Aaron Silber came across an interesting KB article that had some very impressive stats hidden in it!  This particular KB article reinforces a pretty standard Microsoft Windows  practice of removing unnecessary items from the session startup process.  Removing items such as the Java Runtime Update Engine, ICABar and Citrix Online plug-in from users’ startup can yield gains of 10 to 15 percent more users on your XenApp Servers! These tests and benchmarks were done by Citrix Labs and affirm the fact that some extra housekeeping and attention to detail during your server builds can really add up.

The KB Article mentions using msconfig to track down these unnecessary processes.  I’d recommend taking a look at Windows Sysinternal’s AutoRuns for a more complete picture of the processes running on your servers.

For the complete details, check out the rest of the article, it can be found here:  http://support.citrix.com/article/CTX124800

As published on VMwareInfo.com

As Published on VMwareInfo.com:

Last month, Citrix released a new version of their Provisioning Services product.  You can download the latest version here.  Lots of new and updated wizards in this version (Imaging Wizard, Enhanced Config Wizard, AutoAdd Wizard), Support for SQL server database Mirroring, optimized load balancing algorithm, Jumbo Packet support and other fixes and tweaks. 

The prior versions already had offline database support for local network fault tolerance and now with support for the SQL server database mirroring, DR solutions will be easier to architect.

Aaron Silber did notice an interesting blurb in the release notes concerning ARP Cache timeouts for Vista/W2K8 servers that you may want to take note of :

TR-8435: Occasionally server stops responding to booting target device requests for 2-3 minutes.

The default lifespan of ARP cache entries was lowered from 10min in Windows Server 2003 to a random value between 15 and 45sec in Vista/W2K8. As a result, the PVS bootstraps are twenty times more likely to experience a timeout during a Vista/W2K8 boot. The workaround is to increase the ARP cache entry lifespan for PVS-bound NICs:

1 Open a command shell window. At the command prompt, enter the following command:
netsh interface ipv4 show interfaces

2. To set the ARP cache entry lifespan to 600 seconds, enter the following command:
netsh interface ipv4 set interface <PVS interface number> basereachable=600000

3. To verify the new setting, enter the following command:
netsh interface ipv4 show interface <PVS interface number>

The “Base Reachable Time” should be set to 600,000 ms, and the “Reachable Time” – to a value between 300,000 and 900,000 ms.
Source: http://support.microsoft.com/kb/949589

About the Author:
Carlo Costanzo is a very active member in the Virtualization Community. In addition to being a full time consultant at IPM, Carlo also publishes tips, tweaks and article posts on his blog at VMwareInfo.com.

Written by Joel Ramirez:  (Big thanks to Joel for providing us with great coverage of the EMC World event! – Be sure to follow him on Twitter)

EMC World 2010 has wrapped up and it was overwhelming. I loved every minute though because I got everything I was looking for to help me do my job better. Between the hands-on workshops and advanced sessions, my peers from around the world doing their best to stump the chump, and the EMC Engineering folks who are the only bigger geeks out there than our collective, it really was a pretty amazing experience.

At the end of the day, there were plenty of techs there who, like me, just want to be confident in what they do. It certainly benefits the employer and the client when consultants are not only competent and experienced, but enthusiastic about technology. Like so many of my colleagues here at EMC World, I was a kid in a candy store this week.

My focus for EMC World 2010 included:

• Data Domain use cases (VMware and Symantec OST)
• Email archiving
• Fully Automated Storage Tiering (FAST)
• FLARE 30 feature set
• Celerra Celerra Celerra

The bulk of my clientele tend to be small-to-medium sized businesses or in the commercial sector. So much energy is spent on the big boys, so I made my trip up to Boston about the little guys that could really use my help relating the technology to their business needs. So the question is, what did I learn about that can help your business?

Email management

Email is a global, horizontal issue for companies. This means that everyone is having this problem, from law firms with a handful of partners to major corporations. I saw a lot of activity this week around continuing to address the issues with managing email.

Email is the one mission-critical application that everyone in your company touches and it can be a bear to manage. Even if you try to wrangle everyone in with their mailbox usage, employees spin off .pst archives or you have to make exclusions for the senior leadership. That’s why you have to stop fighting city hall and use technology to help you out. I always say that you should use technology to enforce policy. I have seen so many organizations modify business rules or lower expectations because the technology didn’t exist (or they didn’t know it existed) to enable the company to do what it really needed to do which, in this case, is control email sprawl and growth and increasing retentions.

Email archiving has come a long way and it is looking really versatile now. EMC SourceOne Email Management impressed me this week with the ability to ingest and limit future .pst generation. There are comprehensive services and new intuitive tools around migrating archive sources for even the largest environments now. You can migrate old EMC EmailXtender archives into a new SourceOne email archive, you can use SourceOne Email Management as the seamless overlay for all of the EmailXtender/SourceOne archives, and you can even migrate Lotus Notes and competitive product archives into SourceOne with the help of a company named TransVault Software.

The situation will dictate, but centralizing your email archive simplifies management, reduces storage cost, and mitigates risk. You don’t want gigs of .pst’s clogging up file shares and you certainly don’t want people walking out of the door with company information that you have no accountability for.

There are a lot of considerations to make beyond an archiving solution. Where do you put that archive, can I point it at a deduplicating array? Do I just shortcut the email or delete it from Exchange to make it only searchable from the archive? Do I really understand how big of a liability it is to manage my email via the status quo? Will my users benefit and will they have to experience performance issues and outages? Hit me up to talk more about this, it’s a big can of worms.

EMC VPLEX

I don’t have too much thought on this right now because, as I said, I was focused on solutions that affect my clientele now. However, I think that this technology is the first realistic and relevant hardware offering from EMC regarding their private cloud message. This technology will make a significant impact to any companies’ needs for highly-available, flexible storage technology that can leverage previous investments.

The VPLEX comes in an a couple of different configurations, Local and Metro. The Metro configuration extends a storage infrastructure by up to 100KM. The goal, EMC says, is to continue to extend this distance in future releases. The VPLEX is an engine, or a multiple thereof for redundancy and performance considerations, that is composed of two controllers. These controllers are BEEFY. Lots of processing power, lots of memory, lots of front end and back end ports. EMC loves to cross-pollinate technologies to make one product even better, and the VMAX engine architecture is all over these engines. They are modular, scalable, and pack some heat. Keep in mind I’m only emphasizing the intelligent strategy behind leveraging proven hardware architecture; the VPLEX is not VMAX in function.

So you take these engines and you spread them far and wide and attach some storage to them. The VPLEX GUI allows you to manage the storage and create a federated storage environment that has the ability to be geographically dispersed. The essence of the thing is that you can break out of any one array. Out of the gate it supports connectivity to several different vendors’ arrays.

Two things about the VPLEX seem to be what makes it tick: cache coherency and dirty region logs. The engineers kept it simple and used logic to eat this elephant. Cache coherency means that any of the engine’s cache communicates with all of the other engines’ caches to determine if a particular block of data is already out there. Remember that we are talking a lot of cache, and that’s the idea. Leave the spinning disk alone as much as possible! Dirty region logs speak to the high availability of the VPLEX. If a site or subset of the VPLEX engines go down, the hosts can still be serviced by the remaining engines. The logical volumes that the VPLEX is presenting to the hosts may have storage arrays or disks that are no longer available due to the outage, but VPLEX can continue to serve up data and, via the dirty region logs or DRL’s, keep track of the activity to resync the downed disk when they become available again.

There is so much more to say about EMC VPLEX, but I’ll save it for another blog post. All in all, a very cool technology that I can’t wait to see become a little more accessible to the SMB and commercial sectors.

Wrapping it up

I talked about some backup and consolidated storage topics and developments in previous posts, so please check those out to read about some of the things that got me excited about EMC World 2010.

Other things that got me excited were the short ribs on Monday, the hot and cold breakfast every morning, the clam chowder and booze tables at the Solutions Pavilion, and the abundance of attractive, really smart ladies of IT

Talk soon,
Joel

So apparently this actually *IS* the year of the desktop!   Forget about when you heard it last year or the year before.  This is REALLY it.. Or maybe next  year.    In any event, Citrix and VMware have both had plans for a bare metal Client Hypervisor for a while now and with Citrix’s Synergy announcement, they are now the first to actually release it.  XenClient was announced and a public release candidate of it is available for download on Citrix.com.  VMware had also announced plans a while back for its own CVP (Client Virtualization Platform) but has not released anything to the public yet.  VMware has had offline experimental support for VDI though leveraging it’s VMware Workstation application for a while now.

The basic idea behind these client hypervisors are to give users the option of running a secure corporate image alongside a not so secure personal image. This choice of personal and work images is suppose to allow companies to blindly encourage employees to purchase whatever hardware they like so long as they can run the client hypervisor.  From there, the corporate image will get pushed down and the user can do whatever they like on the personal image without putting corporate assets at risk.  Sounds good I guess but I am not sure I am sold on the hypervisor aspect of it.  My work and personal life are so blended at times (from an application and daily workflow standpoint) that booting to and from different images would be an incredible loss of productivity. Inevitably, I would begin to break the basic rules of having the two distinct images by merging applications and functions to whichever one would gravitate to become the dominant image in my life.  This of course would put support and security back to the same state as when I had just one image on my laptop.  So I’m just not sure it would work for me personally. 

Offline desktop access might just be a red herring anyway since almost everything is becoming blanketed in WiFi.  Planes, Trains and Automobiles are just becoming moving hotspots.  These spots were the traditional barriers for mobile professionals to relinquish their data.  ‘How do I access my Hosted Desktop/App/Doc while on a plane?’  That question is quickly becoming irreverent.  Assuming latency and performance will be overcome, why not just access it all remotely from whatever device you happen to have in your hands at the moment.  iPad, NetBook, Laptop, Kiosk, Phone, whatever.  Although Citrix was first to release the XenClient, I think they had the right message all along.

For me, Access Anytime, Anywhere, Any device was on the money.

As Published on VMwareInfo.com

Written By Joel Ramirez

I attended the Celerra FAST Uncovered: Cloud Integration and Storage Efficiency session at EMC World this afternoon and, while the name is a bit of a misnomer for the session in two out of three ways, the veteran unified storage platform that is the Celerra is now coming full circle with storage efficiency.

First off, the Unisphere GUI is great. It’s a long time coming and rest assured it can do everything that we’ve been forced to do separately with Celerra Manager and Navisphere Manager. In fact, Recoverpoint management is built into it as well. Now you can set quotas for a filesystem, expand a meta-LUN, and build a consistency group via one management interface.

The session really didn’t speak to the cloud integration piece at all. Perhaps it’s the fact that fully automated storage tiering, or FAST, has such an integral place in the private cloud. A fellow skeptic and I hammered away at the product engineer to test the relevance of the improved feature set of the Celerra. The idea of FAST is to change the entire storage allocation paradigm. FAST represents a pool of disk that can contain SATA, FC, and EFD. It shuffles 1GB blocks of data among the different tiers based upon user policy and the Celerra’s continuous measurement of data access. Hot spots are accommodated, for example, by moving the data to EFD’s. The window of measurement is continually the last hour so the Celerra is using the latest data to determine where to position data among the different tiers of storage.

The EMC Engineering team has an amazing tool to help scope a Celerra using FAST called Tier Advisor which will unfortunately only be available to EMC Internal. This tool is being engineered to ingest .nar files, the performance logs of the Celerra, to gauge what percentage of disk is doing the bulk of the work. It then allows the engineer or consultant to apply different scenarios to determine their impact and dependencies. It gives information such as the change in power consumption, the number of disks and what types required to meet the change in scope, and it provides a graphical representation of the difference between the status quo and the hypothetical input.

Finally, Flash Cache is a really exciting technology. It allows you to expand the cache in a Celerra or Clariion with enterprise flash drives. This is great news for companies who have to buy bigger Clariions that have higher cache capacities just because they need to feed their applications with bursty performance. Instead they can throw a pair of EFD’s in a CX4-120 and have up to 200GB of cache! By using solid state technology and simply adding more capacity to the fastest part of the array architecture, you can worry less, if at all, about spinning disk being the bottleneck. From here on out, I’m all for CX4-120’s with a pair of EFD’s.

Some other notes:

• Celerra Unisphere is a free upgrade if you are under software warranty (you need a CX4) and will manage older Clariion arrays, too.
• Compression is available in the Celerra as a pay-for feature, a good fit for less frequently accessed data on file systems (use with discretion as it it resource intensive)
• Virtual Provisioning will become free
• Auto-tiering, the meat of the FAST technology, is a pay-for feature
• Zero Space Reclaim is being introduced and will be a pay-for feature

As published on VMwareInfo.com:

While working internally on some System Center configurations, Aaron Silber sent over a quick note on what he was up to.  I personally haven’t played around with System Center or Virtual Machine Manager very much and figure I am not alone so I was glad Aaron found the time to send this over to me.  Internally we are running our critical systems on VMware and Aaron has demonstrated some of the integration possibilities of System Center by bridging the VMware world to the Microsoft World so most things, like vMotion, on the VMware side can be managed from System Center.

In his email, Aaron is quick to state that “One of the cool things about the Microsoft suite is the level of integration between the products, so as an example, VMM manages all things virtual and by connecting the Operations Manager piece, all things virtual that VMM sees, Operations Manager can now see and monitor; this extends to the VMware side of things as well. Operations Manager can be configured to report back on all sorts of issues down to a very low level and even to automatically act upon the issues it finds. As an example, I just received the following alert:

So you can see that Operations Manager is reaching out and reporting back that our VMware server has been hitting 80% utilization on enough of a regular basis to cause an alarm to go off. A basic system action would allow the system to initiate Live Migrations or vMotions of the VM’s to free up CPU on the struggling host.”

In this particular example, System Center could be configured to perform DRS like activities on the VMware infrastructure while only being licensed for regular vMotion.  Companies may begin to start debating Good vs. Good Enough.

If you would like to check out some of the features for yourself, Microsoft has released a slew of preconfigured VHDs with evaluation software that is perfect for lab testing.  Prepackaging the VHDs and making them available for download makes it super easy to begin testing these applications.

Download @ System Center Configuration Manager 2007 R2 VHD and System Center Virtual Machine Manager 2008 R2 VHD

As Published on VMwareInfo.com:

As Posted on VMwareInfo.com:
Joel Ramirez attended the Data Domain Best Practices for VMware session at EMC World and got a little inspired.  So let’s talk backup: (Below are Joel’s thoughts).

DynamicOps, Symantec Backup Exec with OST, and Data Domain together provide the most flexible and efficient backup solution for VMware I’ve seen so far.

DynamicOps provides the ability to retire VMs to archive, or nearline, storage.  It is a right-click operation and when you point the nearline target to a Data Domain NFS share, it invokes an automated PowerShell operation that will Storage vMotion the VM to the appliance.  Now your VM, which could be a reference architecture or a model of a customer environment, but isn’t important or used frequently enough to keep on expensive FC disk, isn’t sapping your tier 1 storage.  And it is still available should you need it in the future.  Not only is it available on lower cost storage, but with a replicating set of Data Domain appliances, you can call back any retired VM at a secondary site for test/dev purposes.  You can even Storage vMotion it back to primary storage in production!

With OST, you maximize the investment in your Data Domain appliances.  Say you have a limited budget but absolutely have a need to make the move to backup to disk and deduplication technology.  You buy a smaller Data Domain appliance and enjoy the magic in a box that simplifies backup and makes it more reliable, in addition to the backup data footprint reduction.  Eventually, you’ll execute the disaster recovery phase and need another Data Domain appliance for the secondary site, and by then, you will be using it as a backup target for your backup infrastructure, an NFS share for your SQL dumps, and as the aforementioned nearline storage for retired VMs.  You want to place a bigger Data Domain for longer retentions and archiving at the secondary site. Because the OST API was jointly developed by Data Domain and Symantec, the backup infrastructure is aware of the secondary Data Domain appliance.  You can account for the remote Data Domain within Backup Exec and apply retention policies to keep data longer on the bigger box.

Use the smaller, local Data Domain appliance for 30-45 days retention and leverage the remote Data Domain for 6-12 month or longer retentions.  Paired with OST and DynamicOps, this is a comprehensive backup solution for any VMware environment that maximizes protection, availability, and cost-effectiveness.  With FC disk at ~$5/GB, Data Domain helps extend the value of that tier 1 storage at 1/10th of the cost.  Even SATA is $2-3/GB, so it has an impact on tier 2 storage savings as well!  This is a reference architecture with many benefits and an incredible ROI, check it out.

-Joel
@JoelJet

As published on VMwareInfo.com:

Tuesday, I had a private mode provisioned vDisk fill up completely and needed to resize it quickly.  A little googling around and not quite finding the right answer (I guess I couldn’t get the right search terms down), I headed off to Twitter.   A couple of tweets later and Jarian Gibson came through with an answer! [BTW: If you are at all interested in Citrix, be sure to follow him on twitter (@jariangibson)]

Jarian pointed me to a Citrix KB article : How to Resize a Provisioning Server 5 Virtual Disks.

The KB article details the steps needed to resize the VHD but I thought I’d post this to help highlight the legitimate cases for using Twitter at work.

Thanks Jarian!

As Published on VMwareInfo

If you haven’t had the opportunity to check out the Citrix Netscalers, do it.  They are some pretty neat devices.  The Tech Dev team at IPM has really, in my opinion, shown how cool and flexible these appliances can be with some of the customizations that they are baking into the Netscalers.  Check out the screenshot below.

The IPMer Chatter Scrollbar scrolls Twitter and Facebook status updates from all of the IPMers. 

The Scrollbar is an excellent example of pulling external information into the portal page.  Although our scrollbar flashes through Twitter and Facebook data, any standard or aggregated RSS feed could be fed into this widget.

There is the Support Information tab which allows a user to give all the important Support information to a help desk support person over the phone if they experience any issues connecting to published applications or network resources.

  and of course the  Weather hover!

Not to mention the Single Sign on pass-through for our INTERNAL Timesheet application, our VoIP Portal and Exchange OWA.

I hope this doesn’t come across sounding all marketing and SALESy but did want to post something that highlights some of the enhancements that can be put right into the Netscaler boxes themselves.  I think they are pretty cool and I wanted to see if anyone else was doing anything outside the box on their own Netscalers.

Don’t forget that you can try out a Virtual Edition of the Netscaler @ https://www.citrix.com/English/ss/downloads/details.asp?downloadId=1857216
Both ESX and Xenserver/Hyper-V editions are available for free download.

-Carlo

As posted on VMwareInfo.com:

Today my NAS emailed me.  It was pretty great.  A short little reminder that a new update to the firmware was now available.  I was lucky enough to receive this in the mail from Chad Sakac around Christmas time and it has been great to work with in my home environment.  I currently have 2 ESX servers connected to it via NFS and all the VMs perform like champs (for my home use).  No complaints from the wife or kids! (Thanks again for the NAS Chad!).

Back to the email.  A quick jump to the web based dashboard of the NAS and there is a nice and easy link to download the new firmware.

A quick shutdown of the Virtual Machines running and I was off to the races on my update.  I’ve said it before but if you are looking for a great inexpensive addition to your Home Lab, check out the IX4.

Literally a couple of clicks later and my firmware was updated.  Check out the release notes below for information on the update.

General Description: This firmware update provides performance improvements for your StorCenter ix4-200d. Iomega recommends this update for all units.

What’s New:

Version 2.1.30.8298

• Improved folder permissions, including AFP
   
• Fixed iSCSI segment length
   
• Improved Video Surveillance integration
   
• Fixed minor security issue with external USB devices
   
• Improved Active Directory integration
   
• Improved content indexing performance
   
• Fixed an issue with RAID arrays
   
• Improved performance when discovering multiple devices on network
   
• Restricts thumbnails to files less than 15 megapixels