Recently, BlueLock’s Brandon Jeffress and I visited Anderson University in Anderson, Indiana to take a look at how cloud computing is being used, or could be used, in the classroom environment. Jeffress, alum of Anderson, arranged for us to meet with Professor Charles Koontz, head of the Information Technology department there. After a tour of the facilities, we sat down with Professor Koontz to discuss the role of virtualization in modern college IT training.


You can’t open a news feed today without reading something about cloud computing, virtualization, or infrastructure as a service (IaaS), so Brandon and I were surprised to learn that these topics are all but absent from the modern college IT major’s curriculum. Professor Koontz explained that colleges follow the guidelines of the Association for Computing Machinery (ACM) which sets standards for the curriculum of IT majors in order to ensure that graduates possess the appropriate body of knowledge upon entering the working world. Cloud computing, it seems, has simply exploded onto the business scene so quickly that education has not been able to keep up. Curriculum changes take time, and so it is left up to the proactive student to engage with these expanding fields through their own research and internships. 

Brandon and I weren’t satisfied with that. Even for the most proactive students, internships and independent exploration cannot compare to the engagement one gets with a project under the direction of a professor. Wouldn’t it be great if there was a way to introduce students to virtualization through a short classroom tutorial, allowing them to interact with the topics they were reading about in the news through actual hands-on experience? 

      
Professor Koontz suggested that the IT major’s senior capstone class might be a great place to start. He invited Brandon to come and teach a day’s class on Cloud Computing, the advantages of cloud hosting, and his role at BlueLock this fall. Brandon agreed that even an introductory level class might really help to give the students some clarity around the whirlwind of topics referred to as “cloud,” including: Platform as a Service (PaaS), Infrastructure as a Service(IaaS), and Software as a Service (SaaS). 

Additionally, Brandon suggested that the advantages of cloud computing could work in the students advantage when it comes to getting hands-on experience in managing environments. Before virtualization, it would have been impossible for an individual student to practice managing their own multiple-server environment. Even just three servers would have cost thousands of dollars in years past. But now, with virtualization, it takes just a few minutes to spin up three new VMs. If a college were to leverage virtualization in its classroom, students could manage their own multi-server environment in the cloud with ease. The student could control everything from creation of the VMs to their retirement, giving them great experience in one of the hottest fields in IT.

Professor Koontz believes such a program would be a great addition to a student’s education and would be willing to experiment with such a program at Anderson. He also recommended we investigate other local universities, including Indiana University, IUPUI, Ball State, Purdue University, and The University of Indianapolis. Together these schools could work to build a standard curriculum and pool their resources to implement this short tutorial series which might give Indiana’s recent graduates a leg up as they step out of the classroom and into a very “cloudy” business world.

Do you remember Google Wave? The new collaboration platform from Google that was supposed to be what e-mail could have been if only cloud computing had existed thirty-seven years ago? The real-time Internet architecture that fully exploited commodity servers spanning the globe and administered by unicorns? The Google Wave that developers gave a standing ovation to and was "dripping with ambition" at the Google IO Conference in 2009?

Yeah, Google officially abandoned the project today.

I sometimes drool over cloud technology and cloud computing infrastructure; consume APIs like candy and dig into whitepapers like they were the latest Kevin Grisham novel. The most important thing our development team does, however, is actually consume everything we construct. Our team develops things that we really, really want to use. Having the power to spin up huge servers ad nauseam is huge fun and we keep coming up with ideas to add even more tricks to the mix. Firewalling! Snapshotting! Catalog management! Instant deployments! Graphing resource utilization trends! It's fun.

New social networking technologies are birthed on a weekly basis and crazy amazing cloud computing companies spring up all year. The ultimate test is looking past the press releases and actually using the service. The architecture is only as good as the need it fills.

Over the weekend the OpenStack project proudly announced its existence and its intended goal: to create an infrastructure cloud platform that can reach the scale of a million machines. NASA has evidently dedicated a team of employees to support these efforts likely to replace their existing Eucalyptus cloud fabric controller used within their Nebula infrastructure cloud.

NASA had already been working on Nova, a next-generation cloud fabric controller, for the better part of this year. Nova had even been released as open source project for public adoption. Meanwhile Rackspace, simultaneously prepping their "Ozone" cloud infrastructure software for public release, approached NASA to see if the two could meld their codebases together. As a result OpenStack was born and now Nova seems to have gone defunct... even Nova's old home at http://novacc.org/ redirects to the Nebula cloud computing platform page.

The announcement of OpenStack has generated quite a bit of buzz. Several out in the grand Interwebs are wondering what this collective brain weight will bring. The goals are quite lofty: allow an open, inter-operable fabric for deployment and provisioning of infrastructure as a service. And while there are many cloud projects ready to pledge support, I wonder if consumer adoption is just as rampant. Will service providers spring up, ready to host an OpenStack cloud? Bear in mind while the hypervisor management may be open source and (presumably) free for use, the capital expense of a data center is most decidedly not.

My biggest wonder is how these two (or three) separate projects, up to now independently architected, will be able to merge and work together as a cohesive whole. Nebula and Ozone appear to be comprised of C, Python and C++ - each of which are definitely complimentary languages to each other - but the codebases may leverage very disparate frameworks. Will code have to be largely re-designed and re-written? Will the separate pieces just end up sandwiched together? Or are the software engineering efforts so vast that it doesn't even matter?

One thing is becoming very apparent - everyone and their mom is racing to push their cloud solution out into public light. Even Oracle just released their Cloud Resource Model API - although it seems that is barely making a din above the OpenStack conversation. Everyone established infrastructure and/or software company seems to be throwing their hat in the ring and handing out orchestration solutions. One big problem exists however: are they going to start handing out blade chassis, too?

Yeah, I don't think so, either.

I could be rolling in free hypervisors but it always comes down to one thing: who is managing the SAN? Or figuring out the resulting layer 2 network craziness? Or keeping the cores stoked? Or keeping the backup generator filled with diesel?

Gartner recently released six "rights" and one "responsibility" for cloud service users/clients to help enable better business relationships between vendor and client. This list, although short, is actually quite exhaustive in terms of outlining some major topics a client should cover BEFORE entering a cloud hosting agreement.

Gartner's list of Cloud Computing Rights & Responsibilities:

The right to retain ownership, use and control one’s own data - Service consumers should retain ownership of, and the rights to use, their own data.

The right to service-level agreements that address liabilities, remediation and business outcomes - All computing services - including cloud services - suffer slowdowns and failures. However, cloud services providers seldom commit to recovery times, specify the forms of remediation or spell out the procedures they will follow.

The right to notification and choice about changes that affect the service consumers’ business processes - Every service provider will need to take down its systems, interrupt its services or make other changes in order to increase capacity and otherwise ensure that its infrastructure will serve consumers adequately in the long term. Protecting the consumer’s business processes entails providing advanced notification of major upgrades or system changes, and granting the consumer some control over when it makes the switch.

The right to understand the technical limitations or requirements of the service up front - Most service providers do not fully explain their own systems, technical requirements and limitations so that after consumers have committed to a cloud service, they run the risk of not being able to adjust to major changes, at least not without a big investment.

The right to understand the legal requirements of jurisdictions in which the provider operates - If the cloud provider stores or transports the consumer’s data in or through a foreign country, the service consumer becomes subject to laws and regulations it may not know anything about.

The right to know what security processes the provider follows - With cloud computing, security breaches can happen at multiple levels of technology and use. Service consumers must understand the processes a provider uses, so that security at one level (such as the server) does not subvert security at another level (such as the network).

The responsibility to understand and adhere to software license requirements - Providers and consumers must come to an understanding about how the proper use of software licenses will be assured.

This list brings light to what BlueLock is already doing right to better our relationships with our own clients. Based on the Gartner list provided, we are certainly in the right place.

Typical data center architectures are based around not just the functions that servers perform, but the capabilities of the hardware in performing it.  In a cloud computing scenario, supported by full-scale virtualization, the capabilities of the hardware change from constants to variables.  Sometimes this makes it more difficult for architects to transition larger-scale deployments, even of specific functions like applications hosting, from physical data centers to the cloud. 

To some extent, Infrastructure as a Service (IaaS) cloud computing (specifically virtualization as the enabling technology for cloud computing) does homogenize the capabilities of the underlying hardware being used.  This is mostly a benefit because it provides economies of scale and allows IaaS providers to maintain higher availability for servers hosted in a cloud.  It does make things like sizing or designing the deployment of applications a bit tougher because typically we deploy the different aspects of a multi-tier application on different types of platforms – i.e., small, scale-out environments for web servers and large, scale-up environments for back-end database servers.

One approach that can be taken is to build “clouds within clouds” each with different characteristics.  A second approach would be to carve things like compute capacity or storage capacity up  into “building blocks” so that when it’s time to deploy an application, an administrator can combine one or more of these “building blocks” to ensure that a specific part of the application is getting the performance it requires. 

BlueLock takes both approaches.  Within our IaaS cloud hosting offering, we have different tiers with different performance and availability characteristics – BlueLock vCloud Express, Virtual Cloud Professional and Virtual Cloud Enterprise.  On the one end, BlueLock vCloud Express is great for things like dev and test.  On the other end, Virtual Cloud Enterprise is a fully-managed IaaS cloud built for performance and availability and perfect for mission-critical or regulated applications.  We try to work closely with prospects to understand their needs and then match those up with the appropriate service.

Cloud Computing Services Expert Chooses Advanced Heat Containment System from Airflow Management Authority

Worcester, MA June 28, 2010 -- Wright Line today announced that it has integrated its patented Heat Containment System (HCS) into BlueLock’s world-class, SAS 70 certified data center. BlueLock is an experienced provider of cloud hosting and managed IT services headquartered in Indianapolis.

“As a result of business growth and increased processing densities, excess heat was being produced in our data center,” said Mike Durham, BlueLock’s Director of Quality. “With Wright Line’s HCS, our ability to contain the hot air exhausted at the rack level, and then return it directly back into the CRAC, provides a predictable and efficient operating environment.”

Wright Line’s HCS was developed in direct response to customers growing concerns about the need to significantly reduce operating and capital costs while conserving energy and eliminating the waste most data centers currently experience.

The system captures, manages and directs the heat exhaust from IT equipment to the top rear of the rack enclosure were it is ducted to the data center’s precision air conditioning units through a ceiling plenum or hot air return.

The HCS can be seamlessly integrated into Wright Line’s own Paramount and Vantage Enclosure platforms, as well as most third-party server enclosures, including APC®, Rittal, Knurr and Chatsworth Products, Inc at the factory or in the field.

If you're a developer, an IT admin, or a cloud computing enthusiast, you've probably heard of Amazon's EC2 and you've probably heard of VMware's vCloud. These two platforms power a large portion of both public and private clouds. But, what's the difference?

In this Whiteboard Wednesday video, Jake Robinson and I outline some of the similarities and differences between Xen-based EC2 and Vmware-based vCloud. If you have questions or comments, please feel free to include them below.

Also, check out the BlueLock vCloud Express Cloud Monkeys contest and submit your use case to win a FlipCam and the Grand Prize - an iPad!

The BlueLock vCloud Express Cloud Monkeys Use Case contest begins today!

Former, current and new BlueLock vCloud Express Beta users will compete for these prizes:

• The first ten submissions will receive a stuffed cloud monkey
  
• The top five finalists will receive a FlipCam which they will use (and keep!) to create a recognition video for the application to compete for the Grand Prize - an Apple iPad!
  
• The Grand Prize winner will receive the engraved Apple iPad!

Open for submissions by current, former and new Beta users, the contest runs from June 16 – September 6 and looks to surface the most innovative use cases of BlueLock’s vCloud Express.

During the 12-week contest, BlueLock vCloud Express developers enter by filling out a simple questionnaire on the BlueLock Web site between June 16 and July 7 2010 that includes a description of the BlueLock vCloud Express use case and why it deserves to win. Participants can promote their own use case through Twitter and other social media outlets. Submissions will be voted on by an open community of voters and judged by BlueLock and VMware on cloud applicability, creativity/innovation, time savings and cost savings to select the top five use cases. The first ten submissions will receive a BlueLock “Cloud Monkey” stuffed animal and the five finalists will receive FlipCams with the option to document their use cases in a two minute “Recognition Video.” Finalists who submit Recognition Videos will then be judged by BlueLock and VMware for the Grand Prize, with the winner receiving an engraved Apple iPad.

“The functionality of BlueLock vCloud Express has proven to be unique and of value to our clients, driving us to design some of the same features into our other solutions within BlueLock CloudSuite,” said Kim Graham Lee, Chief Marketing Officer, BlueLock. “We are excited to not only learn more about how developers have been using vCloud Express, but to also highlight the most unique and interesting use cases.”

“As a top VMware vCloud service provider partner, BlueLock has been able to help shape vCloud Express as it continues to demonstrate that they are ahead of the curve in understanding their clients’ needs in the evolving cloud computing space,” said Mathew Lodge, Senior Director-Cloud Product Marketing, VMware. “We are looking forward to learning about how beta users have taken advantage of the dynamic combination of the industry-leading VMware platform and BlueLock’s secure and reliable cloud hosting and infrastructure expertise.”

BlueLock vCloud Express is a reliable, on-demand, pay-as-you-go infrastructure solution that ensures compatibility with internal VMware environments and with VMware Virtualized™ services worldwide. The technology allows users to create virtual machines as needed and add compute capacity via an online interface. Users pay only for the compute and storage space they use. Since being selected by VMware as one of only five companies worldwide to offer vCloud Express and launching in September 2009, BlueLock has reached 1,100 beta users of the product.

Participants can be past, current or new BlueLock vCloud Express beta users and can submit more than one application. For additional contest details, visit www.bluelock.com.

I remember quite vividly IBM's competition for SQL compliance on their AS/400 platform. 20-some years ago, databases had to be relational, tying together a vast sea of disparate columns. Relations between tables enforced a kind of consistency and normalization. No more brute-forcing random data into your corporate accounting system... now you had to obey the rules!

...or so the thinking went at the time.

Slowly, deep in the seedy database underground, seditious computer scientists sat stewing. They waited for the day when engineers realized that sometimes the process of normalizing data mutated it past the point of recognition. They knew one day some devious developer would see that relationships were too computationally expensive and slow. And one day... ah yes, one day... people would give up their crazy ad-hoc "Standard Query Languages."

While these computer scientists and software engineers were shoved to the margins by enterprise computing a few small companies took note of how well these rogue database systems scaled to the millions of users and petabytes of data. Lilliputian firms such as "Google," "LinkedIn" and "Facebook" started to lead a No-SQL revolution, running contrary to the dominant relational databases and instead storing mind-boggling amounts of data in non-relational tables and retrieving them faster than RDBMS' one-hundredth of their size.

Non-relational databases have become incredibly effective, especially when backed by a scalable pool of resources of a cloud computing provider such as BlueLock. If one takes a look at Redis - a powerful key-value store that can scale to a massive size - such a sense of scale quickly becomes apparent. By removing constraints one can get rid of building a huge number of indexes and instead deal out content quickly and efficiently. Craigslist has already leveraged Redis to an exceptional amount, and VMware sees quite a future in it as a platform as well.

If we take a step beyond we can see an entire landscape emerging: key-value stores such as Redis, Voldemort or Cassandra, hierarchical stores such as Zookeeper and tuple stores provided by JavaSpaces and Apache River. The number of choices seems to grow every day, and without a farm of servers it becomes quite a daunting task to evaluate which one fits your project best.

My recommendation is to take a step back and see which solution best fits the problem you are working within. Re-evaluate your needs and objectively ask yourself:

• What business or logic problem am I really trying to solve?
• How large is this data going to scale within a year? Are we talking about megabytes or petabytes?
• How fast does the data need to be retrieved?
• Do I really need to perform a bunch of ad-hoc queries? Or am I just looking up values based on their primary key?
• Which solution is easiest to deal with? Which makes the most sense to me?
• Do I need relational data? Do I need hierarchical data? Do I even care?

Once you build a matrix comparing each solution you will find some implementations quickly sink to the bottom and others become very tempting choices. Once you have determined a top list of possibilities, it is best to fire up a data store and write a few quick proof-of-concept test applications. A convenient way to do this is to login to your BlueLock vCloud Express account, spin up several virtual machines and load up an array of Linux boxes to test each solution out. Measure how easily the product can be installed and test how easily it can be scaled to multiple servers. Do some performance testing against sample applications on your own fenced network and watch your local resource utilization.

Very soon after you use your vCloud Express account to test the top candidates you should be able to feel one or two "fit" in a much more natural way than other solutions. For example, Zookeeper may be the natural fit for someone wanting to house a slew of centralized configuration data. At this point you can take the next step and test this alongside your web applications and judge more accurately the level of effort to get things running.

If at the end of this arduous process you still can't decide between a couple of top candidates do what I always do: pick the project with the best mascot. You simply can't go wrong.

Don't forget - once you select a data store implementation you can have your own scalable, elastic cloud to grow into. BlueLock can not only help you horizontally scale your data tier, BlueLock can also help design server layouts that best fit the sometimes eclectic world of non-relational databases. Whether it be heaps of disk or mountains of RAM to remain resident within, the BlueLock Enterprise Cloud can help your cabal of data power the next big thing.

BlueLock's healthcare-related client base seems to continue to grow - and they each have one specific need in common: security.

Recently, BlueLock produced a case study with Pathagility in regards to their Software-as-a-Service (SaaS) application which facilitates the generation, management and reporting of clinical data between healthcare providers and healthcare institutions.  Early on in the company's life, they realized that it made more business sense to forge a relationship with a provider who could secure and manage their IT infrastructure environment rather than spending capital on purchasing, implementing and managing their own.  At BlueLock, they were especially happy with the ability to scale and pay as they grew in a secure and SAS 70 compliant cloud hosting environment.  Pathagility's success was dependent on their ability to focus on the development of their SaaS solution and get to market quickly rather than directing their talent towards the day-to-day IT operations.

A very similar story unfolds with a new client, AmeriVeri CR. AmeriVeri CR is an in-depth , fully automated method for verifying the coding accuracy of healthcare claims. The company saves their clients an average of 4% in healthcare claims simply by identifying coding errors. They also needed a highly secure and SAS 70 compliant cloud hosting solution. The company just got its official start in January of 2010, but after quickly receiving funding and attracting great interest from clients and prospects, the comany needed to be able to expand its offering quickly.  Enter BlueLock.

These are just two of many healthcare-related clients BlueLock currently has running in its Enterprise Cloud platform, but each tells a similar story and proves that yes, the cloud is secure.

Looking to quickly get your application into the cloud? Want to build it on a cloud computing infrastructure powered by VMware?

vCloud Express allows you to quickly, easily, and inexpensively deploy your application to a VMware-based public cloud. In fact, right now it's free and you can get started on vCloud Express at the BlueLock website in a matter of minutes.

In this episode of Whiteboard Wednesday, Jake Robinson and I discuss the architecture of the vCloud Express platform and how you can spin up your own virtual servers -- in an instant! If you're a developer, researcher, software tester, or just a technology and cloud computing fanatic; this video is for you.

Be sure to comment below and ask any questions you might have, so we can try to answer them for you in next week's Whiteboard video. 

Cloud Computing means many things to many people.  It excites, motivates, and even scares some.  But what does it mean to you?

The advantages of cloud computing have been touted again and again - from flexibility, speed, versatility, convenience, and cost effective to green, secure and scalable.  But what makes cloud computing so interesting is that everyone has different thoughts and views around what it really does for them and means to them.

So first things first, what is cloud computing? At BlueLock we describe it in terms of 5 major things: on demand self-service, broad network access, resource pooling, rapid elasticity and measured service.  It is provided in three different service models: software-as-a-service (SaaS), platform-as-a-service (PaaS) and infrastructure-as-a-service (IaaS). (BlueLock does IaaS.)

But what's the real value of the cloud? These are the major values our clients have seen:

• Transitioning IT infrastructure costs from Capex to Opex
• Opportunity to lower overall costs
• Better match expenses to revenue
• Rapid provisioning (speed to market)
• Competitive advantage

 But again, I must ask, what does cloud computing mean to you?  

Yesterday, VMware and Salesforce held a joint unveiling of a new platform-as-a-service (PaaS) offering, dubbed "VMforce."

The term "cloud computing" is often overused and overhyped, but this offering appears to be a true cloud play. The VMforce platform is powered by Spring Cloud and enables Java developers to rapidly deploy their applications to the cloud.

To help you understand what VMforce is all about, and what its implications are on the cloud technology landscape, BlueLocker Jake Robinson and I put a quick video together (see below). We outline some of the key benefits as well as the potential downfalls of VMforce. Let us know what you think.

Will you use VMforce for your next Java project?

I'll admit it.  The more I see the term "aaS" used in reference to cloud computing models, the funnier it gets.  Besides being fodder for clever double entendre, it's also the core of some interesting discussion about where all of this is headed.  What do companies really need?  What are they comfortable with?  Where should cloud services organizations be making investments?

It occured to me recently that, in the technology world at least, the more things change, the more they stay the same.  Take, for instance, the current debate and comparisons between Infrastructure-as-a-Service (IaaS), Platform-as-a-Service (PaaS) and Application-as-a-Service (AaaS) is really a new spin on the standard computing stack: 

Applications --> AaaS
OS/Middleware --> PaaS
Hardware --> IaaS

Each of the three cloud computing models matches up fairly well with which each layer of the computing stack.  Applications can't run (at least today) without an OS and hardware resources without some software (an OS) to allow applications to access those resources isn't very useful. 

Cloud computing, similarly, will incorporate IaaS, PaaS and AaaS -focused solutions.  I'm not convinced that one particular model is the future of cloud services.  Krishnan Subramanian's post (hyperlinked in the previous sentence) is a good one and worth reading.  The problem with a world dominated by PaaS solutions is that not every development platform has a cloud platform to build for - leaving more popular development platforms like Ruby on Rails (Engine Yard, Heroku), .NET (Azure), etc. as the only game in town.  PaaS solutions today also require applications to be built with configuration management embedded in the code (using frameworks like Chef) which may be more effort than the developer is able to invest.

Managed IaaS or managed cloud hosting can fill an important gap between non-managed IaaS and PaaS solutions.  Managed IaaS solutions provide a reliable platform on which software developers can deploy their applications without modification on an application framework which they manage, but running on an operating system and hardware platform which they don't have to manage and which can scale as they need it to.
 

Some may argue cloud computing, specifically infrastructure-as-a-service, is moving towards a commodity (I think we're already seeing this) - but does that mean relationships don't matter? I say no. No matter how commoditized the Infrastructure-as-a-service or "cloud hosting" markets get, trust and relationships will continue to be a major factor in choosing and staying with a provider. 

Trust is king.  Clients want a brand they can trust and a provider who cares.  We've seen many prospects choose BlueLock because of the VMware brand, clients like Pathagility and Indigo BioSystems.  Indigo BioSystems, a technology provider for scientific research organizations, was at Amazon EC2, but just couldn't do what they needed - they needed VMware.  At BlueLock they got the security and management they needed on a VMware platform; two things they didn't find at Amazon.  Pathagility, a web-based platform for generating, managing and reporting on clinical data, needed security and compliance as well - on a platform built for mission-critical applications. They chose BlueLock because of the relationship (and VMware, of course).

The VMware brand is also the reason why vCloud Express has been such a hit.  People are out there looking for a self-service platform based on VMware, and now they have it.

Commodity or not, trust is and will continue to be king in the cloud.

In this series of posts (see Part 1 of the series), I'm looking at moving applications to the cloud and the scalability concerns around that. 

The interesting part is that these problems aren’t unique to cloud computing at all.  On one end of the spectrum, the promise of cloud computing and its expansive computing capacities has led customers to believe that simply moving their application to the cloud is going to solve all of these problems.  On the other end, clients who have very important applications running on-premise are concerned that when they move their applications to the cloud they’ll have to share all that wonderful computational goodness with hundreds or thousands of other clients and their applications’ performance will suffer.  Regardless of which perspective you may be coming from, there are two things to focus on when looking at moving to the cloud.

The first is raw computing capacity.  At BlueLock, we’ve chosen to build our cloud computing platform on VMware virtualization technologies.  One of the benefits of virtualizing applications on VMware is that multiple workloads (running within virtual machines) can be configured to run on very high-end server hardware and storage architectures – perhaps mutli-socket, multi-core server hardware with 32GB or 64GB of RAM and high-performance SAN(s).  Those physical hosts can then be combined into clusters and that computing capacity can be even further aggregated.   It’s important to understand how that computing capacity is assigned to your application(s).

Is infrastructure being “over provisioned”?  Since it’s possible to abstract the underlying hardware from the workload running within a VM it’s also very easy to do things like allocate more memory or compute power to the VM than is actually available on the underlying physical hardware. 

Can computing power be scaled (up and down) if needed?  As the business grows, the demand on application performance may grow with it?  It should be easy to assign and re-assign things like CPU and RAM resources.

How high can the underlying hardware platform scale?  Different IaaS and cloud computing models are based on different technologies – VPS (Virtual Private Servers), dedicated physical hardware and virtualization platforms like VMware all work differently, for example.  How much CPU and RAM in total (usually different based on the underlying model being used) can be assigned to the application(s) has an impact on the decisions you make about scaling.

Within the BlueLock IaaS Cloud, compute clusters are carefully divided into building blocks called “cores” and these cores are assigned to customers – never assigning more “cores” to a computer cluster than are actually available.  This goes hand-in-hand with dedicated versus shared computing models – just throwing everyone in the computer pool without regard to expected performance isn’t a good idea.  It’s important to ensure that the capacity to application(s) is both dedicated and somewhat dynamic.  At BlueLock, once one or more of these “cores” is assigned to a client they are combined together into a resource pool.  This pool of CPU and RAM can then be divided among one or more virtual machines, assigning priority to different workloads if necessary and providing the ability (if needed) change how much of the resource pool each VM is allowed to consume.  Behind the scenes, cool features of VMware’s virtualization platform like VMware DRS move VMs around from one physical host in the cluster to another without taking VMs offline.  This ensures that a particular physical host is never over provisioned and that, if needed, the amount of CPU and RAM assigned to a particular VM is always available to it.

This model of cores and dedicated resource pools, along with the abstraction of physical hardware from the resources assigned to a virtual machine, allows clients to provision (and pay for) only what they need.  As their needs change, additional cores can be added to grow resource pools and add to their application’s overall computing capacity.

In the next post, I'll look at the second item to focus on - application architecture.

In this series of posts (see Part I and Part II), I'm looking at moving applications to the cloud and the scalability concerns around that.  The interesting thing to notice is that these problems aren’t unique to cloud computing at all.  

On one end of the spectrum, the promise of cloud computing and its expansive computing capacities has led customers to believe that simply moving their application to the cloud is going to solve all of these problems.  On the other end, clients who have very important applications running on-premise are concerned that when they move their applications to the cloud they’ll have to share all that wonderful computational goodness with hundreds or thousands of other clients and their applications’ performance will suffer.  Regardless of which perspective you may be coming from, there are two things to focus on when looking at moving to the cloud.  

In the last post, I looked at the first issue - raw computing capacity.  A second thing to consider is the application architecture itself.  As Mr. Golden highlighted in the CIO.com article, one could even argue that this is more important than the scalability of the underlying cloud computing platform.  The reason is that there is always a limit to the hardware (virtual or physical) that can be thrown at an application and a lot of applications aren’t even designed from the beginning to scale in this manner.  With VMware a maximum of either 4 or 8 vCPUs (virtual CPUs) can be assigned to a VM depending on the version of ESX being used.  There are even good reasons why arbitrarily assigning the max number of vCPUs to a VM isn’t the best course of action.  

More importantly, if the application (and underlying OS) wasn’t built to support SMP and multi-threading, adding vCPUs will have no effect whatsoever.  If scalability is a concern, ensure that all of the applications components can take advantage of a large number of CPUs and can address > 4GB of RAM.  This is known as a “scale-up” model.

In highly scalable application deployments, though, a “scale-out” model is usually more appropriate.  Applications that are designed to spread load across two or more hosts allow you to add compute capacity simply by adding additional servers.  On the back-end, adding additional database servers and using horizontal scaling relational database tricks like database shards allow you to remedy DB bottlenecks without implementing huge SMP systems to accommodate query load.  The added benefit of this “scale-out” approach is that you get higher availability of the application for free.  You can take one application server offline for maintenance while not affecting the other servers in the application farm.  In addition, if one application instance experiences a crash other users on other instances of the application continue to function normally.

In the on-premise, physical server world both the “scale-up” and “scale-out” approaches I’ve discussed was usually very costly.  4-way or 8-way servers with gobs and gobs of RAM are an expensive way to grow vertically and add performance and though smaller systems used as part of a horizontal scaling approach are less expensive initially, they add a whole new level of complexity and expense in terms of ongoing maintenance.   What’s more, during non-peak times all this compute infrastructure sat largely un(der)-utilized. 

What is exciting about cloud computing IaaS – especially the 100% virtualized IaaS which BlueLock has built – is that this computing model is perfect for these kinds of scalability needs.  For scale-up applications, clients can start small and grow into things.  For scale-out applications, new VMs (running additional instances of the application) can be added and managed much more easily than they can in the on-premise world.

For more info on BlueLock or scaling your application in the cloud, contact us.

“Can cloud computing scale for my [insert client’s business-critical, highly visible application here]?” 

This question comes up a lot from customers as they evaluate moving their on-premise applications to cloud-based IaaS platforms.  This week on CIO.com, Bernard Golden wrote an article on cloud scalability in which he made some very salient points and also sighted typical problems that customers have when moving applications to the cloud.  In summary:

Availability: Applications running within the cloud run fine, but if there is a crash or the application needs to be terminated, it’s out of commission until a new instance can be brought online.

App-Tier Scalability: If an instance of the application gets overloaded, there’s no way to add more resources to improve application performance.

Data-Tier Scalability: Sometimes the back-end database is the bottleneck for performance.

Maintenance: There’s no easy way to update the application without downtime.

The interesting part is that these problems aren’t unique to cloud computing at all.  On one end of the spectrum, the promise of cloud computing and its expansive computing capacities has led customers to believe that simply moving their application to the cloud is going to solve all of these problems.  On the other end, clients who have very important applications running on-premise are concerned that when they move their applications to the cloud they’ll have to share all that wonderful computational goodness with hundreds or thousands of other clients and their applications’ performance will suffer.  Regardless of which perspective you may be coming from, there are two things to focus on when looking at moving to the cloud.

In the next post, I'll look at the first issue - raw computing capacity.

Tom Henderson and Brendan Allen with Extreme Labs did an excellent and even-handed job of comparing three Enterprise Cloud Services.   BlueLock was honored to be considered.  The costs came out exactly as I would have expected and for the reason I’ve pointed out many times in previous blog posts, including this one (see tip #13 from a recent post) – we were the “expensive” option.  I’m not saying that being the expensive option is a bad thing, but if you look at the bigger picture, we just aren’t. The difference is people and expertise.  
 

Our philosophy (and pricing model) is different than Terremark and Rackspace’s cloud hosting offerings.  At BlueLock we’re focused on the relationship and helping companies manage the infrastructure so that they can hire more developers and/or sales people rather than systems engineers.  You’ll pay “more” at BlueLock because there’s more value (hint: people) in our offering than just raw compute and storage.  If you don’t want to read further, just scroll down to the illustration and you’ll see just what I’m talking about.
 

In addition to delivering raw unmanaged infrastructure, BlueLock is delivering systems engineering expertise to manage the environment from the operating system down – which is all part of the bundled price.   I’ve presented many times to an audience of software company executives on making the best infrastructure choices for their SaaS offering and the cost numbers I use during that presentation are almost identical to Tom’s and Brendan’s.   
 

My point here and during the presentation is that when you load your applications to the cloud – someone (a person or people) MUST manage those environments – whether they are internal or external.  When you bring them to BlueLock and load them into our Enterprise Cloud Platform, you do not need to employ infrastructure experts – that’s what we do for you.  Companies that choose our competitors or Amazon for that matter must have infrastructure experts to manage the environment – and that cost is not in their numbers, so they look much cheaper.   
 

The chart below adds a single full-time resource to the costs to represent one employee managing the environment.  That doesn’t cover you if your employee would like to sleep, get sick or take a vacation, but you’ll get the picture without going off the deep end with costs.  
 

Finally, I recently competed against and beat one of these two cloud hosting companies mentioned in the Network World article (link) and based upon my discussions with the client the other guys did have an implementation fee….I’m just sayin’.