Getting Social

Interop debuted its Community Hub among the socially savvy located in the Expo Hall. Aimed at social media types, even the novice tweeter found a ton of valuable info in sessions titled “Building a Global Support Network Using Social Media” and, my favorite, InformationWeek’s “So You Want To Be a Social Influencer.” They even had a space where a video crew could capture your Interop experience. I’ve seen some of those elements used at Cisco Live, but it was nice seeing a tradeshow address the social media needs of its exhibitors and attendees beyond just creating a show #hashtag.

The Main Attraction

Cisco Systems CTO Padmasree Warrior is the main attraction wherever she speaks. Well, except at Cisco Live when her colleagues schedule her opposite William Shatner, John Cleese, and the MythBusters. Sorry, Padma. But at Interop she was the feature presentation. Even with Avaya’s Marc Randall and VMware’s Steve Herrod on the keynote schedule, Padmasree’s only true competition was the cast of Dreamworks Animation’s Madagascar 3. More about that later. What I love best about Padmasree’s keynotes are her slides. Her “Cisco Cloud Strategy” slide in particular offered up how Cisco tailors solutions for building clouds, connecting users to the cloud, and assisting customers with deploying cloud services. Her “Evolution of the Network” slide showed how our networks of tomorrow will have to be prepared for unified workspaces, knowledge clouds, and immersive applications. Another slide detailed how in 2010 there were about two connected devices per person. By 2020, that number will skyrocket to 50 billion total or 6.58 devices per human. I’m gonna have to invest in a pair of cargo pants for sure.

The Silver Lining

Looking over the Interop schedule before the conference, the cloud keynote panel didn’t get my attention. It was stuck between the keynote by VMware CTO Steve Herrod on the software-defined data center and the aforementioned Dreamworks preview. So who could blame me for missing it, right? So as I waited for the Dreamworks show to start, Ellen Rubin took the stage. She’s the VP of Cloud Products for Terremark and is a very compelling speaker. “The hybrid cloud is here to stay because customers want to manage applications in both private and public cloud infrastructures,” she said. My ears perked up. When Rackspace CTO John Engates piped in, as did Steve Shalita of NetScout Systems, I was glad I had stumbled upon the discussion. We could all learn a bit more about the ever-changing cloud landscape. According to the cloud panel, transporting data between numerous cloud providers is a concern for businesses transitioning their IT infrastructures. Moving essential applications to the cloud is a bigger one.

Cloud Computing Gets Animated

HP’s collaboration presentation with Dreamworks Animation SKG was impressive. While a lot of cloud discussions focus on solving the problems of scale and virtualization in the data center, HP shed light on cloud computing as a service and connecting users, like say digital artists across the globe, with needed applications. Dreamworks, the filmmakers behind the Shrek franchise, Kung Fu Panda, and Chicken Run, previewed two 3D clips of the upcoming summer blockbuster Madagascar 3 during the keynote. Just imagine a room full of hundreds of tech geeks in 3D glasses watching a cartoon. It looked great on Mandalay Bay’s theater-quality screens. With the help of HP’s cloud and networking technology, Dreamworks can now produce five feature films every two years. Using HP’s SwitchCloud to store and distribute files, Dreamworks has reduced its dependency on local storage and reduced bandwidth issues. With studios in California and Bangalore, India, Dreamworks digital artists can now work on any project, no matter their location. But it’s not easy. Each film has a ton of data associated with it. Making that data accessible at all times has been easier with HP’s help.

He’s Got Game

I love playing Words with Friends by Zynga on Facebook, even on a plane. So when I saw that Zynga’s CTO Allan Leinwand was speaking, I made sure to add that to my Interop to-do list. Despite overly cute slides in Farmville style, Leinwand discussed Zynga’s transformation from a traditional IT infrastructure to a hybrid cloud solution Zynga calls zCloud. With the addition of data centers on both US coasts, Zynga gamers went from playing 80% on Amazon’s Elastic Compute Cloud to 20%. Leinwand explained that Zynga’s decision to go to a hybrid cloud model was simple: more control. Now if I could just find someone to milk my cow in Farmville when I’m on the road.

Related Post
What Happens at Interop…

Cloud computing is a new business model powered by new technologies. It’s an on-demand, self-service, “pay as you go” model for access to hosting infrastructure (networks, servers, storage, operating systems, applications, support, administration). Cloud providers deliver infrastructure, platforms and applications as a service.

With Cloud Computing the business pays only for what it uses. Compared to traditional models, Cloud can deliver five– to ten-fold improvements in costs and time to market (although 20% is more realistic). Pay-as-you-go eliminates over– and under-provisioning capacity. Over-provisioning wastes money. It also reduces funds available– for other investments. Under-provisioning increases time to value and can result in lost revenue as customer experience degrades. Automated capacity management is built into the Cloud. Adding or removing infrastructure quickly in response to demand offers agility and cost effectiveness traditional IT cannot match. A Cloud has five essential characteristics, three service models, and four deployment models. Each has its pros and cons. Cloud can reduce the time, money, and the number of people it takes to build and deploy applications and related hosting infrastructure. Yet Cloud is not always the right solution.

WHAT YOU NEED TO KNOW

Five key feature characteristics define the Cloud.

1. On-demand self-service to infrastructure, platforms and applications delivered by a “pay-as-you-go” model based on usage.
2. Broad access through mobile phones, tablets, laptops, and workstations.
3. Resource pooling and automation to combine resources into managed services.
4. Rapid elasticity that scales automatically and quickly with demand.
5. Measured service with usage monitored, controlled, and reported.

 Three Cloud service models define decreasing levels of control.

1. Infrastructure as a Service (IaaS) provides network, server, storage, and middleware that IT uses to deploy and run their own operating systems and applications. IT has control only over operating systems and applications. IT can configure storage and some network configurations. IaaS is used to create platforms for service and application development, test, and deployment.
2. Platform as a Service (PaaS) provides application hosting and development tools. Developers create and deploy their applications into Cloud infrastructures. Developers control only their applications and some operating system configurations. PaaS is used to create and deploy applications and services for users.
3. Software as a Service (SaaS) provides pre-built applications, typically available via web browser. Consumers control only application configuration settings. SaaS is used to complete business tasks.

 Four Cloud deployment models describe Cloud ownership and usage.

1. Private Cloud: infrastructure dedicated to a single organization. The organization or a provider owns and operates it. It may be on– or off-premises.
2. Community Cloud: infrastructure shared by a group of organizations with similar needs. One or more of the organizations or a provider owns and operates it. It may be on– or off-premises.
3. Public Cloud: infrastructure for shared public use. It is owned, operated, and hosted by a service provider.
4. Hybrid Cloud: combines services from two or more different Cloud models.

In the next post I’ll talk about what you need to do in order to decide whether or not Cloud is the right option for you and how to integrate it into your business model.

This appendix is significant in that it clearly defines various types of cloud services and the types of clouds in use, as well as different aspects of cloud services. This post will focus on how this appendix describes the various aspects of cloud services.

Appendix C: Service Strategy and the Cloud indicates that there are five characteristics of cloud services. These characteristics are:

• On demand
• Ubiquitous access
• Resource pooling
• Rapid elasticity
• Measured services

I’ll provide a brief description of each of these characteristics, as well as how I’ve recently seen companies offer services that meet these characteristics.

On Demand

An on demand service is a service that can be accessed when and where it’s needed through either the Internet or an Intranet. “On demand” literally means it’s available when the user demands it.

I recently completed an XSLT class. To do the assignments for this class, students activated a host with specific settings. Once the assignment was finished and uploaded to a website, the host was no longer needed, so it was deactivated until the student was ready to begin working on the next assignment. This was helpful because the class required significant configuration to make everything work. Having an on demand service like this meant that students could load a pre-configured environment as needed.

Ubiquitous Access

Ubiquitous access means that various types of clients can use the service. Ubiquitous access requires three things:

1. The use of standard communication methods and network protocols
2. Coarse grained interfaces
3. An effective model for managing security-related aspects

An example of ubiquitous access includes services that allow users to store music, photos, or other information that can then be accessed by various devices such as traditional personal computers, tablets, and smartphones.

Resource Pooling

Cloud services are often provided through a collection of physical and virtual assets that are managed dynamically according to patterns of business activity and user and customer demand.

An example of resource pooling is having a collection of assets that can be quickly arranged as needed to meet customer demand. I once worked for a financial services company that pooled its hardware assets in such a way that we could utilize additional resources from the pool to quickly respond to demand for IT services.

Rapid Elasticity

A service that is elastic can be quickly and appropriately sized in-line with customer demand. What this means is that as patterns of business activity change, the demand for services is affected. A highly-elastic service is able to quickly add more resources in-line with increased demand or reduce resources in response to diminished demand.

Rapid elasticity has existed in IT for some time. An example of rapid elasticity I was recently exposed to was with respect to an IT organization’s network circuits. This organization moves large datasets across a portion of its network. At times, it requires more bandwidth and in order to achieve these needs dynamically reallocates networking circuits and assets. When the need for increased bandwidth diminishes, the organization positions network assets according to normal operating criteria.

Measured Services

Cloud services are often purchased according to a pay per use, or pay per utilization model. In order to offer that type of pricing arrangement, it is critical that there is some method to measure use of the service.

Measuring the utilization of services is nothing new. We’ve been doing it for quite a while in IT. An example of measuring a service comes from mainframe environments but has been used in other environments as well. Many organizations will calculate what is called a “MIPS rate”. “MIPS” stands for “million instructions per second”, and it is the number of instructions that a computer or service can execute per second. An MIPS rate will often not only consider this but also will weigh the average time various instructions take, consider the overall cost of providing the service, and ultimately produce a billing rate equal to the cost of 1 MIPS. I’ve worked with several organizations that billed internally for high-end computing resources in exactly this fashion.

In the most recent version of ITIL quite a bit of coverage is given to cloud-related topics and the impact of cloud technologies on IT service providers. The importance of this is to establish a common meaning of the various aspects of cloud services and how we as service providers can apply these technologies to deliver value to our customers.

Related Posts
Cloud Issues and HDI conference Orlando
Types of Cloud Services — As Defined By ITIL

Related Courses
ITIL® Foundation
ITIL® Service Lifecycle: Service Strategy

I’ve been to Las Vegas before when I attended Cisco Live 2011, but this is my first time attending the Interop IT Expo and Conference. Mandalay Bay, Las Vegas, and Interop — what a combo.

Like the majority of folks attending, as we walk by the slot machines, craps tables, and roulette wheels, I’m looking forward to learning more about emerging IT topics such as cloud computing and virtualization, data center and storage, and wireless and mobility, as well as security and risk management.

With over 100 sessions jam-packed with info from IT experts from across the globe, I am especially impressed with Interop’s list of heavy-hitters for their keynotes. I’m most excited about Padmasree Warrior, Cisco System’s CTO and Senior Vice President of Engineering. I’m a fan. She’s also General Manager of Cisco’s Enterprise Business. In her keynote, she’ll discuss how technologies such as mobility, cloud, and video are transforming IT. I’m sure she’ll be great. Like I said, I’m a fan.

VMware’s CTO and Senior Vice President of Research and Development, Steve Herrod, will also be on hand to discuss the concept of the software-defined data center during his Interop keynote. I’ve heard he’ll explain how IT needs a software-defined data center that’s flexible, efficient, and the peak of automation to be successful in the marketplace.

Avaya’s Marc Randall will discuss application-driven networking. He’ll explain how applications, devices, and networks have seemingly advanced separately since their inception. Current IT trends require a more proactive integration of these technologies that can potentially be met with application driven networks.

Also on the schedule are Google’s Jonathan Rochelle, Dell’s Dario Zamarian, and Rackspace CTO John Engates, as well as Allan Leinwand, who is the CTO of Infrastructure at Zynga, the creator of the Facebook addiction of the moment — Words With Friends. I’m interested to hear what those guys have to say.

Also marked on my Interop calendar is the unveiling of the HP and Dreamworks collaboration, Cloud’s Silver Screen Debut: How HP & DreamWorks Animation Brought Cloud Computing to the Red Carpet.

I’m curious about the downsides to cloud deployment other than the security risks. I hope one session in particular, Building VMware Private Clouds, will hopefully cover how the enterprise can take virtualization to the next level by building cloud networks with VMware’s vCloud Director product, whether they are private, public, and/or hybrid, as well as how to manage ongoing operations.

In Las Vegas, if only for this week of Interop, BYOD still stands for “bring your own device.” There are three sessions addressing this topic: Tablets: Where Do We Go From Here?, Living With (And Prospering From) BYOD, and How BYOD is Driving Change in the Campus Network. Whether it’s supporting these devices and allowing them on your network or learning how to keep them off your network, BYOD has arrived. You might as well be prepared.

One more thing: If your boss cheaped out and told you to read post-show blogs instead of sending you to Las Vegas for Interop, follow my tweets at @GKJohnMarkIvey during the show to keep up with the week’s events. If you’re actually attending Interop, follow my tweets as well (there are three tweetups that I know of), and stop by the Global Knowledge booth to get the last of our Powered Up t-shirts designed by artist Tom Whalen.

My presentation was at 7:30 AM on Thursday morning. I had low expectations for the number of attendees, but I was pleasantly surprised. The rather large room was mostly full. People are interested in hearing what cloud-related incidents other organizations see as well as how organizations are handling those incidents.

My presentation basically concluded with this: based on the data, organizations that address cloud issues with some kind of knowledge management game plan tend to see both a reduction in the number of incidents over time and a decrease in the level of support over time. This is for a couple of reasons. First, with effective knowledge management the user community is exposed to, users develop the ability to diagnose and correct their own cloud incidents. Second, with effective knowledge management the organization’s operational support teams are better able to find knowledge when it’s needed, thus reducing the load on the service desk and other support teams.

Organizations that choose to ignore this by not implementing some type of knowledge management run the risk that the prevalence of cloud incidents will increase over time, and the amount of support required to address those incidents over time will remain high.

It’s clear that as service providers we can’t simply throw money and resources at the incidents we face, regardless of the source. Many of us work in environments that increase in complexity, and, with increasing complexity, the absence of effective knowledge management leads to more incidents that require higher levels of support.

Citrix Systems (CTXS) sees the VDI market as a potential goldmine and has bet much of its future on XenDesktop, a product that is well positioned to capture the VDI space. This post describes five tips for getting the most out of your XenDesktop 5.5 deployment.

Infrastructure Planning and Architecture

As companies move towards VDI, they almost immediately realize that virtual desktops require resources in the data center, many more resources and infrastructure then a traditional desktop environment. In order to get the most out of XenDesktop they will need to answer the question of how to size this infrastructure. Network, Processing, and Storage are all part of the discussion, but how do you scale the environment? How much is required to support the number of VDIs you plan to deploy?

The answers to many of these questions will be unique for each organization, depending on user requirements, applications, and existing infrastructure. For example, most organizations underestimate or miscalculate the network requirements of virtual desktops. This is especially true of organizations that are familiar with XenApp.

Profile Management

No desktop management solution is complete without a good Profile Management design plan. Many customers believe they can take their existing desktop deployment model and just migrate that to a virtual infrastructure. In a traditional desktop/server model, the user’s profile settings and files are stored on the user’s desktop.

Although this might work fine in a physical desktop, in a virtual desktop it becomes problematic. Most virtual desktop deployments leverage a pool of desktops, and a user might be connected to the first available desktop in the pool. Without a Profile Management strategy, the user is presented with a “vanilla desktop” with none of the customizations or files the user needs to do his or her job.

Application Integration

To really get the most out of your XenDesktop Deployment you need to consider the applications. Without applications, all we have is a Start Menu, and we can’t get much work done with that. Corporate applications drive business, and they also drive virtualization projects. The mistake many companies make is to think that they will use traditional methods to deploy applications in their virtual desktops.

Optimize the Operating System for VDI

Once you have built your Master Image, the temptation is to install the required applications and immediately roll it out to your users. In a VDI deployment, there is still more to be done. Specifically, the desktop should be optimized for mass deployment. Optimized is a fairly general term and means different things to different people. In this case, we are talking about the fact that there are certain settings that are appropriate in a traditional desktop environment but which might not be in a VDI deployment. For example, we generally recommend turning off Auto Updates of the desktop and Antivirus since the image is usually in a pooled state and will reset back to its default upon reboot.

Defining Suitable Use Cases

To get the most out of a XenDesktop deployment a full understanding of the Citrix Flex Cast model is essential. VDI is only one option for deploying desktops and applications to end users. It is not necessarily the best option for all users.

A common mistake when deploying XenDesktop is to assume that all users require dedicated VDI desktops. VDI is the most expensive and resource-intensive in the Flexcast delivery model. Dedicated VDI is even more expensive than Pooled or Streamed Virtual Desktops. Other options such as Hosted Shared Desktops or Local Streamed Desktops might be a more cost-effective solution and a better choice for certain users groups.

By following the tips outlined above, your company can make the most of your XenDesktop 5.5 investment. Avoiding common pitfalls and leveraging all aspects of the technology will enable users to leverage the benefits of virtual desktops while keeping costs down and preserving the security requirements of the organization.

Related Courses
Citrix XenDesktop 5 Administration
Citrix Desktop Virtualization Fast Track

Excerpted from Global Knowledge white paper: Five Tips for Getting the Most Out of Citrix XenDesktop 5.5 by Paul Kushlan

Simplified Installation

Beginning with XenApp 6 and continuing with XenApp 6.5, Citrix introduced a number of enhancements to the installation wizard. Before we discuss the roles wizard, let’s look at a typical Citrix Farm Architecture.

Small Citrix farms typically have the following roles:

1. Web Interface Server
2. License Server
3. XenApp Server

Although the Farm Data Store is an essential part of the Farm, it is not technically a “role.” It is a database server, typically SQL, that contains critical Farm Data. During the configuration wizard for XenApp, the administrator designates where the Data Store will be located. If a server is designated as a controller during installation, then it can become a Zone Data Collector (ZDC). If it is designated as a worker, then it will never attempt to become a ZDC and is, therefore, a member server.

Larger Citrix XenApp Farms are much more complex. They include components like Provisioning Services, EdgeSight monitoring, Dedicated Data Collectors, Remote Access Devices and increased redundancy with all essential servers.

Another change to the install process is that there is no configuration required during installation. This makes the install of each of the roles very simple. Once the installation is completed, the administrator is prompted to configure each of the roles.

Single Management Console

One challenge for Citrix administrators over the years has been the need for different consoles for many of the Citrix management tasks. In XenApp 6.0, Citrix has consolidated almost all administrative tasks in one console called the Delivery Services Console. In XenApp 6.5, Citrix has enhanced this console and changed the name to AppCenter.

With the AppCenter Console, an administrator can use one console to perform things like Publishing Applications, Creating Policies, Managing Worker Groups and Zones, along with many other tasks. With one console, you can now manage hundreds of servers and applications in a Citrix Server Farm.

Worker Groups

Worker Groups allow you to manage all servers within an App Silo as one object. This allows you to publish applications and set policies with a Worker Group rather than individual servers, making managing complex real-world Citrix environments much easier.

Another benefit of Worker Groups is the ability to perform Load Balancing between geographical locations or direct users to the XenApp servers closest to their current location.

Policies and GPO Integration

Citrix Administrators can now use GPOs and Organizational Units to manage their Citrix Policies, just like they do their Microsoft Policies, giving Administrators one tool to manage user rights. If you do not have rights at an Active Directory level, Citrix does still allow for IMA-Based Polices that are stored in the Citrix Data Store. These policies are created using the Delivery Services Console; however, they are superseded if GPO Based Citrix Policies exist.

New in XenApp 6.5 is the ability to filter policies in the XenApp Policy Editor. For example, you can look at XenApp 6.0 polices only, or vice versa, only look at what is new in XenApp 6.5.

Merchandizing Server and Windows Receiver

Merchandising Server allows administrators to centrally deploy and update Citrix clients. Merchandising Server is a virtual appliance that is free to download and import within XenServer.

HDX

Many of you have probably heard of HDX. It is short for High Definition Experience and represents a host of technologies (more than 60) that allow the user experience to be the best ever when connecting to hosted desktops or applications. See http://hdx.citrix.com for a list of some of the HDX Technologies included in XenApp.

Provisioning Server

Provisioning Server is a technology that allows Citrix administrators to create a single master virtual disk (vDisk) and then connect multiple desktops or servers to that vDisk, all of which boot simultaneously.

With XenApp, we can leverage this technology to keep all of the Application Servers in a Citrix Farm consistent with the same applications, hotfixes, and patches. This ensures that users receive the same experience regardless of which server they are load-balanced to. With this technology, administrators can now update many servers by updating a single vDisk, making administering large server farms much simpler.

MultiStream ICA

MultiStream ICA is a new feature introduced with XenApp 6.5. In previous versions of Citrix, QoS (Quality of Service) was difficult with the ICA protocol because the various channels all flowed within the port 1494 or 2598 when using session reliability. If QoS were enabled, it would prioritize all types of ICA traffic (graphics, keyboard, mouse, audio, printing, clipboard, drive mapping, etc.). MultiStream ICA Protocol splits virtual desktop traffic into 5 streams – real time, interactive, background, bulk, and RTP Voice – to enable network administrators to prioritize traffic by type and maintain QoS with existing network tools.

Integration with Desktop Director

In previous versions of XenApp, if you wanted to give the Help Desk and other associates within your company administrative access to Citrix, you only had the option of giving them access to the same consoles you, as an Administrator, use. These consoles were notorious for being somewhat complex, and they required some training for new associates. With the introduction of XenApp 6.5, Citrix has given Desktop Director the ability to connect to XenApp. Desktop Director is a web-based dashboard and console that allows Help Desk and other support staff to manage and monitor XenApp sessions.

Pre-launch, Session Linger, and Fast Reconnect

This collection of features improves the user experience by eliminating delays when launching and maintaining sessions. First, with the use of Session Pre-launch policy settings, a session can be started automatically when a user logs on to the farm. By implementing Session Linger policy settings, sessions remain alive for a configurable period before termination, rather than terminating when users close applications. Finally, Fast Reconnect, built into XenApp and requiring no configuration, helps minimize delays when users reconnect to existing sessions.

Related Courses
Citrix XenApp 6.5 Basic Administration
Citrix XenApp 6.5 Advanced Administration

Excerpted from Global Knowledge: XenApp 6.5: New Features and Benefits by Paul Kushlan

If, like many of us, you didn’t get to sign off to attend VMware PEX in Vegas a few months ago, the upcoming changes to the Partner Competencies for Enterprise and Premier Solution Providers may not have come to your attention yet. Competency and certification changes come into force from 1^st October 2012, with the revenue targets being enforced from 1^st October 2013.

So, hot off the press we have highlighted the more poignant changes. Please be aware the details below are relevant to developing countries within the VMware SP Programme.

Enterprise partners are now required to hold a minimum of TWO competencies (of which one must be Infrastructure Virtualisation)

• TWO of the following three accreditations PER competency
  • Sales (VSP)
  • Pre Sales (VTSP)
  • Post Sales (Technical Reqs)

PLUS

• Two core VSPs and VTSPs
• Two current VCPs ( v4 or v5)

Premier partners now need a minimum of THREE competencies (of which one must be Infrastructure Virtualisation)

• TWO of the following three accreditations PER competency
  • Sales (VSP)
  • Pre Sales (VTSP)
  • Post Sales (Technical Reqs)

PLUS

• Four core VSP/VTSPs
• Four current VCPs (v4 or v5)

In addition to these changes, which stand as of February 14^th 2012, new competencies are to be released this year such as VBCA (Business Critical Applications),  the “Management” competency (vCenter suite), as well as the revamp of the Cloud competency which will now be IaaS and focuses on VMware Private & Hybrid Cloud solutions. The partner programme has been upgraded to reflect the improved choice of products now available from VMware, and now they need their partners to step up and take action. October is only a few months away so planning will be key not to miss the October 1^st 2012 deadline.

Linda Smith is a contributor for the Global Knowledge UK blog. For more courses and information, visit Global Knowledge UK.

The unprecedented growth of data, its increasing importance, and business’ dependence on digital information are leading to larger and more complex information storage environments that are becoming more challenging to manage. From the perspective of data availability and protection, the information storage infrastructure is the most critical component of an overall IT infrastructure. It plays a key role in making applications work efficiently, both locally and across multiple sites. With the increasing complexity and criticality of storage, highly skilled and focused storage groups are as mission-critical as the technology being deployed.

Challenges Faced by IT/Storage Managers

These challenges are common to both large enterprises as well as SME (Small and Medium Enterprises) sectors of the industry:

• Managing storage growth
• Designing, deploying, and managing storage in a virtualized server environment
• Designing, deploying, and managing backup, recovery, and archive solutions
• Storage consolidation
• Making informed strategic/big-picture decisions
• Designing, deploying, and managing disaster recovery solutions
• Lack of skilled storage professionals
• Designing, deploying, and managing storage in a cloud computing environment
• Convincing higher management to adopt cloud
• Managing external cloud service providers

Each of these activities is on-going at various levels in each of the companies. Activities such as backup/recovery have been in practice for decades; still the professionals believe that they are not doing enough or not performing these activities well.

Complex Storage Environments

Data centers across the Americas, Europe, and Asia have deployed very similar storage solutions, including hardware and software. The sizes vary based upon business requirements, or in some cases a particular vendor may have a stronger presence in a given environment. However, on the whole, the deployed technology and challenges are very similar.

Storage Technology Segments

Storage technology deployment and its importance to the data center aligned with general market trends for each of the storage technology segments.

Each of the technology segments is unique, bringing its own specific business or operational values. For example, SAN and NAS provide connectivity options with unique functionality, while BR and replication technologies provide options for information protection against planned and unplanned outages. Technologies which enable cloud computing continue to generate significant interest.

Data Movement to Virtualized and Cloud Environments

Migrating to a highly virtualized cloud environment is a significant transformation, requiring a considerable amount of technology and business planning. Companies recognize the need for having an in-house team of professionals to lead the planning, design, and implementation of cloud and related technologies. Since cloud computing requires cross-skill expertise, IT professionals are required to have necessary knowledge across technologies that will be used in cloud infrastructure and services.

Formalized Storage Groups

Storage infrastructure is mission-critical, and a significant part of infrastructure budgets is allocated to storage-related products and services. A well-structured storage group of highly skilled professionals is key to building and maintaining high-performance, highly available storage infrastructures. Job titles and descriptions of dedicated storage professionals are evolving. With the advent of storage virtualization and cloud computing, the industry is expecting the expansion of core skills of storage professionals to include systems and networking skills.

Responsibilities

Storage groups are responsible for the overall planning, design, implementation, monitoring, managing, testing, and operation of all components in the infrastructure. Skills and processes are required to manage these tasks against expected expertise in one or more assigned “specialty” or storage technology segments.

Storage Group Skills and Performance

This is a key challenge for storage managers because it underscores the very real skills gap in their teams. Sub-optimal skills yield sub-optimal storage deployment. On the other hand, a well-skilled team will lead to higher productivity, better technology deployment and management, and optimal utilization of existing staff.

Sources for Hiring and Development

The most significant challenge faced by IT/storage managers is the shortage of skilled storage professionals in the marketplace. In fact, lack of skilled storage professionals is the most serious industry challenge. Considering the aggressive hiring requirements and plans, the lack of skilled resources becomes a serious bottleneck. There is a shortage of skilled manpower in the industry. Capable, experienced, and skilled individuals are usually not available to be hired.

Given that there exists a scarcity of certified or well-skilled storage professionals in the market, managers frequently resort to internal recruitment. Often internal recruitment involves moving an existing valuable employee who has different expertise (such as operating systems, databases, and so on) but has limited storage technology knowledge, which creates a knowledge gap in both technologies.

On-the-job training, technology vendor training, and self development by reading manuals typically cover usage and management of products and technology that is either already implemented or is in the process of being implemented. In addition, there is a need for wider and deeper training focusing on underlying technology concepts, planning, design, and management. This will enable storage professionals to independently and more efficiently design and deploy storage infrastructures that fully leverage the capabilities of all applicable storage technology segments.

Recommendations and Conclusions

From the perspective of data availability and protection, the information storage infrastructure is the most critical component of an overall IT infrastructure. It plays a key role in making applications work efficiently, both locally and across multiple sites. With the increasing complexity and criticality of storage, highly skilled and focused storage groups are as mission-critical as the technology being deployed.

Due to a lack of comprehensive storage technology education in the industry, most storage professionals have relied on on-the-job training, vendor product training, and self development.

Though probably adequate for day-to-day administration, a lack of broad and deep knowledge hampers the ability to make informed strategic decisions and to proactively plan, design, and manage storage infrastructure.

In the previous post, we discussed the need for VXLAN in the cloud along with the issues it solves. In this post, we will focus more on how VXLAN works.

As you can see in the figure above, the packets exchanged between VXLAN enabled devices have four headers that encapsulate the Layer 2 frame. Those headers are:

• Ethernet header
• IP header
• UDP header
• VXLAN header

The first question that might come to your mind is why we need all of those headers? Why do we not just add the VXLAN header? In order to understand this, we need to understand what is VXLAN, and how does it work?

VXLAN

VXLAN (Virtual eXtensible Local Area Network) addresses the requirements of Layer 2 and Layer 3 data center network infrastructure in the presence of VMs in a multitenant environment. It runs across the existing networking infrastructure and provides a means to “stretch” a Layer 2 network. In short, VXLAN is a Layer 2 overlay scheme over a Layer 3 network

Only VM’s within the same VXLAN segment can communicate with each other. Each VXLAN segment is scoped through a 24 bit segment ID hereafter termed the VXLAN Network Identifier (VNI). This allows up to 16M VXLAN segments to coexist within the same administrative domain. Hence we have the usage of the VXLAN header in the figure.

The VNI scopes the inner MAC frame originated by the individual VM. Thus, you could have overlapping MAC addresses across segments but never have traffic “crossover” since the traffic is isolated using the VNI qualifier. This qualifier is in an outer header envelope over the inner MAC frame originated by the VM.

Due to this encapsulation, VXLAN could also be termed a tunneling scheme to overlay Layer 2 networks on top of Layer 3 networks. The tunnels are stateless, so each frame is encapsulated according to a set of rules. The end point of the tunnel (VTEP) is located within the hypervisor on the server which houses the VM. Thus, the VNI and VXLAN related tunnel/outer header encapsulation are known only to the VTEP— the VM never sees it. The VTEP we are talking about here is the Nexus 1000V. Nexus 1000V now fully supports the VXLAN technology. For more information, on that visit: www.cisco.com/go/nexus1000v .

VXLAN In Action

Consider a VM within a VXLAN overlay network. This VM is unaware of VXLAN. To communicate with a VM on a different host, it sends a MAC frame  destined to the target  as before. The VTEP on the physical host looks up the VNI to which this VM is associated. It then determines if the destination MAC is on the same segment. If so, an outer header comprising an outer MAC, outer IP address UDP address, and VXLAN header are inserted in front of the original MAC frame.

Now you might ask yourself why UDP and not TCP — or even why UDP in the first place? Well, the outer UDP header with a source port is provided by the VTEP, and the destination port is a well-known UDP port obtained by IANA assignment. It is recommended that the source port be a hash of the inner Ethernet frame’s headers to obtain a level of entropy for ECMP/load balancing of the VM to VM traffic across the VXLAN overlay, which, as we discussed earlier, will now use VPC or VSS instead of STP which relies mainly on Portchannels.

The final packet is transmitted out to the destination, which is the IP address of the remote VTEP that connects the destination VM addressed by the inner MAC destination address.

Upon receipt, the remote VTEP verifies that the VNI is a valid one and is used by the destination VM. If so, the packet is stripped of its outer header and passed on to the destination VM. The destination VM never knows about the VNI or that the frame was transported with a VXLAN encapsulation.

In addition to forwarding the packet to the destination VM, the remote VTEP learns the Inner Source MAC to outer Source IP address mapping. It stores this mapping in a table so that when the destination VM sends a response packet, there is no need for an “unknown destination” flooding of the response packet.

So to summarize, VXLAN will allow you to increase the number of available layer 2 domains by adding the new VXLAN header. Using this header will allow you to have up to 16 M layer 2 domains, and if two devices hosted on the same physical infrastructure have the same address, it will not be a problem as long as it is configured in a different VXLAN.

Also your VXLAN members can be in the same layer 2 domains or in different layer 2 domains since VXLAN can over-relay over layer 3 domains. Devices with the same VXLAN still can communicate with each other.

Also, your traffic will be effectively load balanced over the over-relay network as the UDP source port will be different for each VM starting a new connection.