I have the pleasure to participate in Dell EMC World 2017 this week in Las Vegas and want to give you some of my impressions regarding Dell EMC strategy, product announcements and various other things.
The key word for this years Dell EMC World is REALIZE.
Realize what is going on in the industry and that there is transformation ahead:
For all the challenges that are coming up with these transformations, Dell Technologies has the right solutions and can support you.
In the general sessions there were some examples like the Dell Canvas and how it will transform the creative process of creating new things and Ken Black (VP of Digital Design Transformation @Nike) explained how they use technology like Dell Canvas to empower their creative design. In addition as a hobby photographer I´m also personally very excited about this and can´t wait to try it in the near future for image editing. Have a look at the video to get to know Dell Canvas.
There were several guru sessions for different topics like AI (Sir Tim Berners-Lee) and Virtual Reality / Augmented Reality (Tim Sweeney, founder of EPIC Games). It was very interesting to listen to those guys and get their point of view to those emerging topics. Furthermore the topic of VR/AR was often found in the solution expo in different forms like VR gaming or the possibility to fly a drone with an integrated camera and you get the direct video feed to your VR headset. Amazing stuff, except the motion sickness 😉
Major product announcements:
VMAX 950F is a new upcoming VMAX All-Flash array which will be the successor of VMAX 450F/850F. In a performance point of view the 950F will exceed the 850F and regarding scale-out it will match the 850F.
Unity 350F/450F/550F/650F: More Performance, higher density (80 drives in 3U)
Isilon Infinity: New generation (6) of Isilon, higher density of nodes, modular design, option of All-Flash nodes
XtremIO X2: Scale-Up and Scale-Out, native and efficient replication, improved data efficiency
Dell PowerEdge Server 14G
VXRail 4.5: based on PowerEdge 14G, support for vSphere 6.5 and VSAN 6.6
Integrated Data Protection Appliance: turnkey data protection appliance converges storage, protection software, search and analytics in a single, easy-to-deploy solution
Despite it isn´t really a product announcement, but in my opinion it is an important topic. There was the introduction of a new flexible consumption model (for HCI) which allows you to consume compute power & storage like you would do in the cloud:
no upfront payment
the option to return all or some of the equipment any time after the first year
establish a simple payment schedule for the entire hardware
Gwen Stefani was rocking the house at the Wednesday evening event – the Customer Appreciation Event.
Did you know Gwen Stefani is 47 years old right now? Hard to believe.
That´s it so far. I hope I could give you some useful information about this week at Dell EMC World. You didn´t attend this year but you want to do next year? Save the date: 28.04.2018 – 03.05.2018 again in fabulous Las Vegas.
Just to recall what are Cloud Pools I have to explain Smart Pools first.
Smart Pools are a policy-based fully automated tiering mechanism of the Isilon system to tier files between different node types. You can use it to move cold data with a file policy like “last access >6 month ago” from fast flash-based nodes (S-Nodes) to NL-SAS based nodes (NL-Nodes).
Now Cloud Pools extend this mechanism and leverage the Elastic Cloud Storage system like an additional tier. As you surely remember one of its characteristic features is the low price per GB and as such the ECS can now be considered the lowest tier for the Isilon system.
Well first of all I tell you what it isn´t – it is not the type of cloud storage where you put your data to a place you don´t know. That´s not why it is called Cloud Storage. It is called Cloud Storage because you can consume storage like you would out of the cloud:
inexpensive – without throwing around with the numbers, but comparable or below the prices of all those cloud storage providers out there
scalability – you will never run out of storage in the cloud, neither you will with a ECS. Scale to Peta- and Exabytes.
agile – if you need more storage, get more and consume more
And why do you know where your data is? Because it is a physical storage array (ok there is also a software only version, the DIY option…) and you can choose where to put it.
First of all we have to login to the VDP Configuration Web Interface (not the one integrated in the VMWare Web Client).
On the tab Storage we have to click “Add Data Domain”
Enter the FQDN of the data domain system and enter the credentials of the previous created DD Boost User
Enter the SNMP Community
Finish – that´s it!
Create a backup job:
Connect to the vSphere Web Client
Enter the VMWare Data Protection section and connect to the deployed VDP appliance
Create a backup job and follow the wizard
Gallery: How to create a backup job in the vSphere Web Client
Testing & Results:
So after creating the first backup job we have to trigger a manual backup of one of the protected virtual machines. Simply done through right-clicking the virtual machine in the vSphere Web Client and click “Backup now”.
Let´s have a look in the CLI of the data domain system. You can display the dd boosts stats with the following command: ddboost show stats int
You can see a few columns with different informations, but the interesting ones are the first three.
Backup: The amount (speed) of data that is backed up. When some segments of the data are saved because of the deduplication or compression it will still be shown as written due to the Data Domain design with deduplication. Every backup is handled like a full backup but due to dedup/compression you only physically write/transfer data like it would be a incremental backup. For example in this test I was backing up a virtual machine with ~120 GB in size and if it is backed up in lets say 1 second due to dedup/compress savings we would see a value of 120 GB / s as Backup speed.
Post-comp Written: The amount that is left after the client-side deduplication & compression and really gets written
Network In: And finally what was transfered over the network
For the first full backup you can see that there is a moderate backup speed (dont forget it´s a homelab NAS with 2 disks and data is read and written to it) and already a fair amount of savings due to DD Boost mechanisms.
Now let´s start a full backup of the same machine again and let´s see what is changing.
Now you can see the amount post-comp written and transfered over the network is reduced. And when we have a look at the backup numbers we see a huge increase in the “backup (transfer) speed”. Here you can see the power of Data Domain and DD Boost when the deduplication and compression kicks in – the more data you save the more efficient the system works and you can enhance your retention time.
That´s it – in three easy steps we have deployed VDP + virtual EMC Data Domain and integrated them. What´s the benefit?
VDP(A) is a part of vSphere 6 and now at zero cost
Data Domain: Deduplication and compression saves a lot of space and makes it very efficient
Data Domain: Self healing filesystem and regular healthchecks make sure your backup WILL work when you need them
Upcoming virtual deployment option, the perfect fit for remote offices! And on top you simply replicate your backup to your data domain in your central data center.
Fully integrated in vSphere! Manage your backups in your vSphere Web Client.
Side Note #1: Remove a Data Domain from VDP:
When you want to remove the data domain from the VDP you have to use a CLI command right now. Delete all backup jobs in the VDP related to the Data Domain as target and then execute the following vdp cli command: