Storage Designed for the Microsoft Administrator


Last week EMC announced the release of 41 new products. Yes you read that right….FORTY-ONE.

Among the products announced is the VNX & VNXe – our new unified family of products. The VNXe (the e is for entry) is a new low end entry model designed for the small & medium businesses with a price tag under $10k.

 Among some of the many features that make this product so attractive is:

  • The VNXe3100 is only 2U of rack space while the VNXe3300 is 3U.
  • The VNXe3100 scales up to 96 drives and the VNXe3300 scales up to 120 drives!
  • Both units support Ethernet based protocols NFS, CIFS & iSCSI.
  • The VNXe3100 supports 1 or 2 storage processors while the VNXe3300 ships with 2 storage processors.
  • The VNXe has customer replaceable modules so it is easy to service.
  • It supports thin provisioning, snapshots and compression.
  • Both units use a 6Gbps SAS back-end.

All this equates to small cost-effective storage system that is 3x the performance of our current unified arrays! People aren't over exaggerating when they say this release is a game changer for EMC.

But what has me excited about the VNXe is that it has been designed for the application owner. Take a look at the new Unisphere management console:

What you notice on the main dashboard is that there are options for configuring storage based on the application or technology that you are running. Novel concept right? Who allocates storage without considering the application (hey can you provision 100GB of storage for me and then I'll figure out what to do with it)?! No of course not! You know that you have 500 mailboxes of Microsoft Exchange that you need storage for or five Hyper-V servers with 10 VMs on each.

So let's drill into the Hyper-V storage task option! The first screen asks the user to name the new datastore (or the name of the storage container on the array for the virtual machine):

Next we are asked what pre-existing storage pool (think of a storage volume or a collection of drives) to assign the new datastore to. You have the option to mark a new datastore to be thinly provisioned at this point of the wizard.

This next screen provides the user with the opportunity to enable snapshots for the new datastore including the ability to set up a new snapshot schedule. In this example we will choose not to enable snapshots but we can easily enable this feature at a later time.

The next step in the wizard provides the administrator with the opportunity to map the new datastore to a particular host (or multiple hosts). In the screen below we have selected Server-A which is connected to the array via iSCSI. If the host wasn't visible to the array yet we could select an option to add it during this point of the setup.

Finally we see a summary of our selections and Wah-lah! our Hyper-V server is connected to the array and ready to store our virtual machines! It really is that easy.

And this isn't just for virtual environments, check out a few screens from the Exchange wizard. In this screenshot the wizard asks the user what version of Exchange will connect to the array.

And then they are presented with a question about how many mailboxes and the average mailbox size so the array can automatically allocate the right amount of space required for the mail environment.

Awesome stuff and this is just the start of some exciting things coming out designed for the Microsoft Windows & Hyper-V admin! Stay tuned…


Free Windows Training from RunAsRadio

People learn in all sorts of ways and I wanted to take a minute to highlight  one of my favorite sources of free IT training – RunAsRadio.  I can download these MP3′s and listen to them on my commute (when I’m not on the phone).

From their website:

RunAs Radio is a weekly Internet Audio Talk Show for IT Professionals working with Microsoft products. The full range of IT topics is covered from a Microsoft-centric viewpoint.    RunAs Radio was launched on April 11, 2007 with a nod from its sister show, .NET Rocks!, which started as a weekly downloadable mp3 in August, 2002! Coincidentally, the first RunAs Radio show featured Patrick Hynds, who was also the first guest on .NET Rocks!  Richard Campbell, the third co-host of .NET Rocks!, hosts RunAs Radio along with Greg Hughes.   Each show promises to be about 30 minutes long and focused on a single topic.     A full range of audio formats and feed options are offered.

Some of the recent webcasts include (recommended ones in bold).

  • Brian Randell Uses Lab Management for IT Testing!
  • Pam Lahoud is Proactive about SQL Server Performance
  • Brad McGehee on the Remote IT Administrator!
  • Susan Bradley Digs Into Small Business Server 2011!
  • Joel Oleson Talks Office 365!
  • Amit Agarwal Analyzes SQL Traces with Project Lucy!
  • Ned Pyle Migrates from FRS and DFSR!
  • Kevin Kline Recaps PASS!
  • Alan Sugano Helps Us Migrate from VMWare ESX to ESXi!
  • Mark Minasi Ask Questions About Cloud!
  • Jeremy Moskowitz Troubleshoots Group Policy!
  • David Mills on System Center Essentials!
  • Clint Huffman Announces PAL 2.0!
  • Sean Deuby Talks Identity in the Cloud!
  • Jeff Stokes Introduces Us to xPerf!
  • Don Jones Revisits PowerShell!
  • Adam Hall Automates with Opalis in System Center!
  • Jose Barreto Distributes Our File System!
  • Alan Burchill Talks Group Policy Preferences!
  • Rhonda Layfield, Queen of Deployment!
  • Chris Jackson on IE6 App Compat!
  • Stephen Rose on the State of Windows Today!
  • Adam Gent Gets Us Started with OCS!
  • Doug Finke on the OData PowerShell Explorer!
  • Adam Machanic Does SQL in Parallel
  • Isaac Roybal on Microsoft, Cisco and the Unified Computing System!
  • Dana Epp Fixes a Security Vulnerabilty!
  • Kim Tripp and Paul Randal Talk about Learning Effectively!
  • Richard Campbell on the Differences Between IT and Dev!
  • Mark Minasi on the Things Windows 8 Needs!
  • Mark Minasi Digs Into Office and Presentations!
  • Robert Smith Debugs Windows!
  • Vijay Tewari Briefs Us on Hyper-V in 2008 R2 SP1!
  • Joel Oleson Upgrades to Sharepoint 2010!
  • Alex Payne Compares Online Offerings!
  • Daniel Parker Adds Linux to Active Directory!
  • Laura Chappell Wires Sharks!
  • Brent Ozar Masters SQL Server 2008!
  • Francois Doremieux Updates us on Office Communicator!
  • Robert Hamilton Prevents Data Loss!
  • Eriq Neale Puts Macs in Active Directory!
  • Nick Simons Puts Office on the Web!
  • Amy Babinchuk Deals with Security in EBS!
  • Dave Sobel on Virtualization for Small Businesses!
  • Robert Crane Manages Sharepoint 2010!
  • Kevin Royalty Uses Home Server in Businesses!
  • Doug Toombs Gets Us More Free Tools!
  • Rolly Perreaux Manages with Microsoft Project!
  • Clint Huffman Does Hyper-V Performance Tuning!
  • Richard Hicks Gets Us Secure on the ForeFront!
  • Rodney Buike Talks Office 2010!
  • Brent Ozar puts SQL in the Cloud!
  • Stephen Rose Deploys Windows 7 to Businesses!
  • Kevin Kline on the State of SQL Server and More!
  • Laura Hunter Upgrades Active Directory with Server 2008 R2!
  • Mitch Garvis Gets Us Deployed with MDT 2010!
  • Rhonda Layfield Deploys Everything!
  • Bhargav Shukla Goes Deep on Exchange 2010 High Availability and Resiliency!
  • Sumeet Bansal and Chris Featherstone Go Solid State Driving!
  • Bhargav Shukla Deploys Exchange 2010!

VNXe – A Redesigned GUI for Application Owners

You’ve already seen a view a bunch of the screenshots of this new product that caters to the IT generalist.

And – you may have heard some of the cool behind the scenes development tidbits from Chad Sakac’s blog here.

GUI and Redesign

The specific thing I wanted to mention is how important (I think) this new interface design is for EMC, especially with this product.  Well before the product released there was a lot of research done, test system pilots performed, and beta products put out.  Time and time again, three major themes kept emerging among this audience of small/midsized businesses:

  1. EMC is known for being a trusted, reliable brand (that’s a good thing)
  2. EMC is too expensive (ouch)
  3. EMC is difficult to manage*

Then someone really smart said, “We really gotta fix #’s 2 and 3 if we want to play in the lower end of the market.”

And we put to work a team of usability engineer and designers with relative storage-industry outsider status who could take a fresh perspective on what it means to:

  • Provision a shared folder
  • Provision storage for virtualization environments
  • Manage and report on storage utilization
  • and so much more…

I think time will show that this investment was well worth it.   So far, every customer I’ve shown the VNXe demo, cannot believe how easy it is to use!

*This is a realization not much different from the excellent conclusions made by Thomas Goetz in his excellent Wired Magazine article and TED talk on how medical data could be redesigned by talented graphic designers to quickly enable clients and doctors to navigate a set of complex and inter-related factors…

Local Disk or SAN for SQL Databases?

* Over at Ask SQL Server Central, I noticed an interesting question:

Why should one use SAN instead of local hard disk?    I am really a novice in hardware configuration.

Some of the answers were great:

My favorite (written by Matt Whitfield):

1. Performance. Most SANs allow you to attach extra shelves to the array, thereby giving you an easy way to up performance by throwing spindles at it.
2. Scalability. Same reason again, more spindles = more space, as well as more performance.
3. Redundancy. Using a SAN puts your data on a separate physical entity. SANs generally offer the facility for redundant controllers, fabric connections and power supplies. That way you can connect the SAN to multiple servers over a totally redundant fabric to achieve full redundancy. That fabric might be fibre channel, or it might be iSCSI, but either way, it’s the redundant part that counts.
4. Maintenance. When a disk dies (which it will), then on a SAN you just pop out the old disk, and pop in the new. If you have been sensible enough to assign a hot-swap spare or two, then one of those will have already caught up with the array, and the disk you put in will become the new hot-swap spare.
5. Flashing lights. People who look round data centres are typically impressed by flashing lights, and SANs tend to have a lot of them.

That summarized a lot of major points (flashing lights especially :) ), but it’s missing a few details:

* No mention of Automated Data Placement type features which leverage Flash drives – EMC’s Fully Automated Storage Tiering is an example of this
* No mention of advanced protection functionality like replication and space-efficient snapshots
* No mention of virtualization awareness and integration (Hyper-V and VMware)
* No mention of dynamic capacity management functionality – i.e. thin provisioning

I helped write a recent Top 5 Reasons Why EMC Unified Storage for SQL Server document – and I think we did a great job capturing a lot of these details without too many words – and used some really talented graphic designers to turn my stick figures and chicken scratch into works of art.

Check it out!