Slower Insert Performance in Couchbase Cluster

Hi,

I am new to Couchbase. I was in the process of checking raw performance of INSERTs with various Couchbase setups (single node, 2-node, 5-node etc.) I am using VMs on Windows Azure to do my testing since I do not have enough hardware to test it on my servers. I am using the .NET client library provided by Couchbase to write my C# based test application that does record insertion in a loop.

Here are the results of inserts thus far:

Time taken for 10 million records on a hardware server - roughly 10 minutes
Time taken for 10 million records on a VM server on Azuere - roughly 21 minutes
Time taken for 10 million records on a 2-node VM cluster on Azure - 63 minutes

So here are my questions:

Is visualization causing the insertion to come down to half of a hardware server?
Is it normal for a Couchbase cluster to take 3 times as much compared to a single node Couchbase?

I have followed the following article to set up Couchbase cluster in Windows Azure:
Couchbase on Azure

Hello,

So when you create a cluster be sure you put at least 3 nodes. (to distribute properly the data and replicas), and use 1 replica.

I do not see why it it that slow, could you give us more information about:

  • the version of Couchbase Server and SDK
  • Size of your bucket (RAM Quota)
  • number of writers
  • Code that inject

Any information that could be helpful to find the source of the problem.

Tug

Well, I tried it with 3 nodes as you suggested with 1 replica. Performance got worse. It’s now taking roughly 75 minutes for 10 million records.

VM Size is: 2 Cores, 14GB RAM
RAM allocated for bucket is 10GB

Any thoughts?

Could you take a screen shoot of the disk write que during a data load.
http://www.couchbase.com/docs/couchbase-manual-2.0/images/web-console-server-stats-diskqueues.png
During data load how many SET()/sec are you doing?