Every day, companies around the world rely on the features of Microsoft Exchange for their business-critical applications like email, calendaring, contacts, mobile support, web-based information accessing and data storage support. While we’ve discussed the importance of maintaining Microsoft Exchange high availability and steps to simpler Exchange HA, we thought this would be a good opportunity to share a case study from one of our customers.

Connecting Employees, Vendors and Customers Without Interruption

A large U.S. furniture retailer was becoming increasingly dependent on Microsoft Exchange 2003 for internal communication and collaboration, and for communication with both vendors and customers. Since the retailers primary revenue-generating activities relied on e-mail, downtime would have resulted in serious consequences. As a result, Exchange protection became a requirement and top priority for senior leaders.

everRun –An Alternative to Clustering

The IT staff had previous experience with traditional clustering and was looking for an easier, more robust solution. They selected everRun and implemented a solution using a pair of IBM servers with local boot disks and a fibre channel SAN for the datastore. Currently, the system supports close to 1200 users.

No Exchange Failures = Increased Competitive Advantage

In over a year of operation, they have not experienced a single unplanned disruption of their Exchange system. In turn, this has allowed them to keep their revenue-generating activities operating at full speed. As a result, they are currently looking at adding the everRun SplitSite option to allow geographical separation of their systems for additional protection.

Do you have a story when protecting Exchange would have been a better option than what resulted? How did it affect you or your company?

Yesterday, Matt Fairbanks, VP Product Marketing, Citrix, and Jerry Melnick, CTO, Marathon, presented the webinar “Virtualizing Exchange – The Cold, Hard Numbers on Why Citrix XenServer + everRun VM is the Best Platform.” Below are a few of the questions asked from participants with Jerry’s response to each:

Q: What happens in a case of a split brain scenario?

Jerry: In our SplitSite products, we have what we call a quorum services capability – it’s actually an additional component that’s added on to manage split brain and arbitrate when you lose all connections between the two machines.

Q: How long does it generally take to set up XenServer with everRun VM to create this kind of a solution?

Jerry: Citrix people have always mentioned “Ten minutes to Xen” which is a pretty good rule of thumb. We say it’s another ten minutes to add the Marathon software. It’s a simple script that gets run on each host, and then you’re off and ready to protect the machines. The actual protection process itself is really a matter of a minute. The simplicity and ease have never been seen before in this industry with this class of availability solution.

Q: In field of limitations and customers that have deployed this kind of technology, are there any things you would council people to consider to set up XenServer and everRun in the most highly available and robust way?

Jerry: With our system, we provide best practice guidelines for configuring networks availability, etc. One of the beauties of our technology – working in conjunction with XenServer – is that once everything is installed and running, we put everything into an active validation mode so that we know components are configured properly. If something is misconfigured or isn’t running redundantly, you’re going to see the status and receive a warning. A key benefit of this system is you will know how to fix it before there are any problems.

There are many cases in availability systems where you have simple failover technologies: you take an error, you failover, you get to that resource, and then you find out the network or disk isn’t working because it wasn’t configured properly. By having this active validation capability and the report out, the status is being monitored in a simple and reliable fashion – you know when you’re redundant and how you’re going to manage failures.

Thanks to everyone that attended. For those that didn’t have the chance to attend or ask questions, please feel free to leave them in the comments section and we will do our best to answer them.

For those of you that may have forgot to put it on your calendars, Marathon and Citrix are holding a webinar today to showcase the benefits of virtualizing Microsoft Exchange servers. Event attendees will learn:

The webinar will begin promptly at 11:30 a.m. EDT, so if you haven’t yet registered you can do so here.

For those that can’t attend, we will be posting the Q&A following the event.

As we noted in our last post, Exchange High Availability has become increasingly important to businesses of all sizes. To help you get started, we’ve put together these five tips, which are easily-digestible pieces from our “Protecting Microsoft Exchange in Physical and Virtual Environments” white paper.

STEP ONE – PROTECT AGAINST SERVER FAILURES WITH QUALITY HARDWARE AND COMPONENT REDUNDANCY

Server core components include power supplies, fans, memory, CPUs and main logic boards. Purchasing robust, name brand servers, performing recommended preventative maintenance, and monitoring server errors for signs of future problems can all help reduce the chances of Exchange downtime due to catastrophic server failure.

Downtime caused by server component failures can be significantly reduced by adding redundancy at the component level. Examples are: redundant power and cooling, ECC memory, with the ability to correct single-bit memory errors, and combining Ethernet cards with RAID.

STEP TWO – GET RID OF STORAGE FAILURES WITH STORAGE DEVICE REDUNDANCY AND RAID

Storage protection relies on device redundancy combined with RAID storage algorithms to protect data access and data integrity from hardware failures. There are distinct issues for both local disk storage and for shared, network storage.

For local storage, it is quite easy to add extra disks configured with RAID protection. A second disk controller is also required if you want to protect against controller failures.

Access to shared storage relies on either a fibre channel or Ethernet storage network. To assure uninterrupted access to shared storage, these networks must be designed to eliminate all single points of failure. This requires redundancy of network paths, network switches, and network connections to each storage array.

STEP THREE – PREVENT NETWORK FAILURES WITH REDUNDANT NETWORK PATHS, SWITCHES AND ROUTERS

The network infrastructure itself must be fault-tolerant, consisting of redundant network paths, switches, routers and other network elements. Server connections can also be duplicated to eliminate failovers caused by the failure of a single server or network component. Take care to ensure that the physical network hardware does not share common components. For example, dual-ported network cards share common hardware logic, and a single card failure can disable both ports. Full redundancy requires either two separate adapters or the combination of a built-in network port along with a separate network adapter.

STEP FOUR – FORGET SITE FAILURES WITH DATA REPLICATION TO ANOTHER SITE

Site failures can range from an air conditioning failure or a leaking roof that affects a single building, a power failure that affects a limited local area, or a major hurricane that affects a large geographic area. Site disruptions can last anywhere from a few hours to days or even weeks.

There are two methods for dealing with Site Disasters. One method is to tightly couple redundant servers across high speed/low latency links, to provide zero data-loss and zero downtime. The other method is to loosely couple redundant servers over medium speed/higher latency/greater distance lines, to provide a disaster recovery (DR) capability where a remote server can be restarted with a copy of the application database, which only misses the last few updates. In the latter case, asynchronous data replication is used to keep a backup copy of the data.

Data replication is combined with error detection and failover tools to help get a disaster recovery site up and running in minutes or hours, rather than days.

STEP FIVE – CONSIDER VIRTUALIZING EXCHANGE FOR BETTER HIGH AVILABILITY

The latest server virtualization technologies, while not required for protecting Exchange, do offer some unique benefits that can make Exchange protection both easier and more effective. Virtualization makes it very easy to set up evaluation test and development environments without the need for additional, dedicated hardware. Virtualization also allows resources to be adjusted dynamically to accommodate growth or peak loads.

To help you make the business case for virtualization Exchange, we’re producing a live webinar with Citrix on November 11th: Virtualizing Exchange - The Cold, Hard Numbers on Why Citrix XenServer + everRun VM is the Best Platform. Register for the webinar here.

For most organizations, email is single-handedly the most important tool for accomplishing business objectives. Without access to email, companies are at an immediate disadvantage in today’s “I want it now” marketplace. For example, let’s look at the impact email downtime has on productivity: Assuming that your employees are 25% less productive when email is unavailable, and their annual salary is $60,000, then every hour of downtime for an organization of 500 people results in more than $7,200 in lost employee productivity. Can your organization bare a $7,200/hour loss? In today’s economy? Probably Definitely not.

Avoiding the aforementioned consequence is an option, but in order to do so you need to guarantee continuous availability for your organizations email server. According to Paul Rubens at ServerWatch, 2007 forecasts from Gartner revealed that Microsoft Exchange 2007 will own 70% of the email market share by 2010. Now, whether Microsoft will actually return those results, it’s still too early to tell. However, as more and more companies rely on Exchange servers to run business functions, all potential causes of unplanned downtime need to be identified and eliminated.

Over the next month, we will be providing you with some recommendations on how to improve Exchange high availability through planned and unplanned downtime – starting with a webinar on November 11 titled “Virtualizing Exchange – The Cold, Hard Numbers on Why Citrix XenServer and everRun VM is the Best Platform.” For this webinar, Jerry Melnick, Marathon CTO, and Matt Fairbanks, VP of Product Marketing for Citrix Virtualization and Management Division, will team up to discuss how the latest server virtualization technologies keep users continuously connected to Microsoft Exchange servers in the easiest and most effective manner. We encourage you to register online for the webinar if you haven’t already.

Is there anything in particular related to protecting your Exchange severs you would like us to address in the next few weeks? Leave us a comment below and we will be sure to put it on our radar.

Everyone is well aware of the recent market turbulence and the pressure this will put on corporate spending. The economic situation is going to have an effect on every aspect of business action. However, despite budget cuts, I think that the increase in virtualization adoption will continue unabated. Here are a few things I expect:

• Since the beginning of 2008, the virtualization industry has seen changes in its landscape. Microsoft’s Hyper-V introduction has made a lot of noise, as more and more customers are seeing and trying it for themselves. One of the big draws for Hyper-V is its price point. In this market, its price will continue to make it an attractive solution. In fact, I believe the price will dip even further by year’s end, keeping it in consideration for enterprises as other solutions are ditched to shore up costs. As additional vendors support Hyper-V it will only become easier for companies to evaluate, justify, and acquire the solution.

• I believe that not only will Hyper-V demand continue, but people will also begin to adopt more than one hypervisor. Almost all companies have more than one operating system or hardware platform. Since IT departments are adapt at integrating heterogeneous technologies, it is only natural for companies to mix and match hypervisors to solve the specific challenges they have while meeting budget requirements.

• In general, many of the companies that made initial virtualization-related purchases capitalized on its “low-hanging fruit.” Most Fortune 1000 companies have already completed relatively low risk actions and virtualized applications like dev and test systems and file and print. Moving forward, I think that customers will now want to implement virtualization on more of their mission-critical applications. Applications like Exchange and SQL, which both have growing needs for high availability solutions.

• Expect the interest level and the actual deployment of both higher-end availability technology and sophisticated management tools, and the infrastructure that surrounds that, to quickly ratchet up. This will allow people to complete the first phase of adoption and will begin a speedier move towards mainstream inclusion of virtual technology in data centers.

How do you think the rest of the year will play out in the virtualization industry? Feel free to leave a comment with your thoughts; I’m interested to hear the insights of others.

We all like to think that we will be prepared in the event of an emergency, or a disaster. Hospitals exist if we fall sick; fire stations surround us if flames break loose; we are constantly preparing so if a catastrophe strikes, we are ready.

Preparing for a system’s disaster is no different. However, how to go about preparing for an event like this can be confusing. There are many options out there when it comes to protecting your system, each best suited for specific requirement. Unfortunately, many vendors use terms like disaster recovery and high availability interchangeably to describe their solutions when in fact they are usually designed for one or the other.

Disaster Recovery (DR) is the way to recover applications and from a system failure. DR is a reactive solution where if a failure occurs, IT relocates the data, builds the system over, and brings everything back up to working order. This takes time, a precious commodity that typically businesses relying on critical applications don’t have. In addition, recovering applications could bring about a number of side effects which you really don’t want to endure every time some minor failure happens.

But what if I could tell you that instead of worrying about how to recover from a computer system failing, you could simply prevent it from occurring at all?

Disaster tolerance (DT) is a proactive way to prevent system failure from impacting application and data availability. A disaster tolerant solution isn’t going to recover the data if there’s a disaster. Instead it will tolerate the fault if a disaster occurs – keeping an organization’s critical applications up and running at all times. It is not recovery, but rather prevention. And with solutions like our everRun SplitSite, separate servers don’t even need to be in the same building – they can be up to 100 miles apart with fault-tolerant protection between the two locations.

DR solutions are good for applications that can afford some downtime while you recover them. But for essential applications like Microsoft Exchange, SQL, and SharePoint, which need to be available all the time, disaster tolerance is often the best way to go.

So what combination of DT and DR protection would work best for your company’s applications?

Can virtualization move beyond test and developmental in most companies and become a key component of the core enterprise IT infrastructure? Jerry Melnick discusses the necessity of high availability in a virtual environment and how it can be achieved with Mike Vizard of eWeek.

If you’re interested in using virtualization for essential applications like Exchange but haven’t yet because you’re worried about keeping them available, we encourage you to listen to this podcast and learn about the proliferation of different virtual machines, why there are different types available, and how Marathon’s HA solution can make virtualization of applications like Exchange and SharePoint a reality.