0800 8047 256

United Kingdom
EACS on LinkedInEACS on TwitterEACS on YouTubeEACS on FacebookEACS on Goolge+EACS on Pintrest
≡ Menu

Faith, Trust and Pixie Dust – What to expect at this year’s Microsoft Worldwide Partner Conference

David Fuller, Partners & Alliances Director at EACS, provides his predictions for Microsoft’s conference…

Next week is Microsoft’s Worldwide Partner Conference (WPC) in Orlando, Florida. My children have just about forgiven me that I will be a stone’s throw from Disney without them. Having assured them that it is a full working week (honest) they have sent me on my way with a shopping list.

June is the end of Microsoft’s financial year so the conference always happens in July to rally the Partners and get us all “super jazzed” for the coming year. I’ll be providing regular updates from the conference but this initial post is my take on what we can expect to hear this year.

It’s been almost 18 months since Satya Nadella was appointed CEO and he has been very active in repositioning Microsoft with his “Mobile First, Cloud First” strap line. A theme in the Ballmer years was the genuine excitement displayed when he and others talked about competition. Nadella wouldn’t be leading Microsoft if he wasn’t a competitive animal however his record suggests his battleground is one based on demonstrating value rather than a fist fight. He still has the likes of Kevin Turner, who isn’t exactly the shy and retiring type, for those who prefer a more confrontational approach.

The top team now is Nadella’s but the realisation of the need to have Microsoft as the default choice for customers from Cloud to End Point has many challenges, not least the still disappointing performance of the phone platform. The departure of former Nokia chief Stephen Elop signalled a change of direction with Nadella creating a new Windows and Devices Group headed by Terry Myerson. This continues the drive towards a single product approach which should enable development at a more rapid pace. I use a Windows phone and really like it but they have a way to go before Android and Apple get too concerned about market share.

Microsoft’s partners will need to evolve if Microsoft is to execute on this strategy. Customers need to consume services if they are to gain sufficient levels of familiarity and expertise that will defend Microsoft from its competitors. I’m expecting to hear more about how Microsoft will support Partner to Partner engagements so that customers maximise the investments they have or are likely to make. Traditional Infrastructure Partners who are building cloud practices may not have experience with Dynamics or SharePoint and although these lines are becoming more blurred it could cause channel conflicts as Microsoft drive customers to consume more and more cloud services. The knock-on effect for customers is more sales calls from more people which could do more harm than good for Microsoft’s reputation.

Cloud has been a major disruptive force, maybe even the biggest in my time in IT. It is reshaping how customers deliver IT, how it is paid for and the business models of Partners. I’m not convinced that Microsoft always understand the need of local markets and is assumed that what works stateside can be applied on a global scale. I understand why they want this but the drive for operational efficiency from within shouldn’t stop the local UK subsidiary from making decisions that impact the unique circumstances and needs of its customers and partners.
Innovate, Engage, Transform and Lead are the four subject areas for this year’s conference with what looks like at first glance a decent balance between technology and business. The proof will be if the sessions are new and updated, or just a rehash of last year.

I’m sure that there will lots of talk about the technology that is about to launch in market. Windows 10 is probably the best publicised but there will also be Surface Hub which is bound to create some “oohs” and “ahhs” from the audience, but this technology could be transformational if deployed with business outcomes in mind.

I would expect to see lots more on how Microsoft is driving customers to Azure with more use cases coming on line almost by the day. DR as a Service, Back up as a Service, Business Intelligence all delivered by harnessing hyper scale will be key messages that Microsoft will want its partners to embrace. As long as they can clearly and easily couple this to Partner Profitability then I’m sure that there will be some big strides in this area during the next Financial Year.

5 years ago when Microsoft declared they were all in for the cloud, partners and customers were asked to believe in Faith, Trust and Pixie dust and that it would all come good in the end. Well today the Cloud is with us and is being embraced on a significant scale and now is the time for these services to be deployed and consumed in the mainstream. Microsoft needs its Partners to do this and so I expect this call to arms to be the main focus for this year’s event.

Often at these events it is what you learn when you are not in a session or a keynote that is the most valuable and I’m looking forward to listening and understanding my peers’ perspectives on Microsoft strategy and how this is impacting them.

It’s important for Microsoft to leave their partners feeling inspired and to help with this a keynote session will be delivered by Tommy Caldwell and Kevin Jorgenson. Nineteen days after they set out to achieve one of climbing’s most difficult challenges, these guys reached the summit of the 3,000-foot rock known as El Capitan in Yosemite National Park, marking the first free ascent of a notoriously difficult section called the Dawn Wall. Overcoming adversity, strength, courage, determination will be key themes but I’m looking forward to hearing their story.

Look out for some more insights from Microsoft’s conference over the next week.



Office 365….the journey continues

The blog below was provided by one of our consultants on the EACS Infrastructure team who specialises in designing and implementing Messaging and Cloud Productivity products…

The quiet after the storm has come… either you’ve gone big bang and migrated all your mailboxes to Office 365 in one hit, or you’ve slowly but surely moved your mailboxes over to the Microsoft Cloud. Time to make a cup of tea and bask in your awesomeness.

With any project, once the work is over and the dust has settled, there are still tasks to complete, loose ends to tie up and servers to be decommissioned from use. The bits and bobs are, in my opinion, just as important as the project itself. I have included some of my own key decision points and notes from working on Office 365 projects here for your perusal. This is not an exhaustive list as every environment is different, but hopefully it will help you once you get to the end of your rollout.

SMTP Relaying

A big decision which needs to be made is regarding SMTP relay. Most companies use some form of relaying, be it scan to email on multi-function devices, or your anti-virus console sending alerts and notifications. Any system which sends mail will likely be connecting to your on premise Exchange server at present. There are various options available to customers with regards to relaying, but for the SMB or Enterprise environment, I believe the most easily manageable options are:

  • A dedicated IIS SMTP Virtual Server. This is configured to connect to an Exchange Online licensed mailbox. This option does require that you pay for an Exchange Online license just for the purposes of relaying, and you will also need to reconfigure your relay devices to point at your IIS SMTP server. If you don’t have an on premise Exchange box, then it is a great option as it is a very secure and manageable way to control relaying. You can find a great guide for settings this up here.
  • Making use of your Exchange Hybrid server as a Relay. If this server is going to stick around anyway for the purposes of user management, why not leave the Hub Transport role installed and make use of it? No additional licenses are required, there is a minimal footprint and no setup changes to your devices or Exchange setup are needed. The Hybrid deployment will do the hard work for you, namely through Remote Domains, the Outbound Hybrid Connector and your Email Address Policy.
  • Custom Inbound connector in Office 365. This involves creating an inbound connector in Office 365 which allows mail to be received from your public IP addresses (this will already exist in some form if you have a hybrid deployment). You can then configure your relay devices to point to your Office 365 MX hostname and configure the appropriate firewall rule/s to allow these devices outbound on port 25. No additional licenses are needed for this, and it has a very small footprint. This setup option is great if you have a single office and no on premise Exchange server, but if you have multiple branch offices then it can become rather complicated to manage.

All of these options help to decrease the attack surface of your network by limiting traffic outbound on port 25. They allow for all relaying situations; authenticated, unauthenticated, external relay and internal relay. You can also use different email addresses as the sending address, so your scanner can still send from iamaprinter@domain.com and your anti-virus console can still send from youvegotavirus@domain.com.

The other options for relaying are direct send and SMTP client submission. However both of these have some kind of limitation, and using a mix of solutions to get around this seems, to me anyway, to be confusing and counter intuitive. Microsoft has published a very useful TechNet article about SMTP Relaying options which may help you decide which is the best option for you.

Post Migration Tidy Up

First thing is first, make sure that all DNS records are pointed to Office 365 if they are not already.

Lock down your firewall. Make sure that SMTP traffic on port 25/587 can only go outbound from your relay devices or central relay point, and block port 25/587 inbound completely. You will also want to remove any rules you had for port 443 inbound to your Exchange server, as this will no longer be used for remote access. Any reverse proxy functionality and/or SSL Offloading will also become irrelevant at this stage and can be removed.

Ensure your Outlook clients are streamlined by configuring Group Policy to only look for autodiscover.domain.com records. This will make sure that your autodiscover lookups complete quickly and that your Outlook clients do not accidentally go looking for an On Premise Exchange Server. I usually disable every option here except for the autodiscover domain and HTTP redirect.

If you use additional Mail Hygiene services above and beyond the Office 365 hygiene service, Exchange Online Protection (such as Mimecast or MessageLabs), you may have already decided to lock down inbound mail flow to only allow mail from these services to come into Exchange Online. This is done by using a custom Inbound Connector in Exchange Online, although there is some extra configuration required if you also use SharePoint Online.

The Last Exchange Server

There are various articles on removing the last Exchange server in your environment scattered around the internet, the most useful one being: Configuring your Exchange Server in a Hybrid Deployment. I won’t add too much to the argument here, however what I would say is that if you have migrated from Exchange 2003 and still want to use some kind of Exchange Management tools to manage your email addresses, then please, please, please don’t leave the Exchange 2003 server there! It is no longer supported, and the Server 2003 OS which it is installed on isn’t either! Install an Exchange 2010 server, you can get a free ‘Hybrid edition’ product key for this purpose and if you want to you can even install 2013 using your free key once you’ve migrated to 2010 and cleanly uninstalled your 2003 server. Exchange Server 2013 is by far the best software to use to manage Office 365 mailboxes. There are licensing based limitations as to what you can do with a Hybrid server (for example, you cannot host mailboxes), but you can manage user objects and use it for relaying, which is more than sufficient.

The tasks which become cumbersome if you remove your last Exchange server are; adding/removing email addresses for users, enabling an Archive mailbox and changing the Owner of a Distribution List. In addition to this, Exchange Online has no capability at present to create an Email Address Policy. This means that if you have multiple SMTP domains and aliases and no Exchange Server on premise, you will need to add all aliases manually every time you create a new user or change SMTP aliases. This is all done via ADSIEdit, which can be a dangerous place to be if you don’t know what you are doing. Generally I would advise that you keep an Exchange Server of some form installed just for management purposes, unless you rarely change email aliases.

My checklist for cleaning up an Exchange environment post-migration goes something like this, but please don’t take it as an exhaustive list as every environment is different!

Cutover/Staged (2003/2007)

  • Convert mailboxes to Mail Enabled Users using the guide below, from the Office 365 Wiki. This removes the mailbox whilst leaving the proxy addresses in place, and then adds a target address to the user, allowing you to manage your Exchange Online mailbox users without using ADSIEdit.
  • Remove or export to PST any remaining mailboxes (Staged migration only)
  • Ensure Public Folders are migrated to Exchange Online or moved into SharePoint Online
  • If required, install Exchange 2010 Management Tools/Hub Transport using your Hybrid edition product key
    • Configure relaying
    • Test management/creation of Exchange Online mailboxes
  • Decommission Exchange 2003
  • Install Exchange 2013 Management Tools/Hub Transport ◦Configure relaying
    • Test management/creation of Exchange Online mailboxes
    • Decommission 2010

Hybrid (2010/2013)

  • Remove or export to PST any remaining mailboxes
  • Ensure Public Folders are migrated to Exchange Online or moved into SharePoint Online
  • Ensure transport rules and/or journaling configuration have been copied over
  • Slim down Exchange server
    • Remove all but single Exchange 2010/2013 Server
    • Remove Mailbox/CAS roles
    • Remove drives used for Mailbox role (Database & Logs)
    • Dismantle Hybrid configuration objects, leaving in place:
      • Outbound Hybrid Connector (if using Exchange Hybrid as Relay)
      • Relay Receive Connector (if using Exchange Hybrid as Relay)
      • Email Address Policies
      • Accepted and Remote Domains
        • Remove resources from remaining Exchange server. I usually go down to 1 vCPU and 8GB RAM

Life goes on

Once all this wizardry is complete, either by your own hands or by the hands of a trusted Microsoft Partner such as EACS, you can sit back and take it easy, safe in the knowledge that you have no more maintenance windows or refresh cycles to worry about. If you have left your hybrid server in place, as most customers do, you still need to treat it as a production server. Patch it and keep it in good shape as you would any other server. Many customers migrate AADSync and Office 365 PowerShell to the hybrid server at this stage and treat it as the Office 365 management server.

But what to do with all this free time? If you haven’t already rolled out Skype for Business, OneDrive and SharePoint Online, now is the time to get stuck in and enable your users to be more mobile and more collaborative! Remember, Exchange Online is just one part of the Office 365 product suite. The fantastic thing about Office 365 is that new services and features are constantly being added to keep you on your toes and make your workplace more modern and productive.

Personally, I always like to keep my eye on the Office 365 Roadmap and the Office 365 Blog so I can learn about the new, shiny nuts and bolts which are planned or in the process of being released. In the Office 365 Admin console, under Service Settings>Updates, you can opt particular users in to the First Release program, giving you and your businesses technology champions the bells and whistles before anybody else!

DD, Infrastructure Team


Flash, Aahhh, The Possibilities…NetApp Makes Fantasy a Reality with the AFF8000 Series

All-flash storage changes the way companies do business by dramatically speeding application performance and improving server efficiency. Unfortunately, the exisitng all-flash options lack core enterprise capabilities such as application integration, built-in data protection, and integration with the cloud. Not to mention there isn’t a flash array out there that can handle unified protocols (SAN and NAS) and deliver top-line performance at an affordable price. Until today, that is. Enter the NetApp AFF8000 Series, a total game changer…

Seriously, what’s the big deal about flash?

Flash storage is any type of data repository or storage system that uses flash memory. Flash memory is ubiquitous in small computing devices and becoming more common for larger applications. All-flash storage changes the way companies do business by dramatically speeding application performance and improving server efficiency.

When flash first emerged it was a “good” storage silo that offered storage administrators the chance to play around with it, see how it performed with their data sets. However, it became just another silo they had to manage, and the general direction companies were taking was to reduce the number of data silos, not increase them. Furthermore, flash storage is notoriously expensive, so storage administrators were reluctant to replace spinning disks with flash for primary storage systems.

The hard truth is that the existing all-flash choices available today lack core enterprise capabilities such as application integration, built-in data protection, and integration with the cloud. Furthermore there isn’t a flash array out there that can do unified protocols (SAN and NAS) and deliver top-line performance at a cost per GB that’s low enough to warrant serious consideration.

The AFF8000 Series, and how NetApp changed the game in a flash of genius…

Everything that’s been holding back the mainstream adoption of all-flash storage has been eliminated in one fell stroke with today’s announcement of NetApp’s All Flash FAS (AFF) 8000 Series. NetApp says the AFF8000 series will deliver the most complete all-flash offering for the enterprise today, an exciting prospect. With the ability to offer unified protocol support on a single storage system, built-in data protection, scale-out performance, and seamless data movement from flash to disk to cloud, this all-flash array is definitely the best solution available for broad enterprise deployment today.

Furthermore, with a new aggressive pricing model, enterprise IT departments are now able to implement a scale-out all-flash storage system at a reasonable price point; this includes software license bundles, support, and a three-year basic warranty. So, for the lowest price out there, you get world-class, enterprise-ready flash storage from the only vendor that can offer enterprise-ready flash storage!

AFF8000 Series Features

Available in four models as standalone systems and in FlexPod; all four models feature:

•Multiprotocol support
•Built-in data protection
•Scale-out performance
•Quality of service

Want to go to the cloud? Great, seamlessly move data from flash to disk to cloud as it ages. You’ll also achieve better ROI for the long term as a result.

Why NetApp?

Yes there are plenty of storage vendors out there offering flash, but only NetApp has the proven track record in enterprise environments and the ability to offer unified protocol support on a single storage system. And only NetApp offers best-in-class application integration for simplified configuration and management of SQL and Oracle databases, virtualized servers, and VDI workloads. In addition, Data ONTAP 8.3.1 offers new inline data compression enhancements including support for all workload environments, which, in some circumstances, brings better performance and lower capacity. Who can argue with that?

Let’s be honest with ourselves, you don’t need flash for everything, but what NetApp offers is the ability to mix flash and spinning disk arrays into the same namespace, thus allowing you to seamlessly move data between flash and spinning disks. No other storage vendor offers this functionality.

NetApp also offers risk-free no-cost options for customers to evaluate their all-flash solutions, making it easy to test the various workloads in your environment and see the performance gains first-hand. Win-win, hands down.

For more information go to http://www.netapp.com/us/products/storage-systems/all-flash-fas/index.aspx.

DB, Infrastructure Team


How can Citrix PVS supercharge your IOPs? Part 3 – Conclusion

Q. What have we learned about Write Cache in the latest incarnation of XenDesktop – currently 7.6 ?

A. If you want your machines to perform faster than traditional storage implement Cache in device RAM with overflow to hard disk. As you can see by the examples above IOPS were increased almost 10 fold.

EACS has also performed the Cache in device RAM with overflow to hard disk on HP BL460 G9’s using XenServer 6.5 and have found the following :
Average I/Os per second 65,657 with 0.2ms average read time and 0.2ms average write time.

blog 4

Speeds of 75,000 IPOS have been reported by our engineers on HP BL460 G9’s using VMware 5.5 as the hypervisor platform.

To read part 1 – The History Click here.

To read part 2 – Testing Click here.

IE, End User Computing Team


How can Citrix PVS supercharge your IOPs ? Part 2 – Testing

EACS has completed some independent testing and has come up with the following results using IO Meter and parameter associated with PVS Targets (80 Writes 20 Reads).

The same parameters for IOMeter were used on each PVS Target, these were specifically set to emulate a VDI workload normally associated with PVS Targets.

Hypervisor Spec : Dell R720 128GB of RAM 2*Intel  E5-2640 processors @2.50GHz, Mirrored O.S. drive and RAID 10 data drive for Virtual Machine storage, running on XenServer 6.2.

Windows 8 machine specs : 4GB of Ram 1 vCPU @2.50GHz, 30GB PVS Disk and 10GB Write Cache disk.

Image 1. Cache on server
Average I/Os per second 2,191 with 6.9ms average read time and 7.3ms average write time.

IE blog pic 1
Image 2. Cache on device hard disk
Average I/Os per second 2,321 with 8.6ms average read time and 6.4ms average write time.

IE blog pic 2
Image 3. Cache in device RAM with overflow to hard disk
Average I/Os per second 19,579 with 0.3ms average read time and 0.3ms average write time.

IE blog pic 3

Look out for Part 3 of this blog, which will look at what we have learned about Write Cache in XenDesktop 7.6.

To read part 1- The History, click here 

To read part 3 – Conclusion, Click here


IE, End User Computing Team