Hi, I am trying to make simple “write speed” tests on my local pc(sata ssd).
I am using c#.
Here is how i am writing 100 000 records very simply;
And this write operation takes 56 second. Am i missing something or this is best performance for my local pc ?
Maybe bulk insert?
You are only using a single thread to perform these ‘puts’ and the ‘puts’ are all synchronous. You could use more than one thread or use the async APIs.
If you are using .NET Framework, there is an AerospikeDemo application that can run multi-threaded benchmarks for you.
Thanks i will check that application.
I couln’t prove my self Aerospike is faster then Postgres(Bulk) on Insert operations. I have tried 100 000 records both(Aero and Postgres) takes around 10 second
Then I have tried 1 000 000 record Aerospike(Batch) / Postgres(Bulk) insert From C# app
Postgres was faster then Aerospike.
Is it normal ?
Could you forward the spec of the machine you are running on and the configuration of Aerospike?
Thanks for helping, I really want to find why
I am using; Processor: Intel(R) Core™ i7-4700MQ CPU @ 2.40GHz (8 CPUs), ~2.4GHz
Memory: 16384MB RAM
SSD Sata
By the way i am using vagrant VM in windows 10 machine here is some stats from aerospike dashboard
I am using IP adress of my vagrant VM (in my pc) Maybe its slow beacuse of this reason. “localhost:3000” is not working.
I can share my test code parts also;
Keys and bins are simple strings.
For postgres iam also generating my entity and bulk insert.
Ahmet
Can you also share your aerospace.conf file with us?
Thanks Tibor
I didn’t found “aerospike.conf” file They say its in etc/aerospike/aerospike.conf path but i haven’t find such path.
I have follwed these steps to install; https://www.aerospike.com/docs/operations/install/vagrant/win/using-vagrant.html
can you try:
vagrant ssh
and from within the prompt type
cd /etc/aerospike/
and then
cat aerospike.conf
and post the output of the cat command to this thread.
# Aerospike database configuration file.
# This stanza must come first.
service {
user root
group root
paxos-single-replica-limit 1 # Number of nodes where the replica count is automatically reduced to 1.
pidfile /var/run/aerospike/asd.pid
# service-threads 4
# transaction-queues 4
# transaction-threads-per-queue 4
proto-fd-max 15000
node-id-interface eth1
}
logging {
# Log file must be an absolute path.
file /var/log/aerospike/aerospike.log {
context any info
}
file /var/log/aerospike/udf.log {
context udf info
context aggr info
}
}
network {
service {
address eth1
port 3000
# access-address <Published IP>
# access-address <NAT IP>
}
heartbeat {
mode multicast
multicast-group 239.1.99.222
address eth1
port 9918
protocol v3
# To use unicast-mesh heartbeats, comment out the 3 lines above and
# use the following 4 lines instead.
# mode mesh
# port 3002
# mesh-address 10.1.1.1
# mesh-port 3002
interval 150
timeout 10
}
fabric {
port 3001
address eth1
}
info {
port 3003
}
}
#namespace test {
# replication-factor 2
# memory-size 4G
# default-ttl 30d # 30 days, use 0 to never expire/evict.
#
# storage-engine memory
#}
namespace test {
replication-factor 2
memory-size 2G
default-ttl 5d # 5 days, use 0 to never expire/evict.
# To use file storage backing, comment out the line above and use the
# following lines instead.
storage-engine device {
file /opt/aerospike/data/test.dat
filesize 5G
data-in-memory true # Store data in memory in addition to file.
}
}
Here is my vagrant file;
# The contents below were provided by the Packer Vagrant post-processor
class VagrantPlugins::ProviderVirtualBox::Action::Network
def dhcp_server_matches_config?(dhcp_server, config)
true
end
end
Vagrant.configure("2") do |config|
config.vm.base_mac = "0800278AA558"
end
# The contents below (if any) are custom contents provided by the
# Packer template during image build.
# -*- mode: ruby -*-
# vi: set ft=ruby :
Vagrant.configure("2") do |config|
config.vm.define "vagrant-as"
config.vm.box = "aerospike-vm"
config.vm.network "forwarded_port", guest: 3000, host: 3000 , auto_correct: true
config.vm.network "forwarded_port", guest: 8081, host: 8081 , auto_correct: true
config.vm.network "private_network", type: :dhcp
config.ssh.insert_key = false
config.vm.box_download_insecure = true
# config.vm.network "private_network", type: "dhcp", virtualbox__intnet: "name"
# virtualbox__intnet: true
# Berkshelf
# config.berkshelf.enabled = true
# Shell - Hello World
# config.vm.provision :shell, :inline => "C:\\vagrant\\scripts\\HelloWorld.bat"
config.vm.provider :virtualbox do |v, override|
v.gui = false
v.customize ["modifyvm", :id, "--memory", 2048]
v.customize ["modifyvm", :id, "--cpus", 2]
# config.vm.network "private_network", :type => 'dhcp', :name => 'vboxnet0', :adapter => 2
# v.customize ["modifyvm", :id, "--vram", "256"]
# v.customize ["setextradata", "global", "GUI/MaxGuestResolution", "any"]
# v.customize ["setextradata", :id, "CustomVideoMode1", "1024x768x32"]
# v.customize ["modifyvm", :id, "--ioapic", "on"]
# v.customize ["modifyvm", :id, "--rtcuseutc", "on"]
# v.customize ["modifyvm", :id, "--accelerate3d", "on"]
# v.customize ["modifyvm", :id, "--clipboard", "bidirectional"]
end
config.vm.provider :vmware_fusion do |v, override|
v.gui = false
v.vmx["memsize"] = "2048"
v.vmx["numvcpus"] = "2"
v.vmx["cpuid.coresPerSocket"] = "1"
# v.vmx["ethernet0.virtualDev"] = "vmxnet3"
# v.vmx["scsi0.virtualDev"] = "lsilogic"
end
config.vm.provider :vmware_workstation do |v, override|
v.gui = false
v.vmx["memsize"] = "2048"
v.vmx["numvcpus"] = "2"
v.vmx["cpuid.coresPerSocket"] = "1"
# v.vmx["ethernet0.virtualDev"] = "vmxnet3"
# v.vmx["scsi0.virtualDev"] = "lsilogic"
end
end
Can you also run the following command within your vagrant vm
asinfo -v service
This should give you the ip(s) address being sent back to your client. I want to confirm its one ip address and its the one you’ve been using for your client to connect. (and its not the NAT ip 10.0.2.15)
This ip adress i am using in c# app.
that is weird. Try monitoring latency while you run your test.
from within the vm run asadm by typing:
asadm
then type
show latency
or
watch show latency
Is this a one node cluster? to rule out an issue with the dhcp assigned IP address you could create a second interface in your vagrantfile
config.vm.network "private_network", ip: "1.1.1.200"
modify you address binding in aerospike.conf to use eth2 instead of eth1. and use that ip address in your application.
