In earlier posts I talked about why we purchase storage the way we do, I showed you my zany storage sub-system, and I talked about my goals for testing Storage Spaces with Tiered Storage.
In my last post, I showed how I created a Storage Pool.
Now I’ll walk though defining a virtual disk and a volume using the available wizards in Server Manager.
Start by going to Server Manager and opening up the File and Storage Services, and selecting Storage Pools on the left side of the screen.
In the VIRTUAL DISK window, select TASKS and click on New Virtual Disk:
This opens the New Virtual Disk Wizard – you can read and then hit Next:
A virtual disk is created from an existing storage pool…I have only one, so hitting Next is a good option:
The next screen allows you to assign a name to your virtual disk, as well as let Windows know that you would like create storage tiers on the virtual disk (use those two separate media types defined in my previous blog post to optimize performance):
The storage layout must also be selected. With my storage, I was allowed to select either a “Simple” striped layout or a mirror (simple in this case has no fault tolerance…but I don’t need that right now for my testing) :
In some configurations, Storage Spaces allows you to enable “thin provisioning” of volumes…I don’t have that option here, based on my setup, but it’s a wonderful capability:
I defined my virtual disk to be the maximum size it could be, based on the available capacity of my storage pool:
The Wizard provides the opportunity to review the virtual disk options, and then create the disks:
The progress, and lastly the results are displayed once the virtual disk is created (note the check box to start the New Volume Wizard at the bottom):
Now the new virtual disk has been created, and is visible within the storage management interface:
As noted, the New Volume Wizard will kickoff by default after a new virtual disk is created:
…and our new virtual disk is there on which to create our volume:
Just like with a physical disk, I can create multiple volumes (if desired) on a virtual disk, up to the capacity of the virtual disk, as are the options for assigning a drive letter or mounting the volume as a folder, picking a file system, and assigning a volume label:
One option you may not have run into previously, is the option to enable Data Deduplication. What’s new in 2012 R2 is the ability to specify the “type” of depuplication – dedup for general purpose file server services, or for hosting VMs for VDI:
Once again you’ll see a summary screen detailing the options for the volume to be created, followed by progress and completion status:
The new volume hosted on tiered storage is now available and visible in all those common places we might want to see it:
That’s it! That’s all that I needed to do to create and use my high-performance, coffee table-based storage sub-system!
I’m going to take a break now for a couple of days, and see what questions and comments come in.
Let me know what you think about this tiered storage walk-through, as well as what else you might want to see me test.
Now I would like to show you how I created my high-performance storage pool with the remaining, unallocated memory sticks and external drives.
The first thing to note, is that you do not use the traditional “Disk Management” interface you may already be familiar with to creating storage pools / virtual disks, although the underlying physical disk:
Instead, you use the 2012 (R2) Server Manager user interface – and select “File and Storage Services”, which will bring up a comprehensive set of storage management screens within Server Manager:
Notice that you can see the same defined volumes as well as attached disks in the user interface, by navigating through the various options on the left side of the screen:
To create my storage pool, I navigate to Storage Pools, and then can see the available physical disks:
The information included in the “PHYSICAL DISKS” window, is extremely valuable, as it includes key information like the Bus (all my disk is USB attached), and the Media Type (listed as Unknown, since USB doesn't apparently report that or the spindle speed).
It’s important for Windows Server 2012 R2 to know the medial type, so it can properly provision storage tiers (I’ll show you how to set media type in bit, once we create our pool.
To create a new storage pool, go to TASKS in the top right, and select New Storage Pool, which will open a New Storage Pool Wizard:
You enter your new Storage Pool Name, and hit Next:
You select the available locally attached disks to add to the pool and hit Next again:
The Wizard displays a summary, and lets you confirm the creation of your pool:
Depending upon the number of disk and the performance of your gear, it could take some seconds to complete the creation of your pool:
You will now see your newly created Storage Pool, and no longer see the unassigned disks:
Interestingly, you will no longer see the disks assigned to the new pool in the traditional Storage Management interface:
This really highlights the point that you should perform your storage management tasks from within Server Manager!
Remember, all your storage is right there in Server Manager – I showed you the volumes before in Server Manager – that hasn’t really changed, as I haven’t created a virtual disk from my Storage Pool and subsequently a volume.
Let’s go back to Server Manager and click on our Storage Pool:
Remember I mentioned that the “Media Type” was not set (displayed as “Unknown”) – we need to update that for each drive to properly utilize Tiering.
We just have to touch each physical disk in our pool with a little PowerShell, and we are all set:
Note that you can’t update the media type until after the disks have been added to a pool (why I waited until after the storage pool was created).
Now the disks have the properly set Media Type to proceed:
The storage pool is created!
In my next post, I’ll create a virtual disk and a volume.
To bring you up to speed, I’ve loaded the preview of Windows Server 2012 R2 and built an extremely low cost storage subsystem out using a USB hub, some memory sticks, and USB attached SATA drives detailed in earlier posts. This is all in an effort to give you and me a look at the new tiered storage feature that is part of Storage Spaces within Windows Server 2012 R2.
I don’t have an unlimited budget, or unlimited time, so my initial testing was somewhat limited.
I’m using a laptop with USB 3.0 that has two physical (SSD) drives in it, with the additional external drives attached via USB 3.0. For fun I’ve included a Powershell listing of the drives (using get-physicaldisk):
Disk2 and Disk3 are 32GB USB 3.0 flash sticks, with the other disk (Disk4-7) being USB 3.0 attached SATA disk I salvaged from old laptops and what not.
For testing, the first step was to create a “baseline” environment to act as a control. I configured one of the four spinning disks (Disk4) as a standalone to perform the same copy tests I would use on the Storage Spaces volumes.
Now I’ll move on to how I configured the volume in Storage Spaces.
I wanted to give you some background on that high-performance, coffee-table-based storage sub-system I showed few posts back.
My goal was to create a high-performance storage sub-system which I could use to experiment with Storage Space and the new capabilities in Windows Server 2012 R2, but do it on the cheep. USB 3.0 supports extremely fast data transfer rates, and I just upgraded to a laptop with USB 3.0, so USB 3.0 attached storage seemed liked a great option (plus I already have two USB 3.0 32GB flash drives for use with Windows To Go that I could repurpose as my SSD tier).
I picked up a powered 7-port USB 3.0 hub, and a bunch of USB 3.0 SATA disk enclosures (bus powered), and found some salvaged SATA disks from old laptops (250GB-320GB), and plugged it all together.
I loaded Windows Server 2012 R2 on my laptop (which already had two drives installed) and added some drivers.
Once everything was plugged in and spinning, I had a bunch of disks, ready to use!
There’s a reason servers spin up one or two drives at a time…it’s because STARTING a spinning hard drive TAKES MORE POWER than keeping a drive in motion! To get my storage system working properly, I have to plug each drive enclosure in to the hub one at a time, so I don’t draw too much current all at once…sort of like Apollo 13.’ If you want to try something like this yourself, do not use bus powered drive enclosures to limit the power draw from the USB hub. My hub has a 1.0 Amp external supply, but I still have to manage drive startup manually with all of the bus-powered enclosures.
The drives on my USB powered storage spin town pretty regularly, based on all the great power management “features” backed into USB and Windows. That’s not optimal for a production (or perhaps even a test) storage rig, but it is what it is. Be aware that this configuration is not meant to be used for real life storage applications, but simply for me to show you how to use Storage Spaces and configure Tiered Storage.
Next up, I’ll cover my goals for testing.
At home I typically purchase storage based on my requirements for storage capacity. For a laptop or desktop, the question might be, “ will the 250GB drive be enough for all my kids stuff, or do they need something larger?”…or,” How large a USB disk do I need to backup all my pictures?”
I spent a number of years “back in the day” working with Exchange, so at work I’m much more sensitive to the performance of a storage platform or solution, with the storage capacity sometimes being a secondary consideration.
Storage subsystems for high-performance applications like databases and messaging are sized and purchased for their ability to support a particular peek load of IOs Per Second (IOPS) based on the anticipated read and write behavior of applications.
When selecting a storage subsystem storage architects have traditionally considered the IOPS that a particular model of disk will support, overall number of disks, the impact of the “protection level” (RAID 1 / RAID 5 / RAID other) on performance, and the bandwidth of the disks and the end-to-end system.
Generally speaking, if you wanted twice the performance from storage, you would double the number of spinning disks in the system. You could also perhaps purchase “faster” drivers. Perhaps your server or sub-system was using SATA attached disks rotating at 7,200 RPM. If you replaced them with drives spinning twice as fast (like much more expensive 15,000 RPM SAS attached drivers) you would be able to support more IOPS (at a much higher cost).
Yes, of course there are other considerations (impact of things like controller-based caching, replication technologies, cloning and snapshots, backup tools, rebuild speed), but IOPS per disk has always been a core consideration, and performance of many storage subsystems is ultimately bound by the IOPS of the attached, individual spinning disks. The performance of subsystems based upon traditional spinning disk is something that can be estimated, and has been for years.
You really don’t need that high IOPS capacity for all your data – not all of your files or their chunks are being used all at once.
What if you could use those same spinning disks…or less expensive slower disks for the same workloads, but still meet the same peek IOPS-related demands? That is exactly what the tiered storage capabilities in Windows Server 2012 R2 can do!
Tiered Storage will let you deploy Windows Server 2012 R2 file services for exceptional $ / IOPS, by allowing you to do the following:
There was a great session at TechEd this past June, 2013 titled, ”Deploying Windows Server 2012 R2 File Services for Exceptional $/IOPS”.
This is a MUST WATCH SESSION if you work with Windows and storage. The presenter went into great detail on the economics of IOPS, as well as how to use Storage Spaces with Tiered Storage…this is what motivated me to create that high-performance, coffee-table-based SAN (yeah, SAN is a stretch) that I showed you in my last blog post.
Next up, I’ll go into details about my cool table-top storage rig.
I took a BUNCH off time off from blogging…but I’m back now and intend to share some cool new stuff around Windows Server 2012 R2 (preview at this point), starting with my experiments with Storage Spaces and Tiered storage.
First let me say that I hadn’t worked much with Storage Spaces in Windows Server 2012. I didn’t have a compelling reason, since all of my existing servers already had RAID controllers or external SAN storage which did a pretty decent job of protecting data. That changed in Windows Server 2012 R2 with the introduction of tiered storage! This innovative little jewel allows you to vastly improve the performance of traditional spinning disks with the introduction and easy integration of faster SSD-based storage.
Over my next few posts I’ll try to make the following clear:
I’ll leave you with this for now. I didn’t go out and make a big investment to build my flexible storage environment…I literally took some old parts off the shelf plus less than $90 to create a high-performance, multi-disk array which demonstrated better storage performance using the preview of Windows Server 2012 R2 than many large SAN arrays that I have used.
Where do I keep this monster performer of a storage array? Well, on my coffee table, of course!
The capabilities of Microsoft Lync (IM, presence, voice, video, web conferencing, other) are such that it eliminates the need for some overlapping technologies. With Lync I don’t (for example) need to use a separate web conferencing infrastructure like LiveMeeting / WebEx / GoToMeeting – because this capability is built right in.
When I used LiveMeeting (before Lync) I would sometimes upload videos that I would play in meetings – LiveMeeting would trickle the video down to the attendee’s PCs and issue start / stop commands and everything (sort of) worked great. While Lync does have the ability to upload presentation content (PowerPoint files), it no longer supports uploading of videos. You can imbed videos in your PowerPoint files, and play then that way, but this makes your PPTX files quite large.
I thought that there must be a way to feed video directly into the Lync video window (intended for web cam use). A co-worker suggested trying one of the many “virtual” web cams, so I did, and it WORKS GREAT! We tried a few different (unsupported) virtual web cams, but the one I liked the best was ManyCam. Before you go off, download and install it know that it has one of those REALLY annoying install wrappers that will change your default search engine, install an unwanted tool bar, and generally mess up your browser settings. Once it was installed, I stopped Lync, stopped ManyCam, restarted both, and then ManyCam appeared a video device for me in Lync!
(You can see the sort of videos I have laying around on my PC – high quality stuff!)
Chris Norman thought that another great use of this sort of software the capability to merge the feeds of multiple web cams (his post is here).
What? You want to disable IM in Lync? Why on Earth would you want to do that!
It turns out there are a number of reasons that some organizations might want to, including:
Whatever the reason, there’s a simple way to disable IM via a registry setting on the Lync Client. It’s actually a hold over from the Office Communicator client, and the setting is described in BLOG post here as well in a KB article (KB954648).
All that’s required on the client machine is:
What happens? The user can’t participate in IM conversation, using the Lync client, that’s what happens! One easy way to see the impact is by looking at contact cards before / after the change – the IM “bubbled” is greyed out. Two pictures of contact cards make the point best:
I mentioned in my last post that I’m now focused on Unified Communications at Microsoft. I’ll go deep on what Unified Communications (UC) is later - should you care.
The BIG news is that as part of my new gig, I get to use lots of cool phones and devices that integrate with Office Communication Server (OCS) and Microsoft Lync (the new version about to be released).
Many of the folks at Microsoft no longer have regular telephones on their desks – many locations have moved to a completely “soft” phone model (voice services through your PC) using a USB headset, USB speakerphone, USB handset, or even the old PC speakers and integrated microphone.
Sure, there are “hard” phones all over Microsoft – some traditional “copper” attached phones that talk to a vintage PBX (like in my rural Microsoft office) as well as network-based “hard” phones that plug right into an Ethernet jack in the wall (these are everywhere in Redmond) – like the awesome Polycom CX7000. The CX700 is great, but it does lots more than just let you dial a number using the key pad!
The Problem for lots of organizations has been that there’s a real need for IP “hard phones” – standalone devices that look and act just like an old telephone. Many users don’t adjust well to a handset / headset / speakerphones without a physical keypad. I have phones I love/use every day without keypads (like the USB Polycom CX200 or CX100) and THEY ROCK but they aren’t for every situation.
Some users and situations mandate a low cost solution with a keypad (like with the USB attached CX300). I REALLY LIKE my CX300…it has almost everything – great price, USB interface, speakerphone, and dial pad. The one limitation (over my CX700) is that it is not a stand alone device – it still requires a PC to function. I have other USB devices with a keypad I use a lot (like the Jabara 520 and the Plantronics 1100M ) This is not a draw back in many dedicate scenarios (where a user has a PC at work or home), but it still doesn’t address a stand alone “hard phone” scenario (like, say, my office at home…where I may not always have a work PC fired up all the time). Yes the CX700 can do that for me, but it’s a high-function, executive phone – and yes, newer more cost effective devices (like the Polycom CX500) are due out soon to work with Lync.
I know I’m completely ignoring video and headsets in this post (give me time!), but I don’t personally use those as much.
I just got a Snom 300 with updated firmware for work with OCS (and Lync) and it’s A GREAT LOW PRICED IP PHONE! I opened the box, plugged it in, and with limited configuration it worked with Lync from my house! I actually got a call from someone at work on it just after I connected it to Lync.
How hard was the installation? I opened the box, plugged the handset in and after that yes, setting it up was only a tiny bit harder than plugging in an RJ11 cable, but not really that much more difficult.
To set it up I had to:
Right away on my home network it would not work (Lync authorization failure) . A quick check of the Snom knowledgebase and I new that the phone’s time was not close enough to the time in my (Microsoft’s) infrastructure. I set it to a known time server on the the Internet to get back on track via the web UI (browsed to the IP address of the phone at my house – something like /advanced_network.htm">/advanced_network.htm">http://<phone IP>/advanced_network.htm) and everything worked like a champ!
I’ll try to pull together at more detailed post on the provisioning process later (let me know if you are interested!).
Sorry I haven’t posted in a LONG time, but I was super busy getting “all in”.
I changed gears at Microsoft this Summer after my presentation at the Red Hat Summit – Microsoft has lots of credibility (and customers) in virtualization now, and I thought I was ready for new challenges.
I’m focusing once again on messaging (what I did BEFORE joining Microsoft) – specifically Microsoft Exchange and Unified Communications. Voice and electronic communications are evolving rapidly, and I’m excited to be part of it – including things like the upcoming release of Microsoft Lync, the cool new archiving features in Exchange 2010 SP1, and the impact of all this cloud “stuff” on costs / day-to-day administration.
I’ll continue to post information about virtualization and Windows Server that I think may be valuable, but I’m certainly focused on messaging and collaboration these days.
If you have any topics in this area you think I should dig into, let me know! Maybe I should write up details on Exchange Hosted Archiving and how it compares to what’s in Exchange already (and if you should keep paying for a 3rd party archiving tool!)? What about how Lync provides telephony support? Let me know!