Take the XSIGO I/O Challenge


So you might be a little curious now about this Xsigo solution and what it can mean to your business and your data center. Well step up and walk through a simulation to solve three common tasks the Xsigo way! Take the Xsigo I/O Challenge (also available from www.Xsigo.com) and in 90 seconds be amazed at what you can do.

In the Xsigo I/O Challenge, you will perform 3 common tasks and find how the Xsigo solution makes these tasks much easier.

1. add a 10GB connection to a server
2. migrate NICs/HBAs to a new server (profiles)
3. adjust storage QoS on the fly!

I think you will agree the Xsigo way is what your data center needs!

XSIGO I/O VIRTUALIZATION

One takeaway from VMworld 2009 for me was the need for us to move to I/O Virtualization and a converged fabric solution as our next step in virtualization. We have been evaluating several vendors: CISCO, Brocade, xSigo, etc. but the xSigo solution really caught our attention and focus. xSigo (pronounced "See-go") ran all the demos in the VMworld Both and powered 4 racks in the VMworld Data Center at VMworld 2009!

I/O virtualization is next-generation server I/O that eliminates the cable maze and lets you connect any server to any network and any storage device in seconds (even from a iPhone), entirely under software control. Instead of 8-10 Network cables and 2 FC cables for SAN you use only 2 20Gbps links (40Gbps total bandwidth) per server! In addition, 40Gbs links (80Gbps total bandwidth) per server are also available. Now that beats 10GBe by a long shot! Plus you get full QoS bandwidth controls to guarantee performance levels and they are vmotion compatible.

Your HBAs and NICS are now virtual, but not emmulation! So you never open your servers to add cards again. Profiles are used so your configuration including all MAC/WWNs can easily be moved to other hardware and you can move, add, or change connectivity on demand. Add NICS and HBAs on the fly on LIVE systems! Also, your links are Active/Active not Active/Passive like with FCoE! Configure up to 64 separate connections per server through a single cable, and manage I/O in software, not hardware. Re-purpose servers or add I/O connections in seconds, not hours. With 50% less capital cost and 70% less infrastructure we can get immediate ROI!


Novant is scheduled to begin a POC (poof-of-concept) in March. xSigo came in and worked with us on designing this new virtual I/O rack and datacenter evolution for Novant. So needless to say we are really excited about our upcoming POC. We also learned that deep inside VMware there is a fondness of this solution and it is used for all the roadshow events. I look forward to sharing our experience and hopefully some video with you all!

Dell is also selling the xSigo solution. For a WebEx from dell about xSigo see this link.

For more info visit the xSigo Website or see the link below:

All xSigo Video: http://www.youtube.com/user/XsigoVirtualIO#p/u


Highlights of the xSigo solution:

high bandwidth, low latency, virtual NIC and HBA adapters--save more money than a port based solution say from CISCO

can drive 10G ethernet at line rate

can drive 4GB FC at line rate

demontrated to deliver 20Gbs from one vSphere 4 server

No need to ever touch the server again to swap cards or add cables.

xSigo Director can be upgraded for new I/O needs such as 8G FC, 40G Ethernet, FCoE, etc…

Nehalem servers can generate over 20Gb I/O traffic

Xsigo is 20G or 40G, active/active, for up to 80Gbps bandwidth

FCoE is 10Gb, active/passive

FCoE today enables only priority queuing (at least Cisco’s box does only that)

Xsigo today offers true bandwidth control on storage and network traffic

70% Less Infrastructure: cards, switches, ports, cables

no need for complexity of Nexus 1000V or the extra expense

Enterprise Plus VMware Licensing not needed

50% less capital cost than traditional I/O

30% less power consumption

QoS bandwidth controls for predictable application performance

Guaranteed bandwidth for applications or specific virtual machines.

20Gb bandwidth to each server through just one cable (two for redundancy)

Add I/O to live servers without re-booting

Single-screen view of I/O resources across all servers

1560Gbs aggregate bandwidth for fast throughput

Redundant components ensure system availability

vSphere 4 Security Hardening Guide Released

All vSphere administrators will want to go out at get this new, just release first draft of the vSphere 4 Security Hardening Guide.

The guide comes in five different sections:

vCenter

vNetwork

Host

Virtual Machines

Console OS

Follow this Download link.

vSphere Real Word Thin Provisioning Savings

I wanted to share a real world business example I did today showing the tremendous immediate ROI from using vSphere's new Thin Provisioning. First some really good news, you can start using thin provisioning right after you upgrade to vCenter, as this is a function of vCenter. I have confirmed this works with  ESX 3.5 and higher Server releases.

In my case I had an urgent request for 540GB of extra disk space for a mission critical project. Now I didn't have that space available in my cluster but I did have thin provisioning. So I converted the three virtual machines needing this space to thin and then added the extra 180GB each needed. With thin provisioning these were my results:

Total storage used without thin provisioning: 762GB

Total storage used using thin provisioning: 111GB

Total immediate ROI Storage Savings: 651GB

Not only did I save storage costs, but I was able to go forward with meeting the emergency needs of the business! And now I can go back and request more storage and prepare in case they really use up that storage I provided.

In addition, vSphere is very flexible with being able to convert to thin and back to thick while the virtual machines are running!

Now you see why vSphere is such an awesome product and the best virtualization solution!

Mike Laverick Honors Me with a Post

The famous, vEXPERT, VMware Trainer, famous blogger, and a famous VMware Guide writer that mentored to many of us as we have used VMware the past 9-10 years has honored me with a mention on my blog. You can see post here. We converse a lot via email and I first met Mike at VMworld events. We got to chat even more when he was a featured speaker at the 2008 NC/SC VMware Summit event, of which I was a speaker that year as well. He returned as the featured speaker to the 2009 NC/SC VMware Summit event and I attended nothing but Technical Breakout discussion which he particated in all day. He has really been a big help in my career focus on Virtualization and VMware.

I mention this not to pat myself on the back, but honor Mike and his efforts and all those bloggers out there that share their knowledge with us all. Also, I think a very interesting blog post from Yellow-Bricks.com dealing with why he (Ducan Epping) blogs and his thoughts about the why you should consider doing it as well are very relevant.  A few key statements from Ducan:

• ...decided to look back and try to convince you why voicing your opinion/views and sharing knowledge is important for your personal development and career
• I personally think everyone can benefit from blogging in terms of personal development
• For me the main driver has always been "documenting my experiences" and gaining knowledge
• Of course everyone who blogs hopes their blog will be a success and is out for some form of recognition and or visibility
• Another reason to "why" start blogging is visibility. I did multiple projects where the customer specifically requested me as the consultant delivering it because they read my articles
• First and foremost for personal development, increasing technical skills, knowledge and expertise. This should always be your prime driver
• Just three months after I started blogging I was asked to take on the Technical Team Lead job role. Two months later I had a job interview with VMware and almost 12 months later I was asked to take on the Practice Lead job role. I'm not saying this is only because of the blogging but I am confident that blogging helped me improve my technical skills, knowledge, expertise and increased my confidence

For me, I started blogging in 2006 mainly as a way to start sharing knowledge I had gained and I wanted to give back to those just getting started much like Mike Laverick had done for my career. I was also using it as a means for those I worked with to learn more without flooding their mailboxes. At times though I have not given it the time I needed too. In 2010, my pledge is to start REALLY blogging and sharing personal knowledge and experiences in greater detail. I also want the coveted vEXPERT award as I have been doing this a long time and feel like I should qualify (I totally didn't notice the nomination process last year). If you haven't heard of this you should be using http://virtualization.alltop.com/ to stay informed as well from many of the top bloggers. So my list would look like this:

• give back to the community of those starting out in Virtualization and VMware or want to get better at it
• share interesting news and information from other blogs that I feel are important
• document what we are doing at Novant and obstacles, technologies, workarounds for issues that we run into
• gain knowledge
• personal and professional development
• recognition as an expert in the technology (I have certifications as well;VCP2,VCP3,VCP4) but I think real-world experience is best!
• increase technical skills, I spend a lot of time reviewing lots of virtualization blogs--you can learn a LOT from these sites!
• career development and exposure-along with sites like Linked-In

And I can say blogging does all this. Just in first few weeks of 2010 I have had 3 solid job offers. I love the role I am in at Novant now as the Virtualization Lead (still waiting on that title and pay though). With this role I and designing and building a technology that is strategically important to Novant and the health care and patients and customers we serve. Virtualiziation should be a top strategy your company is involved with as well!

Stay tuned!

Use Powershell to set Round Robin Multipathing

So as we start moving to vSphere 4 I am sure most of us are going to want to start using the new mulitpathing policy of Round Robin...to use all available paths to our storage. I have one cluster with about 32 LUNS in it so making that change for each of the 12 Servers in the cluster was going to be a pain.

But it appears someone has already made life easy. Check out this powershell script that does all the magic. Now it takes a little time to run..but it certainly was faster than using the GUI to make all them changes.
Below is the actual code I used (small change from the link--apparently get-viserver is not valid any longer). This connects to your vCenter then to the server on the 2nd line of the script (testvsphere in this case), then goes through all the LUNS that have 2 or more paths, and sets it to the Round Robin policy. You can watch the progress in tasks pane at the bottom of vCenter.

Connect-VIServer vcenter.nh.novant.net
$h = get-vmhost testvsphere.nh.novant.net
$hostView = get-view $h.id
$policy = new-object VMware.Vim.HostMultipathInfoLogicalUnitPolicy
$policy.policy = "rr"
$storageSystem = get-view $hostView.ConfigManager.StorageSystem
$storageSystem.StorageDeviceInfo.MultipathInfo.lun |
where { $_.Path.length -gt 1 } |
foreach { $storageSystem.SetMultipathLunPolicy($_.ID, $policy) }

VMware goes after Xenapp Customers for VDI with Amazing Offer

This By-The-Bell story reveals some very aggressive and impressive announcements of VMware going after Xenapp customers and move to the  VMware View VDI offerning.
The highlights I found most interesting are below:

• It offers XenApp customers a free upgrade to VMware View 4 Premier, but with an innovative twist that converts a normal CapEx budget expense into OpEx. Customers just make 36 monthly payments equivalent to the Subscription Advantage costs they had been paying to Citrix.
• AltTech estimates that XenApp customers spend on average around $75 per concurrent user per year in Subscription Advantage licensing to remain current. XenApp customers can instead spend the equivalent in monthly payments, without any cost for financing, over a three-year period and will receive a free VMware View 4 Premier license (also licensed by concurrent user).
• View 4 Premier includes the back-end vSphere Enterprise Plus licensing as well as ThinApp for application virtualization.
• The program, which requires customer credit approval, includes 3 years of basic Subscription and Support on the View 4 Premier licenses.
  
• Customers will receive a free physical PC infrastructure assessment with SysTrack VMP from Lakeside Software
• One of the big appeals of the virtual desktop model to IT administrators is its simplicity – it's just Windows running in a VM. It doesn't require specialists who know how to manipulate applications, dive into printing intricacies and work with registry hacks.
• Providing the assessment, migration planning and financing to make this a nearly cost neutral process for the end user creates exciting new sales opportunities for our resellers."

Linked Clones in vSphere 4

I was reading a very interesting article located at NTPRO.NIL about this feature being available it appears in vSphere 4 via Powershell. Now this feature is used in the VMware View product..but kinda neat we can play around with it now for workload not VDI related.

Linked clones, as it states, use the base disk for an OS, makes a snapshot, and then other virtual machines are based off just the snapshot data, which of course saves disk space, gives you one place to do updates, etc. Now I would not use this in production right now..but definitely something to look at.


See a VMware article this feature.
Highly suggest getting the NTPRO.NL PowerPack for Virtualization EcoShell to make this easier.

vCenter 4 Client Install Gotchas

After our very smooth upgrade to vCenter 4 Update 1 the next challenge I found was getting the client out to the desktop and virtual admins team. As I was testing the install process I ran into these issues and workarounds.

• I suggested people uninstall all plugins they had used and the older virtual center update 5 client first. Some of the older plugins are not needed anymore or don't work with vCenter 4. The popular Andrew Kutz Storage vMotion plug-in is a good example.
• The new vCenter install exe file does NOT contain everything you might need for the client install to work now (which I hope they fix in the next releases). On a new install for example it tries to install needed: Microsoft visual J#, .NET, and C++ which are located under the redist folder on the client install cd.
• So for me the single exe install did not work
  
• I copied the install cd to a share and deleted any of the exe and install folders not related to just the client install, so the ONLY menu item that would work from the autorun.exe install was the client install.

Please feel free to share your experience and tips you found for the vSphere client install!

Watch out for Datastore Permissions in vCenter 4

Recently we performed our upgrade to vCenter 4 that I have discussed before on this blog. vCenter 4 introduced Datastore permissions as a new feature. We have a Desktop Support Role setup prior to the upgrade that allowed out desktop professionals to deploy and manage our VDI (no broker yet) workloads.


They had the "Browse Datastore" permission which required before to do their work. After the upgrade to vCenter they could no longer deploy the desktop template. The error message was very clear as to the problem though "You do not hold the privilege 'Datastore > allocate space' on the Datastores connected to the Cluster."

The fix was simple, edit the desktop support role and check the box under Datastore called "allocate space".

Hyper-V 4-part Cluster Problems/Performance:

Just a heads up to a MUST READ series for those using or thinking about using Hyper-V. Techtarget has a 4 part series on Cluster problems and workrounds/solutions to some of them. You will see many times Hyper-V manager is used and not SCVMM to fix the issues. Perhaps hacking a clustered file system was a bad decision instead of developing a purpose driven shared file system like VMware did with VMFS. I'm not hear to judge though, but more to welcome you Hyper-vers to the world of virtualization and share important knowledge.

Of course I would highly encourage you to take a look at vSphere 4. It is an amazing product.

Part 1: Firmware, Drivers, patches and updates

Part 2: Personal workarounds

Part 3: Personal fixes-GUID related

Part 4: Networking Issues and more

CISCO Virtual Summit Jan 21, 2010

CISCO is holding an all day Virtual Summit January 21st. You can learn more and preregister for the event at the main site: http://www.brighttalk.com/summit/datacenter3

Here at Novant, we have been really focusing on moving to a converged networking and storage solution, since VMworld 2009 last year. CISCO has a powerful solution in their new Unified Computing System. We also have been heavily involved looking at a solution from xSigo. With the new Fault Tolerance feature in vSphere 4 pushing the NIC requirements to 8 to 10 or more NICs (especially if your using ISCSI), converged fabric just makes sense.

I would also point you to the following link for an intersting perspective on the CISCO solution:

http://www.bythebell.com/2009/12/cisco-ucs-vs-hp-matrix-strategic-vs-tactical-approach-to-virtualization.html

CPUID Powershell tools for vMotion Compatibility

Looks like Eric Sloof over at NTPRO.NIL has created a PowerPak for Virtualization Ecoshell or standalone powershell scripts to get this info now.

Could be a handy tool to add to your toolkit.!

http://www.ntpro.nl/blog/archives/1389-CPUID-System-Information.html#extended