Always a content-rich conference, this year’s show does not disappoint. I look forward to all the cloud announcements that have already begun to come out – and just like customers I face the same challenge of separating the “wheat’ from the hype/chaff. A new game – guess how many companies will announce themselves (new and reinvented) this year with some form of “cloud” or even better “clou” in their names…

Back to VMworld – as I listen to Ed  Bugnion from Cisco plug UCS in the background – this year, VMware’s theme is Virtual Road, Actual Clouds. As opposed to the 3 or 4 actual case studies at the show last year, more customers with public, private and hybrid cloud deployments are here this year and ready to tell their ROI stories. An example from the first session I’m in is BT, which went on an 8-month virtualization (maybe cloud) journey and here were the results:

• from 700 racks at 8 sites to 40 racks at 5 sites
• from $2.4 million spent anually for power to $110K
• from 9309 network ports to 840
• from weeks to provision to 12 minutes
• from 96 hours to backup to about half an hour
• from onsite management required – to fully remote management enabled

Pretty impressive – but to be taken with a grain of salt since BT is of course a vendor delivering virtualization/cloud services themselves.

But more than ROI and case studies, another theme is emerging which makes me feel like I’m back at a Gartner show. Service-oriented architecture. IT as a Service – ITaaS. Which comes first – the analyst or the big vendor messaging?

More on this tomorrow from the VMworld keynote session. Register to see it streamed live.

“The committee is supportive of the concepts contemplated in the e-gov account for fiscal year 2011, namely, moving agencies to cloud computing through pilots…The committee is concerned that the electronic government initiative does not provide sufficient guidance regarding consolidation of federal agency data centers into data facilities with multiple federal tenants.”

What does IT management look like in the future? From Accenture’s point of view from working with many forward-thinking clients who are looking to cloud computing, enterprises will manage less infrastructure but will need more expertise in managing end-to-end performance from multiple suppliers. They foresee a greater focus on ITIL-like processes including service management, governance, service integration and more.

Gartner and IDC have both lowered their IT spending outlooks for the rest of 2010. While both predict that IT spending will continue to rebound from the lower amounts spent during the peak of the global recession of 2009, the pace for the second half of 2010 will be slower than anticipated.

It’s always fun to look at lists and see how much you agree…here’s a good one: Game Changers: 12 technologies that changed everything.

Finally, there’s a silver lining in every cloud. One of the first push-backs regarding the cloud is security. View the slideshow on cloud security pros and cons.

(Photo credit: ElviraZ)

Since I last saw him, Cameron Haight’s been very busy. Well, he’s always busy but this time he’s been busy talking to all the major cloud providers in order to distill some lessons learned from their cloud computing operations management for the rest of us.

For most of us, we’ll simply never get to the economies of scale that Amazon or Google can – pretty straightforward arithmetic. But beyond economies of scale (and more important in terms of operations to emulate), cloud providers are able to push their costs way down because of their non-conventional approaches to “technology, people and process”. And many of you service providers might think you’re looking in a mirror too.

So what do public cloud providers do right? And what is this different operational philosophy?

1) Cloud providers try to remove complexity – Think of complexity as “debt” that you have to pay at some point or go bankrupt. Complexity is a business-killer. Wherever possible, cut out complexity. Simple concept but as anyone who has tried to do it, far harder to do in reality. This simple concept is at the core of how cloud providers think – out of necessity and because it just makes sense.

2) So what does this mean for best practice frameworks like ITIL? Cameron has seen a lot of ITIL projects fail over the last 9 years he’s been with Gartner. At heart, ITIL adds to complexity in any organization in the interests of doing things the “right way”. To cloud providers, ITIL is a 4-letter word…They do “process” of course, but it’s not ITIL process.

3) Break down silos (traditional) and roles – a concept near and dear to our hearts here at ScienceLogic. Silos, while they are a natural outshoot of point solutions and building functional expertise, create a highly inefficient operational environment. Service delivery (and troubleshooting issues) doesn’t have functional boundaries, why would you set up your operations that way?

4) Here’s the zen moment: In the cloud world, it’s no longer necessary to keep a system up at all costs. Failure happens all the time – but they accommodate that as part of their system architecture/design. Accepting failure actually makes some things easier. God grant me the serenity to accept that systems fail and the tools and process, hopefully with automation and integration built-in, that let me deal with that failure in the quickest, most efficient way possible. (Don’t fix it, swap it out!)

5) Simplicity leads to Manageability

6) Management influenced by Design – management should lessen complexity, not add to it

7) Think out of the box! Take a leaf from Cameron’s book and do a completely different presentation that is better suited to your audience than the original and gets your point across even better.

.8) It’s not about best practice – but rather “just enough” process

9) Culture – traditional IT and enterprises have a “follow processes and adapt to culture” mentality vs the cloud providers’ mantra of “unmanagement” and “hackers are heroes”

10) The big cloud providers are growing their own new staff starting with courses at university (not an option for most of us but we can dream)

And here’s a final little lesson from Amazon which apparently believes in “two-pizza teams.” In other words, keep your (support) teams small enough that two pizzas can feed them. Operational agility – it’s a good thing.

Who’s just starting out with Virtualization – 1 lone hand popped up

Who’s in the middle of their Virtualization journey – almost everyone

Who’s so far ahead they’re looking down on the rest of us – a few interesting souls

Big points:

There will be more VMs deployed during 2011 than in 2001-2009 combined.

CIO’s consider virtualization the No. 1 technology trend – the business impact is huge (2009 Gartner CIO Survey)

In 2005, VMware’s installed base market share was 100% By 2012, it will be 65%.

In 2009 – midsize company survey (100-999 employees)

About 30% started server virtualization in 2008. another 30% starting in 2009. Midmarket taking off for server virtualization projects.

Enterprises virtualize as they modernize and replace equipment. This means they will be done with their server virtualization project in something like 7 years (server/hardware lifecycle). Midsize companies tend to virtualize as part of one or 2 projects, and will be done in just one or 2 years.

Percentage of installed x86 Workloads Running in a VM

• 2008 12%
• 2009 19%
• 2010 28%
• 2011 38%
• 2012 48%

In the next 2 years, % of installed VMs will double – midmarket growth is a big driver.

Server virtualization market (installed):

2009 – 10.8 million VMs

• VMware 84%
• Microsoft 11%
• Citrix 4%
• Others 1%

by 2012 – 58 million VMs

• VMware 65%
• Microsoft 27%
• Citrix 6%
• Red Hat 2%
• Others less than 1%

VMware’s dominance in the enterprise will continue; true, they will have a smaller share but of a much larger pie.

x86 Server Virtualization Infrastructure Magic Quadrant (2010)

VMware stands alone in the Leaders quadrant

Microsoft alone in the Challengers quadrant (still immature product)

Citrix alone in Visionaries quadrant (problem with execution)

And everyone else – from Red Hat to Oracle/Solaris – in the Nice Players quadrant

How important will heterogeneous virtual machine support be in your enterprise by 2012?

• Very important 49%
• Somewhat important 35%
• Not important 16%

BUT – Gartner thinks it’s more important to simplify than to plan for managing multiple. Also for those organizations that have multiple hypervisors, different VM types exist in different business units for example. Multiple hypervisors not usually used or managed together. By 2014 – fewer than 10% of enterprises with more than 1000 employees will manage two or more hypervisors together.

The Future of Server Virtualization:

• Automated long-distance live migration
• Fault-tolerant VMs
• Service-smart VM management
• Private cloud: self-service, metered usage (look for VMware’s “Redwood” offering late this year)
• Migration (including live) between on-premise and off-premise (e.g., cloud)
• Single console management for on-premise and off
• Disaster recovery either on-premise or off

“Virtualization leads inexorably to cloud computing.”

Stage 1:Server Virtualization

Stage 2: Distributed Virtualization

Stage 3: Private Cloud

Stage 4: Hybrid Cloud (most orgs will stay here)

Stage 5: Public Cloud

Julia’s note: I used to believe that Private Clouds were coming first, and maybe that’s true for large enterprises, but I think I’m about to flip my thinking around this. Certainly, for most orgs, SaaS came first and it’s up to enterprises to manage SaaS services and validate SLAs – and what they pay for that SLA. But when I think about how much harder it is to transform operations to run as a private cloud – all the politics involved just around chargeback or cost accounting – I can’t help but start to lean towards IaaS as the next set of cloud computing projects for enterprises, and perhaps more appropriately for business units and the midmarket. When it comes to cloud, IT has to think at a service level, and if you’re already thinking that way, how much easier is it to make the determination at an individual service level for what to move to the cloud and how. How much faster. How much easier to manage. How much smarter.

Well, what a difference a year makes – or even just the 6 months since the Gartner Data Center show in December. At the time, I’ve written that I felt some frustration that Gartner didn’t do more with service providers. After all, service providers (especially the large ones) are the models for the dynamic infrastructure, aka cloud operations (in particular private cloud), that they’ve been talking about for years. Check out our post on The Year of the Service Provider. As we’ve said, service providers have been running their operations like a private cloud for years. Yet other than one comment about where to find self-service portal tools (the answer – directly from service providers who built it themselves), service providers were clearly absent from the Gartner picture.

Not anymore. Service provider examples popped up in session after session that touched upon the cloud. Of course, it was mainly in reference to the very large cloud providers (Amazon, Google, etc) but small or large, the fundamentals of being a service provider hold true – from metered usage chargeback to the ability to report on multiple service levels tied to what each customer actually selects (vs. one service level – 99.9% in a traditional enterprise). I’m looking forward to more service provider examples and will share them here.

Drivers of Growth are changing:

• emerging cloud computing/storage sector
• increasing interest from mainstream enterprise and smb (vs previously when most early adopters were “internet enterprise”)
• now: banks, manufacturers, hotel chains, retailers

2010 Internet Infrastructure Revenue mix

• 35% IDC
• 35% Managed hosting
• 7% Dedicated hosting
• 16% Shared hosting
• 4% Internet content delivery
• 1% Interconnection 1%
• 2% Cloud services

Antonio sees the cloud as a major driver for datacenter revenue growth. Although it is currently only 2% of the overall Internet Infrastructure Revenue mix (multi-tenant outsourced infrastructure services), it is the fastest-growing revenue segment in this space (16% quarter over quarter). But more importantly, the “cloud” is a discussion point to get more enteprises into the datacenter, and once the discussion is started it’s down to figuring out what mix of outsourcing services makes the most sense for what the enterprise is trying to do, e.g., maximize opex spend efficiency, meet capacity or on-demand provisioning requirements, etc.

Now this is an interesting point, which we are actually seeing on the network management side as well;  IT operations is very interested in talking about cloud monitoring, but in the end the actual solution they want may not focus on this. For many, it’s early days – but if they don’t use the cloud monitoring aspects of EM7 today, they want the comfort of knowing they are working with a vendor partner that can provide those solutions when they’re ready. I suspect this translates to the datacenter space as well.

So what is the impact of the emerging cloud computing services? Tier1 estimates the cloud services market growth as $735 million YE2010 growing to $2 billion by YE 2012.

Hosters with cloud offerings will continue to be best positioned.

Additional managed hosting growth will be fueled by cloud (hybrid) services. The ‘toe dipping’ into the cloud trend will continue and virtual private clouds, in particular, are receiving a lot of interest.

So for you service providers/hosters who have survived the credit crunch, congratulations. You are in good shape with datacenter services and outsourcing on the rise. More CEOs are looking for these kinds of solutions and are actively part of the discussion. And the future looks decidedly cloudy.

Data.gov

The Senate has passed an IT oversight bill designed to help agencies manage the progress of their IT budgets. The bill requires more planning and oversight before agencies invest their money. Once projects are underway, agencies would also have to measure the progress against the estimated costs and schedule and meet performance targets.

“It’s clear that federal agencies are dropping the ball when it comes to deploying the right technology in a timely and cost-effective manner.” Sen. Tom Carper, co-sponsor of the bill.

More news on the government’s data center consolidation efforts. In reducing the 1,100 data centers in use today, their goals are to:

•  increase server virtualization four-fold or more
• double or triple the number of virtual machines per host
• increase server utilization several times over, and
• make significant gains in energy efficiency

Google and VMware have partnered and announced new features that make enterprise cloud services more portal, addressing the major concern of vendor lock-in. With Google’s and VMware’s Java development tools, enterprise programmers can create applications that run on Googl App Engine for Business in a WMware environment (hosted or on-premises) or even on other infrastructure that’s compatible with the standard Java stack like Amazon’s EC3.

Forrester Research outlines four key product categories that every IT manager should consider for controlling a virtual environment:

1. configuration management to understand and orchestrate relationships
2. capacity management and virtual machine placement to guard against running low or over-provisioning
3. performance monitoring for proactive alerting to a slowdown or total failure
4. real-time automation to keep up with demand of a sophisticated and dynamic virtual environment

After a 5.6% decline in 2009, Gartner predicts worldwide enterprise IT spending across all industry markets will rise 4.1% this year, surpassing $2.4 trillion for all of 2010. Gartner analysts believe while IT spending will return to growth in all major industries, ntional and international government will see the strongest gains.

Could IT staffers’ fears about being dispensable as cloud computing takes off actually be sabotaging the success of cloud computing? IT staffers raised concerns about job security at one firm that announced it was moving to a cloud using a software-as-a-service model. This article outlines some of the common front-line workers’ fears and also ways CIOs can help keep the unrest from spreading to the organization at large. Good news for IT in the cloud age: “We’re seeing IT look to reduce the tactical day-to-day support of applications and spend more time developing and delivering applications to the business.” Not so good news: “My view is that IT is becoming more of a service-oriented organization, providing more value-added services, with less emphasis on [maintaining in-house] systems, networks and architectures. You still need some of that, but not as much.”

Finally, the heading caught my attention: Verizon “forgives” $18,000 cell phone bill. Apparently the charges were accumulated over a six-week period in 2006. Verizon first reduced the bill by half and then eliminated the entire charge. The real news is that the FCC is examining ways to alert consumers when their wireless bills reach high levels to avoid “bill shock”. (Thank you!)

If you’ve read the posts I write from the Gartner shows, you might notice the increasing frustration I had with their lack of focus on and coverage around service providers. Understandable perhaps – since their customers are for the most part enterprises. But all that talk about Real-Time Infrastructure, Dynamic Infrastructure, Cloud Computing with Virtualization, Automation, Chargeback, Self-Service Portals, etc – it’s the service providers that are leading the charge on all of this. Distinguished Analyst Tom Bittman even mentioned in one of his sessions that his virtualization/cloud super-users told him that they look to service providers, not vendors, to get code for building their own enterprise self-service portals.

The reason why: most service providers have been running (or striving) to run their operations like a private cloud for years – far before the term “private cloud” ever darkened the Gartner doorstep. They really had to find ways of doing more with less – running lean and efficient IT operations because margins matter in their hyper price-competitive and severely resource-constrained world. They had tens, hundreds or even thousands of customers that needed to be supported in a multi-tenant environment based on shared resources. The consumption of these resources (bandwidth, memory, storage, etc) had to be measured, billed, monitored and displayed at an individual customer level.  And to support all these customers in the manner they expected to be served, service providers built self-service portals where compute resources were available on-demand, using virtualization and automation technologies, not because these are cool but because they had to meet the requirements of scale, speed and service their customers (and the market) were driving them towards.

Sound familiar? With automation and self-service portals in particular, we’ve now entered the realm of the private cloud.

This is the year that many enterprises and government agencies get serious about cloud computing and start running their IT operations like service providers.

We’ve watched with interest the ongoing dialogue and progress made in cloud computing. One by one, our service provider customers have made their own announcements about cloud offerings. There’s a reason why service providers (in their different forms) make up about 40% of our customer base. Many of us here at ScienceLogic, including all three founders, come from a service provider background. In fact, those experiences – running lean, efficient IT operations, being resource-constrained, first-hand frustration with too many point solutions or framework modules that cost too much (in terms of valuable engineering time and money) – are at the heart of why ScienceLogic and EM7 even exist.

The original product was designed with service providers in mind - multi-tenant from the start, bandwidth billing engine for chargeback, a services catalog of customizable SKUs, a holistic view of infrastructure performance and availability reaching all the way down to individual operator views of devices, integrated service desk that automatically tied actual operations issues to workflow and resolution. And last year, on top of the years of perfecting EM7 as the complete monitoring platform for service provider (type) operations, EM7 G3 added the scale, security, continuity of monitoring operations and more necessary for today’s extreme cloud requirements of the largest service providers, government agencies and enterprises building their own private clouds.

At FOSE we’re showing a sneak peek of the EM7 Integration Server, which does pretty much what the name says it does. It provides an elegant, secure and powerful way to pull data out of and push data into EM7 systems. Sounds simple but the possibilities are enormous. For anyone building a self-service portal for their private cloud, Integration Server supports automated monitoring provisioning; once VMs are provisioned, EM7 can automatically set monitoring policies according to what a customer has “purchased” via the self-service portal and, taking this one step further, work with a software  release/distribution tool to ensure that the right software is installed and then monitored to the specifications already laid out by your engineers.

I’m running out of room here – this post is already pretty long, so I’ll save the public cloud monitoring story for another one. But I’ll wrap it up by saying, that EM7 stays true to its original vision to Simplify IT Management by now simplifying data center and cloud management in a single, easy-to-use, easy-to-deploy solution. It provides a holistic view of computing assets wherever they are – in data centers, remote and mobile assets, or public, private and  hybrid clouds. With the demands of agile IT, it’s never been more important to see the full picture because end-to-end IT services delivery now spans multiple computing environments, and resources are increasingly shared, dynamic and expected to be available on-demand. You need the full picture for capacity, planning, troubleshooting and optimizing performance and availability. Multiple solutions that don’t talk to each other don’t cut it anymore, if they ever did. Best of breed solutions have their place, but without that view across everything, you really are only seeing a piece of the ever broadening picture.

Casey Coleman, CIO of GSA sums up her priorities in two words: IT modernization. Two words that cover many specific projects, as Coleman manages GSA’s proposed IT budget of $639 million for fiscal year 2011. She views the GSA as a service provider to the federal government and feels her agency should be a “stellar example” of the services available to other agencies. Coleman notes (and admits) that 83% of the IT budget across the agency is spent on steady state operations and maintenance. Since she doesn’t expect to get a lot of new dollars for innovative capabilities, she hopes to free up money from legacy systems to reinvest in new “stuff”.

IT cost-cutting in 2009 and the new economic reality of doing more with less has pushed technology projects like IT cost management, chargeback and IT financial management to the forefront for many CIOs for 2010. Surveys and roundtable discussions show a big shift in how IT organizations approach financial transparency by formalizing the process, using tools and presenting costs to business units. IT chargeback practices – knowing how much IT services truly cost and billing business units – will enable more responsible consumption and requests for services within a company.

There’s another report of the shortage of IT pros with skills for the “data center of today” – highly automated, dense and virtualized IT infrastructure that relies on the most advanced electrical and mechanical components for ultra power efficiency. The modern data center demands that IT professionals understand many disciplines as well as the facilities infrastructure, giving rise to hiring difficulties in finding applicants with the right mix of expertise for the rapidly changing data center environment.

The federal government’s use of Quest Communications for cloud services is expected to ease security concerns for enterprise. While there’s been tremendous hype, just a small percentage of companies are doing cloud computing in a production environment.

“The fact that Qwest has that federal customer means that it has been audited, it’s received its certification and accreditation from government as a prerequisite for the agency becoming a customer.”

Ted Ritter, analyst at Nemertes Research Group

Here’s some research showing that government agencies, faced with shrinking budgets and resource constraints, are turning to newer technologies to help trim costs. Some challenges include:

• Roughly 62% of federal, state and local agencies are challenged by reduced budgets and lack of resources
• 58% of federal agencies are challenged with keeping up with rapidly changing technology
• 52% of local governments cite challenges with enhancing the technical skill sets of their employees

“Tight budgets often spur creativity and can increase the willingness to try new approaches. Our research points to government interest in software-as-a-service solutions, cloud computing, virtualization and social networking initiatives as evidence of the need to meet the dual goals of containing costs and modernizing technology.”

Tim Herbert, vice president, research, CompTIA.

The primary characteristics of private cloud computing are server consolidation and virtualization. Research shows that most organizations have already consolidated some servers and about 40% have consolidated the majority of their servers into centralized data centers. As organizations increase their adoption of both private and public cloud computing, the WAN will be involved in an increasing percentage of instances when users access application and storage, resulting in additional security vulnerabilities. IT organizations should implement WAN optimization controllers (WOC) functionality in their data centers and remote offices.

For organizations using public clouds, both the IT organization and the public cloud providers must implement WOC functionality. Since WOCs are proprietary, the IT organization and the public cloud provider must implement WOCs from the same vendor, meaning WOC functionality should be part of the overall service being delivered to the customer. For both private and public cloud solutions, Application 2.0 WOCs will need to scale for more throughput and more sessions than Application 1.0 WOCs. Because of its well-known issues and performance problems, the Internet is not always the most appropriate WAN service to use to access cloud computing solutions. Research shows that IT organizations are slightly more likely to use the Internet over other WAN services when accessing public cloud services and they are more likely to use a WAN service other than the Internet when accessing private cloud services.