet to the heap 
size based on our partition sizes.
All queries use cluster key, so I'm not accidentally reading a whole partition.
The last place I'm looking - which maybe should be the first - is tombstones.

sorry for the afternoon rant! thanks for your eyes!

On Thu, Jun 28, 2018 at 5:54 PM, Elliott Sims 
<<>> wrote:
It depends a bit on which collector you're using, but fairly normal.  Heap 
grows for a while, then the JVM decides via a variety of metrics that it's time 
to run a collection.  G1GC is usually a bit steadier and less sawtooth than the 
Parallel Mark Sweep , but if your heap's a lot bigger than needed I could see 
it producing that pattern.

On Thu, Jun 28, 2018 at 9:23 AM, Randy Lynn 
<<>> wrote:
I have datadog monitoring JVM heap.

Running 3.11.1.
20GB heap
G1 for GC.. all the G1GC settings are out-of-the-box

Does this look normal?<>

I'm a C# .NET guy, so I have no idea if this is normal Java behavior.

Randy Lynn<>


859.963.1616 <tel:+1-859-963-1616> ext 202

163 East Main Street - Lexington, KY 40507 - 


Randy Lynn<>


859.963.1616 <tel:+1-859-963-1616> ext 202

163 East Main Street - Lexington, KY 40507 - 


Reply via email to