From user-return-64589-archive-asf-public=cust-asf.ponee.io@cassandra.apache.org Mon Oct 21 14:43:17 2019 Return-Path: X-Original-To: archive-asf-public@cust-asf.ponee.io Delivered-To: archive-asf-public@cust-asf.ponee.io Received: from mail.apache.org (hermes.apache.org [207.244.88.153]) by mx-eu-01.ponee.io (Postfix) with SMTP id C517A180626 for ; Mon, 21 Oct 2019 16:43:16 +0200 (CEST) Received: (qmail 37436 invoked by uid 500); 21 Oct 2019 14:43:11 -0000 Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: user@cassandra.apache.org Delivered-To: mailing list user@cassandra.apache.org Received: (qmail 37426 invoked by uid 99); 21 Oct 2019 14:43:11 -0000 Received: from pnap-us-west-generic-nat.apache.org (HELO spamd2-us-west.apache.org) (209.188.14.142) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 21 Oct 2019 14:43:11 +0000 Received: from localhost (localhost [127.0.0.1]) by spamd2-us-west.apache.org (ASF Mail Server at spamd2-us-west.apache.org) with ESMTP id 128351A4334 for ; Mon, 21 Oct 2019 14:43:11 +0000 (UTC) X-Virus-Scanned: Debian amavisd-new at spamd2-us-west.apache.org X-Spam-Flag: NO X-Spam-Score: 0.452 X-Spam-Level: X-Spam-Status: No, score=0.452 tagged_above=-999 required=6.31 tests=[DKIM_SIGNED=0.1, DKIM_VALID=-0.1, HEADER_FROM_DIFFERENT_DOMAINS=0.249, HTML_MESSAGE=0.2, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_NONE=0.001, SPF_NONE=0.001, URIBL_BLOCKED=0.001] autolearn=disabled Authentication-Results: spamd2-us-west.apache.org (amavisd-new); dkim=pass (2048-bit key) header.d=jonhaddad-com.20150623.gappssmtp.com Received: from mx1-he-de.apache.org ([10.40.0.8]) by localhost (spamd2-us-west.apache.org [10.40.0.9]) (amavisd-new, port 10024) with ESMTP id 2ufirDgckoua for ; Mon, 21 Oct 2019 14:43:09 +0000 (UTC) Received-SPF: None (mailfrom) identity=mailfrom; client-ip=2a00:1450:4864:20::443; helo=mail-wr1-x443.google.com; envelope-from=jon@rustyrazorblade.com; receiver= Received: from mail-wr1-x443.google.com (mail-wr1-x443.google.com [IPv6:2a00:1450:4864:20::443]) by mx1-he-de.apache.org (ASF Mail Server at mx1-he-de.apache.org) with ESMTPS id 85C6F7E1B2 for ; Mon, 21 Oct 2019 14:43:08 +0000 (UTC) Received: by mail-wr1-x443.google.com with SMTP id q13so9334230wrs.12 for ; Mon, 21 Oct 2019 07:43:08 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=jonhaddad-com.20150623.gappssmtp.com; s=20150623; h=mime-version:references:in-reply-to:from:date:message-id:subject:to; bh=7odOWErYTUQ3vP+moepPop4HGwpDgrbZgOHVKEWpncI=; b=BYyaP0wRRFV27tQhp+spKp6tMIxgXhu/cXLSS6N743+7TA+b5Nb+NlXICDki/eNkDP mRdfdYgaGEjABDLArcSqj5ivqpoHXxLrhVZws6G3Lb7kKHgYpkVunPa3h+Ok65uYDWHJ FnB9o5NO1k5KG4XuJNv2fpiJMIWiHpY7tvyw+C9khtEK0uEGJo/sjyLlhVrb/aQ+PkDe M0evCSiOFyrbrDMQ4vCNI6hcjosaU6yzK7BKdDY1jorZxKeQL7Zu5XCBkFXQWnQIuVVt ezvuNwWB/tf1evx97n1uXWMvXaYyb8seKuORJjrGWoJ9ZFFX1550txMoZday4PzdHZRE Vhdw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to; bh=7odOWErYTUQ3vP+moepPop4HGwpDgrbZgOHVKEWpncI=; b=pkClL9jq66PJFJeDVl6XrGiNL06smt30ENN6gCT24HaottoloEkS2Mpu8nfjwV00jy dR/GhR0YXYK1zcl0lY60+DUO2RuTGwFC2H2wGIBEjDws1xw3HmZoXsTIXNoDwBUlkM6X lemAt524KPgk0OupPRdcm86zRZFEfzCMnUfJpe58Qmar8acPXOTFwgA2RKprU5Xnxnw9 sqDcyU2rdiTQcf0xtHq8rFGaRku8A0J8hCDb/zLaT6IGWswOx3wRLioo+fYYEW+LPewX 2vchvcL2Wb5As0JI5/L2m+mfTziMgR1ghl7iB9WWWvIGwsJMWzCgTlsDM1hNju2uPXjz RHzA== X-Gm-Message-State: APjAAAWlyf49f0t12e5WvZV0MS7yaIaTe1mUyEVpCqw6koem4Bu0ZMPb cAl1x6YGIZy2yamrv9Rw6CqqJwPKMW1846uFkYaDkzmdJFc= X-Google-Smtp-Source: APXvYqyxGeNPEAPDg4IFK9yv5e9m6yTskawXTw1LxGBcpQQDZNI6POO41pFbfFyjEo4GnYqSv1EgZW72mv07sqNtzRw= X-Received: by 2002:adf:f087:: with SMTP id n7mr6133091wro.1.1571668987658; Mon, 21 Oct 2019 07:43:07 -0700 (PDT) MIME-Version: 1.0 References: <669BB091-E495-41AD-8E7E-6AE317C9269A@tripadvisor.com> In-Reply-To: <669BB091-E495-41AD-8E7E-6AE317C9269A@tripadvisor.com> From: Jon Haddad Date: Mon, 21 Oct 2019 10:42:56 -0400 Message-ID: Subject: Re: GC Tuning https://thelastpickle.com/blog/2018/04/11/gc-tuning.html To: user@cassandra.apache.org Content-Type: multipart/alternative; boundary="000000000000b6d29605956cb30d" --000000000000b6d29605956cb30d Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable I still use ParNew + CMS over G1GC with Java 8. I haven't done a comparison with JDK 11 yet, so I'm not sure if it's any better. I've heard it is, but I like to verify first. The pause times with ParNew + CMS are generally lower than G1 when tuned right, but as Chris said it can be tricky. If you aren't willing to spend the time understanding how it works and why each setting matters, G1 is a better option. I wouldn't run Cassandra in production on less than 8GB of heap - I consider it the absolute minimum. For G1 I'd use 16GB, and never 4GB with Cassandra unless you're rarely querying it. I typically use the following as a starting point now: ParNew + CMS 16GB heap 10GB new gen 2GB memtable cap, otherwise you'll spend a bunch of time copying around memtables (cassandra.yaml) Max tenuring threshold: 2 survivor ratio 6 I've also done some tests with a 30GB heap, 24 GB of which was new gen. This worked surprisingly well in my tests since it essentially keeps everything out of the old gen. New gen allocations are just a pointer bump and are pretty fast, so in my (limited) tests of this I was seeing really good p99 times. I was seeing a 200-400 ms pause roughly once a minute running a workload that deliberately wasn't hitting a resource limit (testing real world looking stress vs overwhelming the cluster). We built tlp-cluster [1] and tlp-stress [2] to help figure these things out. [1] https://thelastpickle.com/tlp-cluster/ [2] http://thelastpickle.com/tlp-stress Jon On Mon, Oct 21, 2019 at 10:24 AM Reid Pinchback wrote: > An i3x large has 30.5 gb of RAM but you=E2=80=99re using less than 4gb fo= r C*. So > minus room for other uses of jvm memory and for kernel activity, that=E2= =80=99s > about 25 gb for file cache. You=E2=80=99ll have to see if you either wan= t a bigger > heap to allow for less frequent gc cycles, or you could save money on the > instance size. C* generates a lot of medium-length lifetime objects whic= h > can easily end up in old gen. A larger heap will reduce the burn of more > old-gen collections. There are no magic numbers to just give because it= =E2=80=99ll > depend on your usage patterns. > > > > *From: *Sergio > *Reply-To: *"user@cassandra.apache.org" > *Date: *Sunday, October 20, 2019 at 2:51 PM > *To: *"user@cassandra.apache.org" > *Subject: *Re: GC Tuning > https://thelastpickle.com/blog/2018/04/11/gc-tuning.html > > > > *Message from External Sender* > > Thanks for the answer. > > This is the JVM version that I have right now. > > openjdk version "1.8.0_161" > OpenJDK Runtime Environment (build 1.8.0_161-b14) > OpenJDK 64-Bit Server VM (build 25.161-b14, mixed mode) > > These are the current flags. Would you change anything in a i3x.large aws > node? > > java -Xloggc:/var/log/cassandra/gc.log > -Dcassandra.max_queued_native_transport_requests=3D4096 -ea > -XX:+UseThreadPriorities -XX:ThreadPriorityPolicy=3D42 > -XX:+HeapDumpOnOutOfMemoryError -Xss256k -XX:StringTableSize=3D1000003 > -XX:+AlwaysPreTouch -XX:-UseBiasedLocking -XX:+UseTLAB -XX:+ResizeTLAB > -XX:+UseNUMA -XX:+PerfDisableSharedMem -Djava.net.preferIPv4Stack=3Dtrue > -XX:SurvivorRatio=3D8 -XX:MaxTenuringThreshold=3D1 -XX:+UseG1GC > -XX:G1RSetUpdatingPauseTimePercent=3D5 -XX:MaxGCPauseMillis=3D200 > -XX:InitiatingHeapOccupancyPercent=3D45 -XX:G1HeapRegionSize=3D0 > -XX:-ParallelRefProcEnabled -Xms3821M -Xmx3821M > -XX:CompileCommandFile=3D/etc/cassandra/conf/hotspot_compiler > -Dcom.sun.management.jmxremote.port=3D7199 > -Dcom.sun.management.jmxremote.rmi.port=3D7199 > -Dcom.sun.management.jmxremote.ssl=3Dfalse > -Dcom.sun.management.jmxremote.authenticate=3Dfalse > -Dcom.sun.management.jmxremote.password.file=3D/etc/cassandra/conf/jmxrem= ote.password > -Dcom.sun.management.jmxremote.access.file=3D/etc/cassandra/conf/jmxremot= e.access > -Djava.library.path=3D/usr/share/cassandra/lib/sigar-bin > -Djava.rmi.server.hostname=3D172.24.150.141 -XX:+CMSClassUnloadingEnabled > -javaagent:/usr/share/cassandra/lib/jmx_prometheus_javaagent-0.3.1.jar=3D= 10100:/etc/cassandra/default.conf/jmx-export.yml > -Dlogback.configurationFile=3Dlogback.xml > -Dcassandra.logdir=3D/var/log/cassandra -Dcassandra.storagedir=3D > -Dcassandra-pidfile=3D/var/run/cassandra/cassandra.pid > -Dcassandra-foreground=3Dyes -cp > /etc/cassandra/conf:/usr/share/cassandra/lib/airline-0.6.jar:/usr/share/c= assandra/lib/antlr-runtime-3.5.2.jar:/usr/share/cassandra/lib/asm-5.0.4.jar= :/usr/share/cassandra/lib/caffeine-2.2.6.jar:/usr/share/cassandra/lib/cassa= ndra-driver-core-3.0.1-shaded.jar:/usr/share/cassandra/lib/commons-cli-1.1.= jar:/usr/share/cassandra/lib/commons-codec-1.9.jar:/usr/share/cassandra/lib= /commons-lang3-3.1.jar:/usr/share/cassandra/lib/commons-math3-3.2.jar:/usr/= share/cassandra/lib/compress-lzf-0.8.4.jar:/usr/share/cassandra/lib/concurr= entlinkedhashmap-lru-1.4.jar:/usr/share/cassandra/lib/concurrent-trees-2.4.= 0.jar:/usr/share/cassandra/lib/disruptor-3.0.1.jar:/usr/share/cassandra/lib= /ecj-4.4.2.jar:/usr/share/cassandra/lib/guava-18.0.jar:/usr/share/cassandra= /lib/HdrHistogram-2.1.9.jar:/usr/share/cassandra/lib/high-scale-lib-1.0.6.j= ar:/usr/share/cassandra/lib/hppc-0.5.4.jar:/usr/share/cassandra/lib/jackson= -core-asl-1.9.13.jar:/usr/share/cassandra/lib/jackson-mapper-asl-1.9.13.jar= :/usr/share/cassandra/lib/jamm-0.3.0.jar:/usr/share/cassandra/lib/javax.inj= ect.jar:/usr/share/cassandra/lib/jbcrypt-0.3m.jar:/usr/share/cassandra/lib/= jcl-over-slf4j-1.7.7.jar:/usr/share/cassandra/lib/jctools-core-1.2.1.jar:/u= sr/share/cassandra/lib/jflex-1.6.0.jar:/usr/share/cassandra/lib/jmx_prometh= eus_javaagent-0.3.1.jar:/usr/share/cassandra/lib/jna-4.2.2.jar:/usr/share/c= assandra/lib/joda-time-2.4.jar:/usr/share/cassandra/lib/json-simple-1.1.jar= :/usr/share/cassandra/lib/jstackjunit-0.0.1.jar:/usr/share/cassandra/lib/li= bthrift-0.9.2.jar:/usr/share/cassandra/lib/log4j-over-slf4j-1.7.7.jar:/usr/= share/cassandra/lib/logback-classic-1.1.3.jar:/usr/share/cassandra/lib/logb= ack-core-1.1.3.jar:/usr/share/cassandra/lib/lz4-1.3.0.jar:/usr/share/cassan= dra/lib/metrics-core-3.1.5.jar:/usr/share/cassandra/lib/metrics-jvm-3.1.5.j= ar:/usr/share/cassandra/lib/metrics-logback-3.1.5.jar:/usr/share/cassandra/= lib/netty-all-4.0.44.Final.jar:/usr/share/cassandra/lib/ohc-core-0.4.4.jar:= /usr/share/cassandra/lib/ohc-core-j8-0.4.4.jar:/usr/share/cassandra/lib/rep= orter-config3-3.0.3.jar:/usr/share/cassandra/lib/reporter-config-base-3.0.3= .jar:/usr/share/cassandra/lib/sigar-1.6.4.jar:/usr/share/cassandra/lib/slf4= j-api-1.7.7.jar:/usr/share/cassandra/lib/snakeyaml-1.11.jar:/usr/share/cass= andra/lib/snappy-java-1.1.1.7.jar:/usr/share/cassandra/lib/snowball-stemmer= -1.3.0.581.1.jar:/usr/share/cassandra/lib/ST4-4.0.8.jar:/usr/share/cassandr= a/lib/stream-2.5.2.jar:/usr/share/cassandra/lib/thrift-server-0.3.7.jar:/us= r/share/cassandra/apache-cassandra-3.11.3.jar:/usr/share/cassandra/apache-c= assandra-thrift-3.11.3.jar:/usr/share/cassandra/stress.jar: > org.apache.cassandra.service.CassandraDaemon > > Best, > > Sergio > > > > Il giorno sab 19 ott 2019 alle ore 14:30 Chris Lohfink < > clohfink85@gmail.com> ha scritto: > > "It depends" on your version and heap size but G1 is easier to get right > so probably wanna stick with that unless you are using small heaps or > really interested in tuning it (likely for massively smaller gains then > tuning your data model). There is no GC algo that is strictly better than > others in all scenarios unfortunately. If your JVM supports it, ZGC or > Shenandoah are likely going to give you the best latencies. > > > > Chris > > > > On Fri, Oct 18, 2019 at 8:41 PM Sergio Bilello > wrote: > > Hello! > > Is it still better to use ParNew + CMS Is it still better than G1GC thes= e > days? > > Any recommendation for i3.xlarge nodes read-heavy workload? > > > Thanks, > > Sergio > > --------------------------------------------------------------------- > To unsubscribe, e-mail: user-unsubscribe@cassandra.apache.org > For additional commands, e-mail: user-help@cassandra.apache.org > > --000000000000b6d29605956cb30d Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
I still use ParNew=C2=A0+ CMS over G1GC with Java 8.=C2=A0= I haven't done a comparison with JDK 11 yet, so I'm not sure if it= 's any better.=C2=A0 I've heard it is, but I like to verify first.= =C2=A0 The pause times with ParNew=C2=A0+ CMS are generally lower than G1 w= hen tuned right, but as Chris said it can be tricky.=C2=A0 If you aren'= t willing to spend the time understanding how it works and why each setting= matters, G1 is a better=C2=A0option.=C2=A0=C2=A0

I woul= dn't run Cassandra in production on less than 8GB of heap - I consider = it the absolute minimum.=C2=A0 For G1 I'd use 16GB, and never 4GB with = Cassandra unless you're rarely querying it.=C2=A0=C2=A0

<= /div>
I typically use the following as a starting point now:
=
ParNew=C2=A0+ CMS
16GB heap
10GB new gen=
2GB memtable cap, otherwise you'll spend a bunch of time cop= ying around memtables (cassandra.yaml)
Max tenuring threshold: 2<= /div>
survivor ratio 6

I've also done some= tests with a 30GB heap, 24 GB of which was new gen.=C2=A0 This worked surp= risingly well in my tests since it essentially keeps everything out of the = old gen.=C2=A0 New gen allocations are just a pointer bump and are pretty f= ast, so in my (limited) tests of this I was seeing really good p99 times.= =C2=A0 I was seeing a 200-400 ms pause roughly once a minute running a work= load that deliberately wasn't hitting a resource limit (testing real wo= rld looking stress vs overwhelming the cluster).

W= e built tlp-cluster [1]=C2=A0and tlp-stress [2] to help figure these things= out.=C2=A0

[= 2] http://thelastpickle.com= /tlp-stress

Jon

<= div>


On Mon, Oct 21, 2019 at 10:24 AM Reid Pinchba= ck <rpinchback@tripadvisor= .com> wrote:

An i3x large has 30.5 gb of RAM but you=E2=80=99re u= sing less than 4gb for C*.=C2=A0 So minus room for other uses of jvm memory= and for kernel activity, that=E2=80=99s about 25 gb for file cache.=C2=A0 = You=E2=80=99ll have to see if you either want a bigger heap to allow for less frequent gc cycles, or you could save money on the instance size.= =C2=A0 C* generates a lot of medium-length lifetime objects which can easil= y end up in old gen.=C2=A0 A larger heap will reduce the burn of more old-g= en collections.=C2=A0 There are no magic numbers to just give because it=E2=80=99ll depend on your usage patterns.

=C2=A0

=C2=A0

Message from External Sender

Thanks for the answer.

This is the JVM version that I have right=C2=A0now.

openjdk version "1.8.0_161"
OpenJDK Runtime Environment (build 1.8.0_161-b14)
OpenJDK 64-Bit Server VM (build 25.161-b14, mixed mode)

These are the current flags. Would you change anything in a i3x.large aws n= ode?=C2=A0

java -Xloggc:/var/log/cassandra/gc.log -Dcassandra.max_queued_native_transp= ort_requests=3D4096 -ea -XX:+UseThreadPriorities -XX:ThreadPriorityPolicy= =3D42 -XX:+HeapDumpOnOutOfMemoryError -Xss256k -XX:StringTableSize=3D100000= 3 -XX:+AlwaysPreTouch -XX:-UseBiasedLocking -XX:+UseTLAB -XX:+ResizeTLAB -XX:+UseNUMA -XX:+PerfDisableSharedMem -Djava= .net.preferIPv4Stack=3Dtrue -XX:SurvivorRatio=3D8 -XX:MaxTenuringThreshold= =3D1 -XX:+UseG1GC -XX:G1RSetUpdatingPauseTimePercent=3D5 -XX:MaxGCPauseMill= is=3D200 -XX:InitiatingHeapOccupancyPercent=3D45 -XX:G1HeapRegionSize=3D0 -XX:-ParallelRefProcEnabled -Xms3821M -Xmx3821M -= XX:CompileCommandFile=3D/etc/cassandra/conf/hotspot_compiler -Dcom.sun.mana= gement.jmxremote.port=3D7199 -Dcom.sun.management.jmxremote.rmi.port=3D7199= -Dcom.sun.management.jmxremote.ssl=3Dfalse -Dcom.sun.management.jmxremote.authenticate=3Dfalse -Dcom.sun.management.j= mxremote.password.file=3D/etc/cassandra/conf/jmxremote.password -Dcom.sun.m= anagement.jmxremote.access.file=3D/etc/cassandra/conf/jmxremote.access -Dja= va.library.path=3D/usr/share/cassandra/lib/sigar-bin -Djava.rmi.server.hostname=3D172.24.150.141 -XX:+CMSClassUnloadingEnabled = -javaagent:/usr/share/cassandra/lib/jmx_prometheus_javaagent-0.3.1.jar=3D10= 100:/etc/cassandra/default.conf/jmx-export.yml -Dlogback.configurationFile= =3Dlogback.xml -Dcassandra.logdir=3D/var/log/cassandra -Dcassandra.storagedir=3D -Dcassandra-pidfile=3D/var/run/cassandra/cassand= ra.pid -Dcassandra-foreground=3Dyes -cp /etc/cassandra/conf:/usr/share/cass= andra/lib/airline-0.6.jar:/usr/share/cassandra/lib/antlr-runtime-3.5.2.jar:= /usr/share/cassandra/lib/asm-5.0.4.jar:/usr/share/cassandra/lib/caffeine-2.= 2.6.jar:/usr/share/cassandra/lib/cassandra-driver-core-3.0.1-shaded.jar:/us= r/share/cassandra/lib/commons-cli-1.1.jar:/usr/share/cassandra/lib/commons-= codec-1.9.jar:/usr/share/cassandra/lib/commons-lang3-3.1.jar:/usr/share/cas= sandra/lib/commons-math3-3.2.jar:/usr/share/cassandra/lib/compress-lzf-0.8.= 4.jar:/usr/share/cassandra/lib/concurrentlinkedhashmap-lru-1.4.jar:/usr/sha= re/cassandra/lib/concurrent-trees-2.4.0.jar:/usr/share/cassandra/lib/disrup= tor-3.0.1.jar:/usr/share/cassandra/lib/ecj-4.4.2.jar:/usr/share/cassandra/l= ib/guava-18.0.jar:/usr/share/cassandra/lib/HdrHistogram-2.1.9.jar:/usr/shar= e/cassandra/lib/high-scale-lib-1.0.6.jar:/usr/share/cassandra/lib/hppc-0.5.= 4.jar:/usr/share/cassandra/lib/jackson-core-asl-1.9.13.jar:/usr/share/cassa= ndra/lib/jackson-mapper-asl-1.9.13.jar:/usr/share/cassandra/lib/jamm-0.3.0.= jar:/usr/share/cassandra/lib/javax.inject.jar:/usr/share/cassandra/lib/jbcr= ypt-0.3m.jar:/usr/share/cassandra/lib/jcl-over-slf4j-1.7.7.jar:/usr/share/c= assandra/lib/jctools-core-1.2.1.jar:/usr/share/cassandra/lib/jflex-1.6.0.ja= r:/usr/share/cassandra/lib/jmx_prometheus_javaagent-0.3.1.jar:/usr/share/ca= ssandra/lib/jna-4.2.2.jar:/usr/share/cassandra/lib/joda-time-2.4.jar:/usr/s= hare/cassandra/lib/json-simple-1.1.jar:/usr/share/cassandra/lib/jstackjunit= -0.0.1.jar:/usr/share/cassandra/lib/libthrift-0.9.2.jar:/usr/share/cassandr= a/lib/log4j-over-slf4j-1.7.7.jar:/usr/share/cassandra/lib/logback-classic-1= .1.3.jar:/usr/share/cassandra/lib/logback-core-1.1.3.jar:/usr/share/cassand= ra/lib/lz4-1.3.0.jar:/usr/share/cassandra/lib/metrics-core-3.1.5.jar:/usr/s= hare/cassandra/lib/metrics-jvm-3.1.5.jar:/usr/share/cassandra/lib/metrics-l= ogback-3.1.5.jar:/usr/share/cassandra/lib/netty-all-4.0.44.Final.jar:/usr/s= hare/cassandra/lib/ohc-core-0.4.4.jar:/usr/share/cassandra/lib/ohc-core-j8-= 0.4.4.jar:/usr/share/cassandra/lib/reporter-config3-3.0.3.jar:/usr/share/ca= ssandra/lib/reporter-config-base-3.0.3.jar:/usr/share/cassandra/lib/sigar-1= .6.4.jar:/usr/share/cassandra/lib/slf4j-api-1.7.7.jar:/usr/share/cassandra/= lib/snakeyaml-1.11.jar:/usr/share/cassandra/lib/snappy-java-1.1.1.7.jar:/us= r/share/cassandra/lib/snowball-stemmer-1.3.0.581.1.jar:/usr/share/cassandra= /lib/ST4-4.0.8.jar:/usr/share/cassandra/lib/stream-2.5.2.jar:/usr/share/cas= sandra/lib/thrift-server-0.3.7.jar:/usr/share/cassandra/apache-cassandra-3.= 11.3.jar:/usr/share/cassandra/apache-cassandra-thrift-3.11.3.jar:/usr/share= /cassandra/stress.jar: org.apache.cassandra.service.CassandraDaemon

Best,

Sergio

=C2=A0

Il giorno sab 19 ott 2019 alle ore 14:30 Chris Lohfi= nk <clohfink85= @gmail.com> ha scritto:

"It depends" on your version and heap size= but G1 is easier to get right so probably wanna stick with that unless you= are using small heaps or really interested in tuning it (likely for massiv= ely smaller gains then tuning your data model). There is no GC algo that is strictly better than others in all scenarios u= nfortunately. If your JVM supports it, ZGC or Shenandoah are likely going t= o give you the best latencies.

=C2=A0

Chris

=C2=A0

On Fri, Oct 18, 2019 at 8:41 PM Sergio Bilello <<= a href=3D"mailto:lapostadisergio@gmail.com" target=3D"_blank">lapostadiserg= io@gmail.com> wrote:

Hello!

Is it still better to use ParNew + CMS Is it still better than G1GC=C2=A0 t= hese days?

Any recommendation for i3.xlarge nodes read-heavy workload?


Thanks,

Sergio

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@cassandra.apache.org
For additional commands, e-mail: user-help@cassandra.apache.org

--000000000000b6d29605956cb30d--