Along with its lower price points, the sleek and centralized management of the new EMC Unity all-flash array may help the product appeal to smaller businesses with less-expansive IT budgets.
During EMC World 2016, Stefan Voss, a member of EMC technical marketing, demonstrated the Unity dashboard so attendees could see how IT staff will actually manage the system. Carol Sliwa, a senior writer for TechTarget's Storage Media Group, filmed the demo.
The EMC Unity UI combines a variety of monitoring and management resources into a central dashboard that runs on HTML5. "[It's] not Java-based," Voss said. "So we can run it on any device." The HTML5 system may allow the Unity dashboard to be accessible on mobile devices, which could be an important feature to help the new array stand out.
The EMC Unity dashboard allows users to customize the exact information they see when examining the array's performance. Administrators will have the ability to look at specific performance metrics in real time when deciding how to best make use of the array.
Some of the administration features available through the dashboard involve provisioning storage to servers and computing systems and managing data protection. Voss noted that the dashboard "is a unified method for snapshotting and replication," and snapshot schedules and data retention periods can be created for different pools of storage on the array. The dashboard also provides remote replication, including asynchronous and synchronous.
Voss stressed ease of use as one of the benefits of the Unity dashboard. "Very easy to use," he said repeatedly as he demonstrated specific features of the array. Creating an easy-to-use UI was a priority for EMC Unity, and this may prove to be an important marketing point, as Unity is targeted largely at small and medium-sized businesses. For smaller organizations that may not have large IT budgets, simplicity can be an important selling point.
Transcript - EMC Unity dashboard demo reveals management features
Hi, my name is Stefan Voss. I'm with EMC technical marketing, and I thought I'd take you through the Unity demo. It's HTML5-based, very scalable and not Java-based, so we can run it on any device, which is very nice. What you typically would see when you go in is the dashboard. It gives you basic statistics on system health. So had we configured it, we would be seeing replication sessions, hardware hosts. If there are any issues, it will color code it.
We would see also the health of the logical elements of the system, LUNs, consistency groups, VMware, file systems. What's also nice is we can see capacity utilization, and we can customize it and get that view by pools, and of course, we can see system performance.
I'm just going to go top to bottom, but normally I would just dial in to a specific task that I want to administer. But when I click on system view, it gives me a nice, friendly view of the actual system I'm using. We have the serial number. We get the view. We get some statistics: model number, software version installed, power consumption and so forth. So the stats that you would like to see.
What's very nice about the display of the system, [is that] it does give me performance views that are very relevant. It is a customizable dashboard. I can filter read, write. I get basic statistics, IOPS, bandwidth. I get the same for file systems, and I can actually optimize this or customize it.
If I wanted to see the storage processor utilization, it's very easy to do. You see there's a lot of counters and statistics available. I would go to system I/O or system resources, and essentially say, "I want to see CPU utilization." It would generate the chart, and voila, it gives it to me right here. I can keep track [of things:] Am I running out of CPU resources? How am I doing for performance? And this is how I can essentially get real-time information around the performance of my system.
I have the ability to then get into management vs. monitoring -- again, very easy to use. If I click the plus button, I can create the name of my pool. I can give it a description. I can walk through the wizard.
I can assign the tier; if it's all-flash, it is obviously going to be an all-flash tier. And then off I go, and I can essentially provision my storage. This is the underlying pool I would leverage to then carve out volumes that I then expose to the host environment. It could be the foundation for file systems [or] VVOLs. And, of course, it could be foundations for block. So I would basically cancel out, because I don't want to lose time going through it.
Then I essentially have the ability to pick what I want to provision to the host. It would be a LUN, for example, in this case. I can give it a name, very similar. I can map it to the storage pool. I can map it to the host that I basically have configured into the systems.
I can set up a snapshot, a snapshot schedule and essentially manage my replication sessions specific to that host.
Same for file systems -- there are some that already exist. It would be a similar wizard. Of course, I have to specify if it is a Windows file system based on SMB 3.0 or NFS. Similar wizard, I give it a name.
I select the storage pool I want to use. I can mount the shares, create snapshots. It's a unified method for snapshotting and replication, which is nice. And then it gives me a summary and off I go.
We have VVOL support. I can specify my endpoints, create them. I can create virtual volumes. All these activities are wizard-driven. I can create data stores. So this gives me a nice full end-to-end integration with VMware as the orchestration layer.
I can specify host access. I can discover my initiators -- all these things are very common [and] haven't really changed in terms of the logical elements. And then I can manage my data protection from here as well: so snapshot schedule, [and] we have a built -n scheduler. We have a policy engine available. I can set the retention period.
There is additional orchestration that I can make available with AppSync. So this is just the native scheduler that I have built-in. I can create a schedule. I can specify my times, set the retention period and off I go. So that would then apply to the snapshot.
Same for replication, this is remote replication. Very similar, and I can create my sessions. I can manage my connections, specify my interfaces, and then specify if it's asynchronous or synchronous replication that I want to do. That basically is the data protection bucket. And then I can see, "Are there any systems alerts?" This gets more into event management.
Everything seems to be running fine; otherwise, it would give me some flags. Are there any jobs scheduled? Had I pulled the trigger on any of these management activities, it would basically give me an updated view in terms of the progress of creating the LUN, the replication session and so forth, everything in real time, and it keeps a log.
And then last but not least, which is very nice, we have this notion of proactive assist, proactive support. We want to make sure the customer has the ability to self-manage, if you will, self-support. We have access to support forums, a product portal; we can download software, as well as documentation and release notes. We have a lot of online training available, and we could take you straight to the EMC Store so you can shop for products you would need as you build-out your system.