Snapshot performance comparison EMC Celerra vs NetApp FAS

Posted on 2010-09-16
Last Modified: 2013-11-14
I am comparing NetApp and EMC solutions for a VMware solution. The VMs will be on NFS. I know that EMC and NetApp use different methods of snapshots. Correct me if I am wrong on this, with EMC when a snap is made, all old data that is modified is written to hidden snap volume and the new data is written to the original location. With NetApp a pointer is put in the original data and the new data is written elsewhere. So with EMC there is a read operation and 2 writes, and with NetApp there are 2 writes although one of them will be small.

However though when it comes to reading that data, doesn't this mean that NetApp will have to make 2 read IOs to access the data, one for the pointer and one for the actual data? EMC will only have one read since the new data already replaced the old data. Does this sound correct?

Also from what my understanding is, with an NFS share EMC and NetApp will have roughly the same size usage by snaps?

Any real word experience with any type of performance impact having snaps of a file share running on an hourly basis for 2 days time would have with either system?

Question by:ryan80
Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

  • Help others & share knowledge
  • Earn cash & points
  • Learn & ask questions
  • 2
  • 2
LVL 42

Expert Comment

ID: 33698650
I don't believe that Netapp has a penalty since the snapshot data is stored in the same volume (since only keeps pointers).  The snapshots for Netapp only take up room when there is a deletion of original data, there is no penalty for new data.    EMC and Netapp snapshtos work differently, from what I have been able to find and I may be in incorrect but Celerra is still copy on write which means that any changes are being written to a snapshot location.

On the Netapp CIFS shares I normally perform a snapshot every 1-2 hours with no percievable performance issues at all (setup for 36 hourly, 7 daily, and 4 weekly snapschedule).  Users can restore items on their own and see "previous versions" when right clicking on a file.  With deduplication the storage is smaller in size then the NTFS share that it originated from and snapshot usage is very limited in size even when dozens of snapshots are availalbe (up to 255)

LVL 12

Author Comment

ID: 33703125
I would think though that there would be an IO hit on reading current data since for every read that has to take place of data that has been snapped, it will have to make 1 IO to read the pointer (not a large amount of data, but still an IO) and then read the current data from the other location.

From what I see EMC will not have the read performance hit, but will have a bigger performance hit when taking the snap. It will have to read the data being changed, write the old data to the snapshot hidden location, and then write the new data. Now I don't know much but this sounds like it could take some time depending on the amount of changed data.
LVL 42

Accepted Solution

paulsolov earned 500 total points
ID: 33703225
On the Netapp it's still at the same location just that new data is also in the same volume as well.  In your environment (I have configured much larger environments) I have not see any performance issues.  If you are going to have SQL databases with high I/O on the VMs it is better to have them outside of the NFS volume by either using MS iSCSI, NPIV directly to LUN, or RDM.  Otherwise you should be ok.
LVL 12

Author Comment

ID: 33703296
No I am not concerned about the impact in my environment at all, just more curious about the differences between the two vendor types of snapshots.

Featured Post

Why Off-Site Backups Are The Only Way To Go

You are probably backing up your data—but how and where? Ransomware is on the rise and there are variants that specifically target backups. Read on to discover why off-site is the way to go.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

This article aims to explain the working of CircularLogArchiver. This tool was designed to solve the buildup of log file in cases where systems do not support circular logging or where circular logging is not enabled
Concerto Cloud Services, a provider of fully managed private, public and hybrid cloud solutions, announced today it was named to the 20 Coolest Cloud Infrastructure Vendors Of The 2017 Cloud  (…
This tutorial will walk an individual through the process of installing the necessary services and then configuring a Windows Server 2012 system as an iSCSI target. To install the necessary roles, go to Server Manager, and select Add Roles and Featu…
How to install and configure Citrix XenApp 6.5 - Part 1. In this video tutorial we have explained step by step installation of Citrix XenApp 6.5 Server on Windows Server 2008 R2 is explained in this video. We have explained the difference between…

717 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question