YCSB Benchmark

classic Classic list List threaded Threaded
7 messages Options
Shay Alon Shay Alon
Reply | Threaded
Open this post in threaded view
|

YCSB Benchmark

Hi,

I have followed the instructions written in the README page here:

https://github.com/brianfra... 
I used the following configuration:
IgniteClient (not IgniteSqlClient).
JAVA 8
Linux
RAM 500G
72 CPU
threads 2
3 backups
1M entries
And run workloads a,b and c and got the following results:
workloada: [OVERALL], Throughput(ops/sec), 323.51989647363314
workloadb: OVERALL], Throughput(ops/sec), 385.2080123266564
workloadc: [OVERALL], Throughput(ops/sec), 375.7985719654265

This results are very low compare to the results that are published in this page.
Can you please direct me? What am I doing wrong?

Thanks


Ivan Artiukhov Ivan Artiukhov
Reply | Threaded
Open this post in threaded view
|

Re: YCSB Benchmark

Hi!

What results are you using for comparison? Please provide URL. There are
plenty of factors which affects performance. Hardware, number of server and
client nodes, Ignite memory mode, replication factor (number of backups),
number of benchmarking threads, data amount (record count) -- just to
mention a few.



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Shay Alon Shay Alon
Reply | Threaded
Open this post in threaded view
|

Re: YCSB Benchmark

Thanks for your reply.
I was comparing to the results published here:
I run the benchmark on very powered machine with 72 cpu and 500G RAM. I used 32 threads.





On Tue, Sep 18, 2018 at 8:38 AM Ivan Artiukhov <[hidden email]> wrote:
Hi!

What results are you using for comparison? Please provide URL. There are
plenty of factors which affects performance. Hardware, number of server and
client nodes, Ignite memory mode, replication factor (number of backups),
number of benchmarking threads, data amount (record count) -- just to
mention a few.



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Ivan Artiukhov Ivan Artiukhov
Reply | Threaded
Open this post in threaded view
|

Re: YCSB Benchmark

Hi!

1) Are you running Ignite and YCSB nodes on the same machine? How many YCSB nodes do you start? 
2) What DataRegionConfiguration do you set in ignite.xml config file? Does the whole dataset of 1M entries fit in RAM ()?
3) Do other Ignite parameters have the same value as in the blog post you are referencing to?

вт, 18 сент. 2018 г. в 9:18, Shay Alon <[hidden email]>:
Thanks for your reply.
I was comparing to the results published here:
I run the benchmark on very powered machine with 72 cpu and 500G RAM. I used 32 threads.





On Tue, Sep 18, 2018 at 8:38 AM Ivan Artiukhov <[hidden email]> wrote:
Hi!

What results are you using for comparison? Please provide URL. There are
plenty of factors which affects performance. Hardware, number of server and
client nodes, Ignite memory mode, replication factor (number of backups),
number of benchmarking threads, data amount (record count) -- just to
mention a few.



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Shay Alon Shay Alon
Reply | Threaded
Open this post in threaded view
|

Re: YCSB Benchmark

Hi,

1. Yes I'm running Ignite and YCSB nodes on the same machine and starting one YCSB node.
2. This is the DataRegionConfiguration:
<property name="defaultDataRegionConfiguration">

                    <bean class="org.apache.ignite.configuration.DataRegionConfiguration">

                        <property name="name" value="default_data_region"/>

                        <property name="persistenceEnabled" value="true"/>

                        <!-- Setting the max size of the default region to 10GB. -->

                        <property name="maxSize" value="#{10L * 1024 * 1024 * 1024}"/>

                        <!-- Setting the initial size of the default region to 10GB. -->

                        <property name="initialSize" value="#{10L * 1024 * 1024 * 1024}"/>

                        <property name="checkpointPageBufferSize" value="#{1L * 1024 * 1024 * 1024}"/>

                        <property name="metricsEnabled" value="true"/>

                    </bean>

                </property>


This are the console printings:


[17:23:45] Ignite node started OK (id=9115dfdb)

[17:23:45] Topology snapshot [ver=1, servers=1, clients=0, CPUs=72, offheap=10.0GB, heap=1.0GB]

[17:23:45]   ^-- Node [id=9115DFDB-8AC2-42C7-BCB5-7870BC1050AE, clusterState=INACTIVE]

[17:23:45]   ^-- Baseline [id=0, size=1, online=1, offline=0]

[17:23:45]   ^-- All baseline nodes are online, will start auto-activation

[17:23:45] Data Regions Configured:

[17:23:45]   ^-- default_data_region [initSize=10.0 GiB, maxSize=10.0 GiB, persistenceEnabled=true]


I guess not all 1M dataset is on heap but almost (each entry is 1K, I'm writing 1M and the heap is 1G).


3. Regarding the Generic Parameters, it not using the same hardware and network, but a very strong machine as you can see in my first mail.

   Regarding the "Apache Ignite Parameters", I'm using the same. As you can see here:

<property name="cacheConfiguration">
<list>
<bean class="org.apache.ignite.configuration.CacheConfiguration">
<property name="name" value="usertable"/>
<property name="atomicityMode" value="ATOMIC"/>
<property name="cacheMode" value="PARTITIONED"/>
<property name="backups" value="1"/>
<property name="writeSynchronizationMode" value="FULL_SYNC"/>
</bean>
</list>
</property>



On Fri, Sep 21, 2018 at 8:43 AM Ivan Artiukhov <[hidden email]> wrote:
Hi!

1) Are you running Ignite and YCSB nodes on the same machine? How many YCSB nodes do you start? 
2) What DataRegionConfiguration do you set in ignite.xml config file? Does the whole dataset of 1M entries fit in RAM ()?
3) Do other Ignite parameters have the same value as in the blog post you are referencing to?

вт, 18 сент. 2018 г. в 9:18, Shay Alon <[hidden email]>:
Thanks for your reply.
I was comparing to the results published here:
I run the benchmark on very powered machine with 72 cpu and 500G RAM. I used 32 threads.





On Tue, Sep 18, 2018 at 8:38 AM Ivan Artiukhov <[hidden email]> wrote:
Hi!

What results are you using for comparison? Please provide URL. There are
plenty of factors which affects performance. Hardware, number of server and
client nodes, Ignite memory mode, replication factor (number of backups),
number of benchmarking threads, data amount (record count) -- just to
mention a few.



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Shay Alon Shay Alon
Reply | Threaded
Open this post in threaded view
|

Re: YCSB Benchmark

Hi,

Can someone can help here?



On Sat, Sep 22, 2018 at 9:39 AM Shay Alon <[hidden email]> wrote:
Hi,

1. Yes I'm running Ignite and YCSB nodes on the same machine and starting one YCSB node.
2. This is the DataRegionConfiguration:
<property name="defaultDataRegionConfiguration">

                    <bean class="org.apache.ignite.configuration.DataRegionConfiguration">

                        <property name="name" value="default_data_region"/>

                        <property name="persistenceEnabled" value="true"/>

                        <!-- Setting the max size of the default region to 10GB. -->

                        <property name="maxSize" value="#{10L * 1024 * 1024 * 1024}"/>

                        <!-- Setting the initial size of the default region to 10GB. -->

                        <property name="initialSize" value="#{10L * 1024 * 1024 * 1024}"/>

                        <property name="checkpointPageBufferSize" value="#{1L * 1024 * 1024 * 1024}"/>

                        <property name="metricsEnabled" value="true"/>

                    </bean>

                </property>


This are the console printings:


[17:23:45] Ignite node started OK (id=9115dfdb)

[17:23:45] Topology snapshot [ver=1, servers=1, clients=0, CPUs=72, offheap=10.0GB, heap=1.0GB]

[17:23:45]   ^-- Node [id=9115DFDB-8AC2-42C7-BCB5-7870BC1050AE, clusterState=INACTIVE]

[17:23:45]   ^-- Baseline [id=0, size=1, online=1, offline=0]

[17:23:45]   ^-- All baseline nodes are online, will start auto-activation

[17:23:45] Data Regions Configured:

[17:23:45]   ^-- default_data_region [initSize=10.0 GiB, maxSize=10.0 GiB, persistenceEnabled=true]


I guess not all 1M dataset is on heap but almost (each entry is 1K, I'm writing 1M and the heap is 1G).


3. Regarding the Generic Parameters, it not using the same hardware and network, but a very strong machine as you can see in my first mail.

   Regarding the "Apache Ignite Parameters", I'm using the same. As you can see here:

<property name="cacheConfiguration">
<list>
<bean class="org.apache.ignite.configuration.CacheConfiguration">
<property name="name" value="usertable"/>
<property name="atomicityMode" value="ATOMIC"/>
<property name="cacheMode" value="PARTITIONED"/>
<property name="backups" value="1"/>
<property name="writeSynchronizationMode" value="FULL_SYNC"/>
</bean>
</list>
</property>



On Fri, Sep 21, 2018 at 8:43 AM Ivan Artiukhov <[hidden email]> wrote:
Hi!

1) Are you running Ignite and YCSB nodes on the same machine? How many YCSB nodes do you start? 
2) What DataRegionConfiguration do you set in ignite.xml config file? Does the whole dataset of 1M entries fit in RAM ()?
3) Do other Ignite parameters have the same value as in the blog post you are referencing to?

вт, 18 сент. 2018 г. в 9:18, Shay Alon <[hidden email]>:
Thanks for your reply.
I was comparing to the results published here:
I run the benchmark on very powered machine with 72 cpu and 500G RAM. I used 32 threads.





On Tue, Sep 18, 2018 at 8:38 AM Ivan Artiukhov <[hidden email]> wrote:
Hi!

What results are you using for comparison? Please provide URL. There are
plenty of factors which affects performance. Hardware, number of server and
client nodes, Ignite memory mode, replication factor (number of backups),
number of benchmarking threads, data amount (record count) -- just to
mention a few.



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Ilya Suntsov Ilya Suntsov
Reply | Threaded
Open this post in threaded view
|

Re: YCSB Benchmark

Hi,

Did you try to run Cassandra on the same YCSB/Cassandra configuration? 

вс, 30 сент. 2018 г. в 16:23, Shay Alon <[hidden email]>:
Hi,

Can someone can help here?



On Sat, Sep 22, 2018 at 9:39 AM Shay Alon <[hidden email]> wrote:
Hi,

1. Yes I'm running Ignite and YCSB nodes on the same machine and starting one YCSB node.
2. This is the DataRegionConfiguration:
<property name="defaultDataRegionConfiguration">

                    <bean class="org.apache.ignite.configuration.DataRegionConfiguration">

                        <property name="name" value="default_data_region"/>

                        <property name="persistenceEnabled" value="true"/>

                        <!-- Setting the max size of the default region to 10GB. -->

                        <property name="maxSize" value="#{10L * 1024 * 1024 * 1024}"/>

                        <!-- Setting the initial size of the default region to 10GB. -->

                        <property name="initialSize" value="#{10L * 1024 * 1024 * 1024}"/>

                        <property name="checkpointPageBufferSize" value="#{1L * 1024 * 1024 * 1024}"/>

                        <property name="metricsEnabled" value="true"/>

                    </bean>

                </property>


This are the console printings:


[17:23:45] Ignite node started OK (id=9115dfdb)

[17:23:45] Topology snapshot [ver=1, servers=1, clients=0, CPUs=72, offheap=10.0GB, heap=1.0GB]

[17:23:45]   ^-- Node [id=9115DFDB-8AC2-42C7-BCB5-7870BC1050AE, clusterState=INACTIVE]

[17:23:45]   ^-- Baseline [id=0, size=1, online=1, offline=0]

[17:23:45]   ^-- All baseline nodes are online, will start auto-activation

[17:23:45] Data Regions Configured:

[17:23:45]   ^-- default_data_region [initSize=10.0 GiB, maxSize=10.0 GiB, persistenceEnabled=true]


I guess not all 1M dataset is on heap but almost (each entry is 1K, I'm writing 1M and the heap is 1G).


3. Regarding the Generic Parameters, it not using the same hardware and network, but a very strong machine as you can see in my first mail.

   Regarding the "Apache Ignite Parameters", I'm using the same. As you can see here:

<property name="cacheConfiguration">
<list>
<bean class="org.apache.ignite.configuration.CacheConfiguration">
<property name="name" value="usertable"/>
<property name="atomicityMode" value="ATOMIC"/>
<property name="cacheMode" value="PARTITIONED"/>
<property name="backups" value="1"/>
<property name="writeSynchronizationMode" value="FULL_SYNC"/>
</bean>
</list>
</property>



On Fri, Sep 21, 2018 at 8:43 AM Ivan Artiukhov <[hidden email]> wrote:
Hi!

1) Are you running Ignite and YCSB nodes on the same machine? How many YCSB nodes do you start? 
2) What DataRegionConfiguration do you set in ignite.xml config file? Does the whole dataset of 1M entries fit in RAM ()?
3) Do other Ignite parameters have the same value as in the blog post you are referencing to?

вт, 18 сент. 2018 г. в 9:18, Shay Alon <[hidden email]>:
Thanks for your reply.
I was comparing to the results published here:
I run the benchmark on very powered machine with 72 cpu and 500G RAM. I used 32 threads.





On Tue, Sep 18, 2018 at 8:38 AM Ivan Artiukhov <[hidden email]> wrote:
Hi!

What results are you using for comparison? Please provide URL. There are
plenty of factors which affects performance. Hardware, number of server and
client nodes, Ignite memory mode, replication factor (number of backups),
number of benchmarking threads, data amount (record count) -- just to
mention a few.



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/


--
Best Regards,
Ilya Suntsov
GridGain Systems