How do you keep it all straight?

At Microsoft, they utilize their Bing application, Bing Maps, to sort it all out. One of the more interesting demos that I saw this year at the Microsoft Partner Conference was the “context-based search experience” from Bing.  If you have ever searched for something and found plenty of nonsensical results, you appreciate the work that Microsoft is doing to bring some order to searches.  Bing Maps does a great job of understanding all of the complexities of searching and tries to tie in not only what you are looking for, but the context around it.  And that takes a lot of horsepower.

To achieve their objectives for storing and correlating petabytes of information, Microsoft needed a massive number of cores, along with high power efficiency.  In their data center in Colorado, they deployed tens of thousands of AMD Opteron™ processor cores inside Dell Nucleon servers.  While the Nucleon may not seem like a household name, there is a reason for that.  These servers are customized through the Dell Data Center Solutions group, a team of experts who focus on large, scale out data centers and have a particular expertise around very large, dense data center deployments.

These servers delivered the density and most importantly, the power efficiency that Microsoft needed to make their Bing Maps come to life.

One of the most interesting aspects of this deployment is not what was being deployed, but how the servers were deployed.  Dell delivered complete containerized solutions to Microsoft that allowed them to quickly deploy and manage the server farm. The Dell Modular Data Center actually utilizes fans and cooling at the container level versus managing the cooling at a server level.  In the world of homogeneous workloads where thousands of servers are all running the same application, this makes tremendous sense because you scale the workload across nodes, minimizing power and heat fluctuations and allowing you to manage the power and cooling aspects at the container level for much greater efficiency.  Microsoft believes that by doing this, Dell helped them reduce power requirements by as much as 80% while simultaneously delivering up to 5X the compute density.

One of the key components in this solution was the low-power AMD Opteron 4100 Series processor, designed specifically for these environments where density and power efficiency are key.  But with 4 and 6 cores, it is clear that while customers are getting exceptional power efficiency and density, they are not doing so by sacrificing their core density – something that a large scale out data center really demands.

As the world continues to evolve towards the cloud, and these large scale out data centers continue to become more plentiful, we’ll hopefully see more customers opting to deploy the AMD Opteron 4100 Series processors in solutions like the Dell Nucleon server from the Dell Data Center Solutions group.  And that ultimately means clouds with high performance, density and power efficiency.

To me, that sounds like a perfect map to the cloud.  Bing Maps has is all laid out for you.

For more information, you can download the complete case study at: http://content.dell.com/us/en/corp/d/corporate~case-studies~en/Documents~2011-bing-10009998.pdf.aspx

John Fruehe is the Director of Product Marketing for Server, Embedded and FireStream products at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

We live in an age where virtualized environments have become indispensable to supporting an ever growing number of connected, internet-enabled devices, social networks and enterprise cloud computing environments. As a result, the need for legacy chip support will grow exponentially as IT departments play the catch-up game to expand their data centers to support this proliferation of technology. It is unrealistic to expect businesses to always “trade up” to the newest technologies in order to gain access to the latest features, regardless of the size of your company or wallet.

At AMD, we believe that virtualization is a mainstream technology. It’s not just meant for the elite. We foresaw that hardware virtualization would be indispensable to our customers large and small and designed processing platforms with the goal of democratizing performance and scalability. Instead of forcing customers to always cater to infrastructure vendors, AMD instead brings virtualization to the masses for the long-term, so information can flow more easily no matter what generation of hardware you have in your data center or PC. It’s also worth reiterating that AMD Opteron™ 6100 Series processors stand out from the pack because they are equipped with dedicated pipelines for up to 12 integer threads, resulting in dedicated power and performance for virtualized environments and full speed memory on all models.

Every dollar invested into IT hardware and software has a great impact on our customers’ bottom line, therefore, we do not want to force our customers into purchasing the latest shiny thing. We also understand that a one-size architecture does not fit all situations, so we want to work with our customers to see that their time and dollars are spent wisely on the tailored solutions. That way, they can gain the most utilization and performance out of their AMD-based systems to ensure improved operational efficiencies.

Vlad Rozanovich is a director of Commercial Business Development at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

Designing world-class semiconductors, platforms and cores for our customers and OEM partners around the globe is one of AMD’s most complex and demanding engineering tasks.  From an IT perspective, it means we have to meet the massive demand of computing power and the need for a flexible infrastructure while finding a way to deliver these resources to our engineers across multiple geographies on an as-needed basis. It’s been a daunting challenge but one that we have embraced.

Our motto was “Compute Anywhere” – inspiring our IT team to create the high-performance private AMD Cloud, an enterprise-wide grid that now powers all R&D at our company. This multi-year initiative received considerable attention and praise when AMD was recently recognized as one of CIO Magazine’s 100 most innovative organizations.

At AMD, we have spent a lot of time thinking about the cloud.  Our recent study on “Adoption, Attitudes and Approaches to Cloud Computing,” revealed that 60% of the organizations that use cloud solutions already see business value.  Moreover, the success of our private cloud demonstrates we are walking our own talk and embracing cloud computing’s greatest benefits – flexibility, increased efficiency, and hardware cost savings.  If your business is thinking about the cloud, you may want to consider what we have been able to accomplish.

More and more companies need to be able to function fully anywhere, anytime. This is particularly essential when your business relies on technology to deliver technology. The AMD Cloud runs on more than 115,000 AMD CPU cores and over 4 PB of storage around the world. It supports thousands of engineers worldwide with approximately 40 million jobs executed in a month.  It is an integral part of how AMD is able to deliver true R&D innovation to our customers. As IDC’s Senior Vice President & Chief Analyst, Frank Gens states: “this year, the market will recognize “the cloud model as a fundamental part of IT service delivery.”

We have stretched our IT muscles through the cloud and are reducing our spending on hardware and data center costs while responding dynamically to the project needs of our engineering teams. By consolidating the compute infrastructure into a logical cloud of a few large geographically diverse physical cloud clusters with common standards, we have achieved a sustained utilization rate of greater than 90%.

Our flexibility has also dramatically expanded what we are able to do. Consider that projects are no longer bound by geography due to the location of the data that they require.  And, over half of our storage was once consumed by huge data sets, and had to be replicated between sites.  This used to take weeks to carry out. Today, we are leveraging available engineers, regardless of their location, to engage on any project using consolidated data sets.

By way of example, in 2010, we needed to shift massive amounts of compute power to a single project. The IT team dynamically reallocated capacity in one night, giving the project approximately 45% of the cloud – up from their typical consumption of only 4%.  Consequently, the project team was able to get two months of testing done in only five days – an outcome simply not possible without the cloud.

So next Monday morning, while you’re heading to your office, think about what task you and your team can achieve before Friday ends.  Five well spent days can result in a dramatic impact on time to market for you and save millions of dollars.

Additionally, you may be asking other questions, such as, what does flexibility mean to your business? How about improved customer service, 24/7? What about being able to sell and deliver products online to a family in Beijing?  How should you assess the value of investing in the cloud model?

Our strategic approach to the AMD Cloud has included a strong focus on standardization across platforms resulting in reduced investment costs for equipment upgrades, and basing the cloud entirely on AMD technology.  Specifically, our servers based on AMD Opteron™ chips  offer a unique capability to swap out processors and achieve in-place upgrades, allowing us  to double our capacity without buying a single new server. Over $6 million in savings were realized with just an in-place upgrade from dual to quad-core chips, demonstrating a direct ROI that comes with the right IT investment.

The AMD Cloud demonstrates daily how IT innovation can be used to create more innovation. We are now better able to respond to dynamic needs, enable resources across the globe, and maintain high levels of service delivery. Our virtualization-ready technology –cloud clusters of servers equipped with AMD Opteron™ processors, running AMD-Virtualization™ technology – has helped us drive and materialize new ideas that will turn into enhanced solutions for our clients.

Our team at AMD is now able to fuel our extremely compute-heavy R&D activities – and to do so with the knowledge that we have the resources we need to embark on any new research opportunity. We, therefore look at our future with even greater optimism, as we contemplate what we will be able to do with this new greatly expanded capability.

The sky is the limit, right there among the clouds.

Mike Wolfe is Senior Vice President and CIO at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

Thanks in part to AMD’s technological leadership in low-power processors and optimization of the server platform roadmap for density, the issues around power in the data center power are fairly well understood, and we have helped OEMs develop some smart solutions on the market today with more coming very soon.

However, it is not enough to think about this year or even the next 5 years.  What happens in the timeframe that is “off the roadmap” of today’s technology?  This is where AMD’s research team has posed some key questions and NYSERDA has stepped up along with Clarkson University in upstate NY, HP, and other industry partners to help answer the questions.

We know that renewable energy – solar and wind power – plays a major role in our future.  How do we link this vital resource to the data center and I mean directly link power source to servers?  (You know AMD is all about eliminating the bottlenecks!)  That is one key issue – getting power from a wind turbine directly to a data center like an HP POD without building a traditional electrical grid between the two. 

Renewable sources can also be intermittent.  What do you do if the sun does not shine one day or there is an atypical calm in the wind one evening?  A data center’s reliability cannot tolerate lulls in the action, so a big question from our AMD Research labs that takes this all one step further is “How can we shift a compute load automatically and reliably between renewable energy sources without resorting to a traditional electrical grid?”

This is a multi-faceted problem and we expect the solutions will fundamentally alter how we design and build computing resources in the future.  The economic implications are not insignificant.  For example, the cost of laying optical fibers is orders of magnitude less than the cost of building power lines.  Estimates can range from $500/mile for “dark” fibers to $15,000/mile for new optical fibers compared to $750k-2 million/mile for new electric transmission lines. The potential gains for the IT industry and the energy sector are exciting and, frankly, huge.  You can bet you will be hearing more about this in the months and years to come.

Alan Lee is the corporate vice president of Research and Advanced Development at AMD.  His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

Many people in the cloud and dense computing world have data center racks that are packed that tight.  Every watt counts because at load, they are maximizing everything in their power budget per rack. 

For customers like this we have a new feature in our upcoming processors codenamed “Interlagos” and “Valencia”: TDP Power Cap. [eWeek takes a look at TDP Power Cap and potential impact for cloud/mega datacenter customers here.]

In today’s products you can cap the power of a server by turning off processor states, though that ultimately can lead to somewhat lower performance. There are a lot of our customers who value power efficiency over performance, so it is not a real issue for them.  But what if you could set low power limits AND get the performance that the platform offers?

With today’s AMD Power Cap Manager, you can limit the processor P-states and cut power consumption although this limits the processor’s ability to get to the top frequency (which is also the most power-hungry spot on the curve as you can imagine…).  By essentially “locking out” the top P-state, the processor never gets into that state, even under heavy utilization, helping cut down total power to the processor.

With the new TDP Power Cap for AMD Opteron™ processors based on the upcoming “Bulldozer” core, customers will be able to set TDP power limits in 1 watt increments.  This means that instead of having to choose between different TDPs for processors, you can actually buy any power range and then modulate down.

Why would someone want to limit TDP?

Well, let’s say that you have a maximum power draw on your fully configured server of 300W, and you have 42 slots in your rack.  The simple math says that you have 12.6Kw of power load that you need to be able to support.  Now, if your power budget only allows you to bring 12Kw to the rack, you essentially have 2 slots that need to be left open in the rack because you can only support 40 and not 42 servers.  But, by utilizing a custom TDP, you could drop the max power that some servers could draw, bringing you in under the limit of 12Kw and still getting 42 servers in the rack.

Best of all, if your workload does not exceed the new modulated power limit, you can still get top speed because you aren’t locking out the top P-state just to reach a power level.

This is truly essential for blades, clouds and other dense environments where every watt counts.

John Fruehe is the Director of Product Marketing for Server, Embedded and FireStream products at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

For many enterprises, data center concerns for the past 5 years focused on IT sprawl and inefficiency, power and cooling shortfalls, and inadequate disaster recovery plans. IT organizations are addressing these challenges with a range of solutions:

• New technologies such as server/storage virtualization to consolidate IT assets
• New IT deployment models such as converged IT infrastructure to improve utilization and maximize asset use
• New data center designs to optimize power and cooling while enabling more rapid IT recovery

Customers are increasingly deploying modular computing infrastructures optimized at the chasis and rack level for optimum processing, memory, I/O and storage performance. Capacity will be delivered in partial and full rack increments. Transaction and content delivery environments, used in traditional web serving and on-line transaction processing, are sensitive to latency, response time, and availability variations, so some geographic dispersion in multiple data centers makes sense.

As an organization makes the transition from Big Data as a “junior science project” to Big Data as a core business resource, concerns about the impact of this new, and rather different, workload on current and future data center plans will increase. Today, the IT architectural approach used in clustered environments such as a large Hadoop grid is radically different from the converged and virtualized IT environments driving most organizations’ data center transformation strategies. They have different server/storage configurations, different environmental (power and HVAC) profiles, and different data ingest/migration patterns.

In a number of larger enterprises (especially those with global needs like major financial institutions and retail organizations), IDC expects to see the emergence of separate data centers designed specifically for Big Data workloads. Concentration of data streams and compute resources makes more sense for both performance and telecommunications cost reasons.  Basically, the data center becomes the Big Data system.

This concentration requirement will also make “the Cloud” a critical part of the “Big Data” picture. Conceptually, “The Cloud” is about dispersion of computing and storage resources, but in reality, the organizations building the underlying cloud infrastructure are concentrating IT and telecommunications resources to more efficiently deliver applications and information. Those organizations are also playing a role in every facet of the Big Data space:

• They are one of the most important generators/collectors of data inputs
• They are among the most important aggregators/forwarders of data streams and content
• They are the most aggressive users/innovators in the development of Big Data hardware and software solutions for use in their own businesses

Given this combination, IDC expects cloud service providers to play a critical role (through simple, temporary provisioning of large processor, memory, and data pools) in the early and rapid adoption of Big Data use by technically savvy, but resource constrained, organizations. Already, cloud-based Big Data platforms are making it practical for smaller engineering and architectural firms to access massive compute resources for short, semi-predictable time periods without having to build their own Big Data IT farms.  Look for solutions targeting other sectors such as healthcare, consumer products, and logistics.

As an IT leader, you need to both understand what your organization is planning around Big Data and begin developing a Big Data IT infrastructure strategy (be it part of your own data center or with a cloud partner). In doing so, you must understand the potential value Big Data represents for your organization and industry. IDC believes that building successful business cases around Big Data can only be accomplished through a tight alignment of critical thinking across both IT and the business. This will require out-of-the-box business thinking as well as moving outside traditional IT comfort zones like traditional data warehousing.

Big Data deployments will also require you to rethink and realign IT administration and application developer skill sets. In this fast evolving area people with these skills are likely to be in short supply for quite a while. You may be able to retrain some existing team members, but once you do they will be highly sought after by competitors. Cloud service providers, making their own considerable investments in Big Data solutions will be among your best “friends” when it comes to getting started, getting serious, and getting to be a leader in Big Data.

Richard Villars is Vice President of Storage and IT Executive Strategies with IDC. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

Not surprisingly, the government and academic IT leaders we spoke with confirmed that they really get the benefits of cloud services.  They know about the strong potential for cost savings and operational efficiencies, as well as accelerated upgrades, access to innovative applications and the many other advantages cloud solutions can provide.  In fact, nearly 25 percent of the officials reported that they have already moved some of their operations to the cloud, and the majority said they intend to integrate cloud-based solutions into their operations over time. And 68 percent of the public sector respondents view cloud solutions as a tactical move or a necessity for their operations, compared to only 52% in the private sector.  In today’s tight economic times, with decreasing tax revenues and increasing demand for more efficient and effective public services, public sector IT managers clearly understand that cloud solutions need to be part of their overall IT strategy

We were surprised to learn, however, that  75% of the respondents who said they are evaluating cloud computing strategies said do not feel they have the IT skills in place to support a cloud environment. The survey proved that public sector organizations are interested in cloud services, are likely to move some of their services to the cloud (and indeed some said they already had moved to the cloud), but they are worried that their staff doesn’t have the skills that are needed to implement cloud solutions. They are worried they have a knowledge gap. And they need to overcome this gap to be able to pursue a cloud strategy for their IT operations – be it a private, public, or hybrid approach.

To be fair, different cloud models require different skills for “in-house” staff.  Managing private or hybrid cloud services is much more complex and labor intensive than managing outsourced public cloud services, although that too requires specific in-house expertise to manage it effectively.  The fact that public sector IT managers think that they have a knowledge gap within their own operations is important because it demonstrates they are thinking about what they need to put in place so they can evolve their operations. Of course, it also means the majority don’t think they have what they need to get the job done.  That’s important feedback for the IT industry, given that public sector represents a significant portion of the global market for cloud solutions.  Sure there is a lot of information on the cloud out there, but not all of it is accurate or helpful.   What IT managers and their staff need is help navigating so that they get the specific information, training, and tools they need to evaluate, select, and implement the right cloud solutions to meet their needs.

In addition to the knowledge gap issue, the survey respondents told us something we already knew, but were really very happy to have confirmed independently.  A full 86% of public sector cloud customers stated that the IT infrastructure used in a cloud service was an important part of their decision to implement cloud computing.  In other words, they believe that “architecture matters” and IT hardware and software is an important part of their evaluation of cloud services.

And why shouldn’t it be?  IT architecture, beginning with the foundation of cloud computing – the processor – helps  define the level of performance, reliability, and energy efficiency of a solution, as well as enabling crucial optimizations in the areas of virtualization, security, and manageability.   These features are vital to delivering the benefits of cloud solutions, and without careful attention to the IT infrastructure a cloud solution can end up delivering a host of problems, including downtime, security breaches, cost increases and other nightmares.  Further, IT managers need to be careful to select open standards approaches to cloud solutions rather than proprietary technologies that promote “Vendor Lock-In,” which can increase costs and decrease interoperability while limiting choice and future system changes.  Indeed, the public sector IT managers we spoke with understand that they need to do more than take a test drive of the cloud applications before they commit to a purchase, they need to look under the hood.

Another thing we found interesting was that when it comes to cloud adoption, the gap between the private sector and public sector is not all that great (37 percent to 21 percent, globally, who are using cloud solutions today).  And with mounting pressure in many countries to accelerate the public sector migration to the cloud to reap the cost savings and other benefits, we expect the difference to continue to close.   This response provides a counter to the longstanding stereotype of the plodding public sector bureaucracy, slow to adopt change and innovation.  At least in terms of cloud computing, it turns out the public sector is closing the gap with the private sector.

Interested in learning more about AMD’s global cloud survey, go to: www.amd.com/cloud

To learn more about how the AMD Difference is helping to transform Public Sector IT services, go to www.amd.com/public

And, lastly, for a visual representation of the findings, check out the below:

Steve Kester is a Director of US Government Affairs at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

Just as London is upgrading its infrastructure, many organizations are looking to Cloud Computing as an approach that allows IT to match new technology with older infrastructure. The findings of AMD’s global cloud computing study clearly show that the cloud is rapidly maturing, gaining widespread adoption with customers willing to store significant amounts of data in the cloud. For example:

• 70 percent of organizations worldwide are either using or evaluating cloud computing today
• 63 percent of organizations using cloud estimate they store more than $250,000 worth of data in the cloud

However, there are still some areas where we need to “mind the gap.” The applications primarily hosted in the cloud (email, finance/accounting, web serving) all have different computing demands that require a unique mix of performance, power and cost efficiency in the data center. With this in mind, it can be perceived that the IT industry is currently approaching a slippery slope in the way we think and talk about cloud computing. There is a risk in framing certain cloud technology as universally applicable to all clouds, but the reality is that cloud IT is just as dynamic and ever-changing as traditional IT. AMD takes a workload-based approach and coordinates closely with OEM partners, software vendors, and customers to deliver flexible solutions based on their IT needs in the cloud.

At AMD we are closing the gap in cloud computing by addressing both sides of the cloud – with power efficient AMD Opteron^TM processor technology that delivers more cores to feed steady streams of transactions to growing numbers of mobile client devices.  On the client front we are marrying CPU and GPU cores via our AMD Fusion APU technology to provide the underpinnings for an emerging world of mobile and thin clients that can deliver a superior visual experience with minimal power consumption.  And if you look forward over the second half of 2011 you find us delivering redesigned core architecture via our “Bulldozer” technology, the next generation of APU technology, and graphic processors that can be integrated into cloud clusters.

With that in mind, are we experiencing the rise of cloud computing? Absolutely. But are all clouds creating equally? Absolutely not. It is imperative to approach cloud computing not as a solution that is an “easy fix” to on-premise servers and software, but rather a new model with its own challenges and opportunities that we must address with better education and tailored solutions for customers navigating their foray into the cloud.

What do you think? For a visual representation of the findings, check out the below infographic:

Margaret Lewis (@margaretjlewis) is a Product Marketing Director at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

As my colleague, Matt Eastwood, mentioned in an earlier blog , data creation is occurring at a record rate. After noting this development, business and IT executives tend to ask IDC one question, “So what’s driving all this data growth?”

In the financial services industry, the answer is easy to provide. The volume of diverse financial transactions occurring around the globe continues to accelerate. Players in this industry face ever stricter anit-money laundering regulations that require big investments in “Big Data” solutions. For most other industries, however, the answer to this question of data sources provides some interesting insights into what use cases are driving Big Data developments. Industries that just recently began to digitize their content are or soon will be joining financial services as some of the biggest Big Data consumers.

• Media/entertainment: Moved to digital recording, production and delivery and are now collecting large amounts of rich content as well as user viewing and gamer playing behavior data
• Healthcare: Quickly moving to electronic medical records and images which they want to use for short term public health monitoring and long term epidemiological research programs
• Life sciences: Will soon be generating large volume of low cost (<$1,000) gene sequencing data that needs to be analyzed to look for genetic variations and potential treatment effectiveness
• Video surveillance: Transitioning from CCTV to IP TV cameras and recording systems that organizations want to automatically analyze for behavioral patterns. Generating data at an accelerating rate from fleet GPS transceivers, RFID tag readers, “smart meters”, and cell phones (call data records, or CDRs) that they want to use to optimize operations
• Transportation, logistics, retail, utilities, telecommunications: Web and social media solutions such as TMZ, Facebook, and Twitter are among the “newest “new data sources. A number of new businesses are now building “Big Data” environments that leverage consumers’ (conscious or unconscious) nearly continuous streams of data about themselves (e.g., likes, locations, opinions).

IDC had a chance to speak with Cameron Befus, VP of Engineering at Tynt, an emerging leader in analysis for media and advertising industries. A couple years ago, Tynt took notice of the growing volume of “copy and paste” actions on many information websites. They wondered if anybody cared. They wrote a little program that would track these actions (they added other actions such as “print” later). As Cameron noted, “we started showing it to publishers; their eyes got big; and they wanted to know more about what they could do with it.” At the one year mark that were collecting and analyzing over 4 billion data points (e.g., web site cut and paste operations) per month. Today (less than a year later), they’re approaching 20 billion data points per month with peaks loads approaching 30,000 events per second. To handle this load they deployed a set of Hadoop clusters (production and development) with over 100 nodes and half a petabyte of capacity.

After talking about these data sources, the next question IDC hears is, “so how do I decide if a Big Data solution can deliver business value for my organization?” Regardless of industry or sector, the ultimate value of a specific Big Data use needs to be judged based on one or more of three criteria.

Does it provide more useful information?

IDC spoke to a major retailer that is implementing a digital video system throughout its stores, not only to monitor theft, but to analyze the flow of shoppers through the store at different times of day, week, and year. It also wants compare flows in different regions. This effort makes it easier for the retailer to tune and assess layouts and promotion spaces on a store by store basis.

Does it improve the fidelity of the information?

IDC spoke to several earth sciences and medical epidemiological research teams using “Big Data” systems to monitor and assess the quality of data being collected from remote sensor systems; they are using Big Data not just to look for broad patterns (The obvious and traditional HPC use case), but to identify and eliminate anomalous and false data caused by malfunctions, user error or temporary environmental anomalies (think birds nest in the sensor).

Does it improve the timeliness of the response?

As my colleague, Steve Conway, noted in another blog post , several private and government health care agencies around the world are deploying Big Data systems to reduce the time to detect insurance fraud from months (after checks have been mailed and cashed) to days (eliminating the legal and financial costs associated with fund recovery).

As a CIO or senior IT executive, you need to understand what your organization is planning around Big Data. Only then can you begin to develop a Big Data IT infrastructure strategy to support it.

Richard Villars is Vice President of Storage and IT Executive Strategies with IDC. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

This data explosion also means that datasets are becoming increasingly large and difficult to manage via conventional database management tools. As data sets grow in size – typically ranging from several terabytes to multiple petabytes – businesses face the challenge of capturing, managing, and analyzing the data in an acceptable timeframe. Organizations are also struggling to understand the opportunity information provides through advanced analytics. The IDC study suggests that organizations with high rates of change in their business think about business operations differently. These customers often put data analytics to use as part of a wide range of business decisions often using data analysis to develop business strategies. In short these users make data based decisions both more efficiently and at faster speed than peers typical of their industry.

Barriers to big data adoption are generally cultural rather than technological. In particular many organizations fail to implement big data programs because they are unable to appreciate how data analytics can improve their core business. Business executives need to improve their ability to convey complicated insights to the organization and drive an effective action from the data analysis process. When getting started it is helpful to think of the following:

• Identify a problem that business leaders can understand and relate to – one that commands their attention.
• Don’t get too focused on the technical data management challenge. Remember that resources need to be invested to understand the uses for the data inside the business.
• Define the questions needed to meet the business objective and only then focus on discovering the necessary data.
• Understand the tools available to merge the data and the business process so that the result of the data analysis is more actionable.

Traditional business intelligence systems have historically been centrally managed in an enterprise datacenter with the scalable server and high-performance storage infrastructure built around a relational database. Now enterprises are working to extract competitive business value – and ultimately revenue – from a growing sea of data. These big data implementations leverage diverse sets of distributed semi-unstructured and unstructured data types which frequently start with mathematics, statistics and data aggregation efforts. Big data analytic software is increasingly deployed on massively parallel clusters leveraging Apache Hadoop framework, distributed file systems, distributed databases, MapReduce algorithms and cloud infrastructure platforms (for time-to-market and scale needs). These big data applications are becoming a source of competitive value for enterprises as firms monetize information by building data products and services. Despite the uncertainty many organizations face, several things are clear:

• Big data is not an extension of traditional BI and requires new thinking.
• New skills will be required including mathematics, statistics and business capabilities necessary to build revenue models around data.
• New tools and architectures such as Hadoop and MapReduce algorithms will be required to deal with largely unstructured or semi-structured data.
• Many of the resulting business opportunities created through big data will be industry specific (genomics, healthcare, oil & gas, finance, etc.) making the focus on time to revenue more critical.

Because speed is strategically important, it will be tempting for business to move forward without IT support. IT needs to recognize that it needs to think differently (and quickly) and fight for a seat at the table as big data strategies are developed. CIOs need to both understand what their organization is planning around big data and begin developing a big data IT infrastructure strategy. In doing so CIOs need to understand the potential value big data represents to their organization and industry. Skill sets will also need to be evaluated with a focus on training as core analytical skills are not widely available in the market and in many cases will need to be developed internally. IDC believes that building successful business cases around big data can only be accomplished through a tight alignment critical thinking across both IT and the business. This will require out of the box thinking as well as moving outside the traditional IT comfort zone as traditional data warehousing models may not be appropriate in order to effectively monetize the big data opportunity.

Matthew Eastwood is the Group Vice President of Enterprise Platforms for IDC. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

The consumerization of IT – once just a buzzword – is today very much a reality. While work and personal computing haven’t fully merged, the line between personal and professional tech is undeniably blurry.

The same overlap is true for the cloud. I frequently hear cloud computing framed in terms of “personal vs. professional cloud”, which makes the two types of cloud seem somehow diametrically different from each other. I believe that artificial distinction can be misleading.  Think of it more like a Venn diagram. There are some fundamental features which are consistent regardless whether the cloud serves business or personal uses.

A very basic similarity between the two types of clouds is what end users expect in terms of service. We have little patience for a work PC that can’t perform as well as our home computer; and we want our experience with the professional cloud to be just as seamless and flexible as the way we experience personal cloud services like  banking, audio and video data streaming, and online gaming.

Why such high expectations, especially for a relatively new mode of computing in terms of consumers? If you consider just how ubiquitous personal cloud computing has become -,– it’s much easier to understand.  Personal cloud computing underlies much of our online activity: online banking, e-filing taxes, online gaming, downloading movies and music, even buying groceries. It’s the engine that enables us to truly “live globally” – keeping up with family and friends around the world, and buying beyond the walls of where we live.

All of this online activity is done via what we call the “public cloud,” where the computing infrastructure is owned by the provider, and applications, data and services are delivered to the end user. Given its popularity, innovations for the public cloud have been a major driver of broader cloud technology. Look no further than Facebook’s recent announcement of the Open Cloud Platform, an effort which AMD is proud to support.

The cloud for professional use has developed alongside the personal cloud, making similarly big strides. But professional cloud-based services also face unique deployment challenges – namely, how do you implement  a cloud computing solution in a way that protects business interests?

Looking at it from a high level, there are three main implementation options. A good rule of thumb is that security is often inversely related to convenience:

• Public cloud: Using a third party infrastructure is often more cost-effective and convenient, but can be scary for businesses. Can you trust vital business functions to IT departments other than yours? High convenience, comparatively low security.
• Private cloud: This is the other end of the spectrum and there are some that argue it’s a great way to get the functionality without the worry. In this scenario, you essentially host your own proprietary cloud. More secure, but not remotely convenient – you really need to know what you’re doing to make this work for you.
• Hosted private cloud: For most businesses, this is the goldilocks scenario. Here you use a service provider to build and house your private cloud, delivering both moderate convenience and security.

Which option do you use for your professional cloud? For more info on how we define public vs. private clouds, have a look at this blog from my colleague Tim Mueting.

Margaret Lewis (@margaretjlewis) is a Product Marketing Director at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

The year was 1999 and I was engaged in experiments designed to simulate Internet performance variation and the use of dynamic DNS resolution to optimize the network performance experienced by web browsers. The problem was that while we were running experiments in our lab, someone had forgotten to turn on egress filtering for our router and our simulation traffic in the lab was propagating out of our lab and into the wild.  Oops.

In 1999, pulling off this kind of scalable experimentation involved the purchase of multiple servers, and at this time 1U servers were not the norm. 4U servers were common. Cobbling together a lab with 50+ servers, racks, cabling, patch panels and switches could easily cost $250,000.

And bandwidth? A T1 line providing ~1 mbps of bandwidth ran close to $1,000 per month. To run the experiments we needed to run involved assembling the server and networking infrastructure for the lab and provisioning 30mbps of wide area connectivity along with the monthly cost of bandwidth utilization. In short, the servers and bandwidth needed to do our experiments required an investment of close to half a million dollars.

But something happened between then and now. The emergence of public cloud computing platforms and the ability to consume compute capacity in a “bursty”, elastic way has totally altered the economics of large-scale compute. Today, compute capacity and bandwidth that, in the past, would have cost hundreds of thousands of dollars to provision can now be had by individuals for about the cost of a family’s monthly cell phone bill.

I recently needed to gather a statistically meaningful slice of web content to mine for insights into the implementation choices being made by web developers. You can buy market analyst projections or read media reports, but I was much more interested in gathering empirical data from real web sites. So I acquired a list of 100 million domain names and decided to harvest the home pages from all 100 million domains. I wanted to use those pages to analyze interesting information regarding web developers’ use of JavaScript, rich media content (e.g. Flash) and to correlate the use of different implementation techniques with specific geographies. And I didn’t want to take weeks to do this nor did I want to spend a bunch of money. Enter cloud computing.

I employed a cluster of virtual machines from one of the major cloud vendors and over a 3 day period crawled all 100 million domains. While my crawlers were running, I found a free IP-to-GEO database and converted the IP addresses for each domain to country codes. Then I modified an HTML parser to siphon the information I was interested in out of the web pages I had gathered.

I did my crawling during a time when my cloud provider was offering free in-bound bandwidth. As a result, it only cost me about $60 to crawl the home pages of all 100 million domains. The ongoing storage costs for the data is a little over $100 per month. In short, for a little less than $200 I could do a project that only 10 years ago would have cost several hundred thousand dollars.

The Internet is not just a place to hang your web site. It is a source of strategic and competitive insight about your business and emerging opportunities. The most successful companies are increasingly savvy to the criticality of using Internet data to provide differentiated insight. In the past, the infrastructure costs required to tap into Internet data in sufficient volume to gain meaningful insight would have only been possible for large companies with fat wallets. No longer. Any individual, sitting at home, can command the compute resources of hundreds of machines to gather and sift data at volumes that would have been financially unthinkable only 10 years ago.

We are living in a great age for data.

For more background on this story and my experiences, be sure to sign up to watch my keynote address at the virtual Cloud Slam 2011 conference on April 20: www.cloudslam.org

Keith Lowery is an Engineering Fellow  with AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

Data-intensive applications for things like cryptography and weather forecasting have been part of high performance computing (HPC) from the start, ever since the 1950s. But the power of today’s HPC systems — large clusters and purpose-built supercomputers — has made it feasible to tackle bigger versions of familiar tasks and a host of previously intractable problems.

The challenges cover a broad spectrum, including fraud detection, anti-terrorist analysis, social and biological network analysis, semantic analysis, financial and economic modeling, drug discovery and epidemiology, weather and climate modeling, oil exploration, power grid management, and many other areas.  The common denominator is that the problems are large and complex enough to require modeling/simulation using HPC resources.

Where HPC is concerned, IDC defines data-intensive (“big data”) problems broadly to include tasks involving sufficient data volumes and complexity to require HPC-based modeling/simulation. The problems can employ structured data, unstructured data, or both. They can come from traditional HPC domains in government, industry and academia–or they can be upward extensions of commercial problems that have grown large and complex enough at the high end to require HPC. In addition, “big data” can accumulate from the multiple results of iterative problem-solving methods in sectors such as manufacturing (parametric modeling) and financial services (stochastic modeling).  So, small and medium-size enterprises (SMEs) are also encountering “big data” challenges.

Some problems involve “finding a needle in a haystack,” that is, locating a discrete item that already exists in a database. This style of problem-solving usually employs relational databases (RDBMS) and traditional search methods.

Other problems are more complex and involve “finding patterns in shifting sand.”  Problems of this kind tend to involve unstructured (NoSQL) data and newer methodologies and special software frameworks such as MapReduce and Hadoop. They involve similar tasks: pattern matching, scenario development, behavioral  prediction, anomaly identification, and analysis of relationships using graphs. They’re for things like catching terrorists before they leave the airport, or catching bank fraud before the criminal gets the money, or protecting the US power grid before it crashes. Some of the powerful algorithms in this domain originated in classified government.

The stakes can be high in relation to economic value, competitiveness or national security. Take fraud detection as an example. Business fraud detection could save millions of dollars, and government fraud detection could save billions. Recently, EBay bought supercomputers to combat fraud in the PayPal system.  Italy’s big government agency, INPS, acquired a supercomputer to attack health care fraud on a national basis.

The U.S. may be heading in the same direction. The FBI estimates that 10% of transactions in federal health care programs – Medicare, Medicaid, Veterans Affairs and so forth – are fraudulent, costing about $150 billion a year. Price Waterhouse Coopers thinks it’s three times that amount.  Today, the health care data is spread across five gigantic databases. As a result, no one can see all the data at once, fraud is detected after the fact, and the government recovers only about $1 billion a year.

Oak Ridge National Lab has submitted a proposal to unify all these databases and perform fraud detection using a Cray supercomputer nicknamed “Jaguar” that features 224,000 AMD Opteron™ processor cores. This solution could save $50 billion a year by analyzing the data in near-real time. The same methods could be applied to other criminal behavior, terrorist activities and many of the other applications I mentioned.

Most “big data” problems requiring HPC-level solutions will be run on large clusters, but the most daunting data-intensive problems are already going to more purpose-built HPC systems (e.g., the Ebay/PayPal, INPS and U.S. federal  health care examples). IDC expects this trend to continue.

In sum, “big data” has long been an important part of the HPC market, but recent technology advances have given data-intensive computing much higher potential as a horizontal market. It’s back to the future, with a new twist.

Steve Conway is Research Vice President, High Performance Computing for IDC. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

If you use the internet, that massive “series of tubes,” there is a good chance that you are passing your electronic bits through an AMD Opteron^TM processor.

Last year we did some quick calculations and realized that we had sold more than 2 million AMD Opteron processors to companies that are involved in the web and cloud business.  So if you are searching for data, sharing vacation photos, streaming media, collaborating or sharing those funny dog videos, odds are you are relying on AMD Opteron technology somewhere in the chain. And typically those bits are passing through some of the largest data centers on earth, dubbed “mega datacenters.”

The interesting thing about these large web/cloud companies is that their datacenters are very different from what you typically find in most glass houses.  Imagine tens, or even hundreds of thousands of servers all packed very densely into the same building, all chugging away on the same application. That is so different from the typical environment where servers can be like snowflakes with every one being slightly different.

The guys who run these mega datacenters know a lot about efficiency, so much so that they often build their own systems.  One extra USB port might only be an extra watt to you, but when it is multiplied across 100,000 servers, suddenly it starts to add up.  But the problem is that as you start to branch out into this world of mass customization, you lose the ability to share best practices.  Your infrastructure becomes your secret sauce and you don’t want anyone to know your secrets – even if that could achieve a greater good for society. 

This week AMD joined Facebook, a  worldwide leader in social media, in an ambitious new project that is very counter to the way the industry has run up until now. Today, Facebook announced the Open Compute Project for server / data center designs and architecture. The goal is to make many of Facebook’s design decisions, requirements, and best practices available throughout the industry.  Companies will be able to share best practices around software, hardware, and datacenter design. Through this open communication, collaboration and sharing of best practices, more of the “know how” will seep out into the industry, and at the same time, more of it comes back to Facebook and the people on the Open Compute Project as standards.  And standards are good.

Cooperation between the design and development players in the cloud computing space can help to drive a more efficient energy consumption model, ultimately helping to get out in front of the challenge of power consumption in the data center.

As a representative for AMD on the Green Grid, I certainly have had exposure to a lot of different companies that are wrestling with their power efficiency challenges today, and seeing a leader like Facebook stepping in to open up their knowledge and best practices is a great benefit to the rest of the market.

AMD and Facebook have collaborated in the past on platforms and power efficiency and we see participation in this project as being critical to our continued mutual success, as we both continue to drive towards more power-efficient compute capacity.

Today marks a new chapter in the era of power efficiency in the mega data center, and AMD is happy to be there with Facebook as a part of this new initiative.

John Fruehe is the Director of Product Marketing for Server, Embedded and FireStream products at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

WorldHostingDays is an annual event in Germany that covers the European hosting market, a market that is rapidly changing to meet the new realities of the cloud.  Cloud has such an impact on the hosting market, that the event changed from “WebHostingDay” from last year to its new name, reflecting that there is so much more than just web hosting that happens in the hosting business – and it takes more than a day to cover it all.

This week Dell introduced their new Dell PowerEdge^TM C5125, a 12-server dense chassis that I wrote about in an earlier blog.  I had an opportunity for some hands on time with this system and was even more impressed after seeing it live – quite an accomplishment to fit 12 servers into a 3U rack space.

I also had an opportunity to drop by the Rausch Netwerktechnik booth to see the new “Bigfoot” server and storage system.  This server is a bit larger, occupying 4U in a rack, but what it can deliver is truly amazing.  The chassis is divided into 2 different sections, the first holding up to 48 3TB hard drives, giving a total storage footprint of 144TB – all in one box.  But the more amazing part is that the other half of the chassis is occupied by a server motherboard.  Rausch offers several different choices of AMD Opteron™ processor-based motherboards, from as little as one AMD Opteron 4100 Series processor with four cores, to as many as four AMD Opteron 6100 Series processors for a full 48-core system.  So, whether you are configuring as a NAS and simply need a small amount of processing power to serve up storage, or whether you want to create a core-dense cloud platform that includes a large storage footprint and plenty of storage, this system will meet your needs.

My conversations with the hosting community were very straightforward.  The challenges that they face, especially with the onset of cloud computing, continue to be the need for more cores and more power efficiency, a combination that AMD excels in, delivering more cores per socket than anyone else in the x86 server business. And we do it with more power efficiency; today’s dense 12-core processors are in the same mainstream power band as yesterday’s 6-core processors.* As we move forward into the “Bulldozer” era later this year, the core count will go up to an amazing 16 cores, but the overall power and thermal footprint should stay the same – making for a very interesting back half of the year.

John Fruehe is the Director of Product Marketing for Server, Embedded and FireStream products at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

*12-Core AMD Opteron™ processor Model 6100 Series and Six-Core AMD Opteron™ processor Model 2400 Series share the same mainstream power band at 115W thermal design power (TDP)

The next couple of months promise to be big for the cloud. Just take a look at the lineup of events that AMD is sponsoring that focus on cloud-related topics: World Hosting Day 2011, the global extension of the popular web hosting conference; Cloudslam 2011, the virual cloud computing conference that is now going hybrid;  Red Hat Summit with the catchy slogan “Platform. Middleware. Cloud. Mix well.” – and we can’t forget GigaOM Structures 2011 with its controversial tone and impressive line-up of cloud computing leaders.

To celebrate, we’re hosting an ‘Accelerating Cloud Computing’ blog series with IDC right here on the AMD@Work blog. Our goal is to highlight the trends that are fueling  the cloud frenzy and take a closer look at topics such as Big Data, cloud hosting and innovative platform designs.

Speaking of Big Data – it is becoming increasingly clear that Big Data is tightly woven into the future of the cloud. In June of 2010, we asked “how do you measure the growth of the cloud?” In the short time since we posed that question, we’ve continued to see staggering growth. For example, there are now an estimated 1.99 billion Internet users worldwide (source). Even more shocking, we’re now uploading the equivalent of 150,000+ full-length movies each week in the form of online video – that’s 35 hours a minute for those who are counting (source). This explosive growth rate – combined with no signs of slowing – begs a new question: how do we accelerate cloud computing?

One group where this question is top of mind is what has been the traditional web hosting community. As the cloud transitions from hype to reality, these ISPs have recognized the opportunity to broaden the scope of their services. They have been quietly evolving and reinventing their business models, to morph themselves into cloud service providers – a category that didn’t exist just a few short years ago. Forrester predicts that in 2011 hosted private clouds will outnumber internal clouds 3:1, and cloud services providers appear to be ground zero for this activity.

Don’t forget that to run any type of cloud  successfully, you need a solid computing platform; may I suggest a platform based on leading-edge processor technology from AMD! There are a growing number of innovative AMD-based cloud platforms emerging from our hardware partners, like Dell’s PowerEdge C6105, C6145, and the newly announced C5152 – check out my colleague John Fruehe’s blog on this “sled” system.

Make sure you check back and catch our “Accelerating Cloud Computing” blog series. Next up we’ll have reports from World Hosting Day and Cloud Slam, followed by an examination of the business value of big data and its practical applications. We’ll also touch on the role of GPUs in the cloud before closing out the series.

Margaret Lewis (@margaretjlewis) is a Product Marketing Director at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

The world of cloud computing got a wake-up call recently when Vivek Kundra, the Chief Information Officer for the U.S. Federal Government, announced his “Cloud First” policy requiring federal agencies to identify at least three in-house IT functions they could migrate to the cloud, and then move at least one of those services to a cloud provider (either private or government) within the next year.  

When the announcement was made, the government and vendor IT community started scrambling to prepare for the migration this year, as well as greater demand for cloud services in years to come.  Many government agencies, including NASA, Department of the Army, and the Securities and Exchange Commission, are ahead of the game having already adopted cloud services.*  Some have been firmly in the “let’s wait and see” camp.  Those agencies are now working quickly to put a strategy in place. 

As part of the announcement, Kundra also set an aggressive target of closing more than 800 of the 2,100 data centers the federal government currently operates over the next four years.  This creates additional pressure for agencies to move to the cloud and significantly improve the efficiency of the remaining facilities through virtualization, higher core counts, greater density, and other best practices.

According to Kundra’s 2011 Federal Cloud Computing Strategy, the plan is designed to reign in the federal government’s $80 billion annual IT budget (not including the electric bill!).  It’s also validation of the considerable benefits of outsourcing certain IT functions using a public, private, or hybrid approach to the cloud.

Like private sector businesses, government and education institutions are attracted to the flexibility, economies of scale, mobility and service oriented features cloud services can provide.  Yes, they may also share concerns about the loss of control, security, interoperability, and other issues, but those concerns begin to fade as one learns more about the different cloud service models, enhanced security, redundancy, performance and operational features that are available.  In fact, many public sector agencies are opting to operate their own private cloud as a way to maintain full control of their operations.  Others are adopting a hybrid approach that allows them to keep their most sensitive services and data in their own private cloud operation and outsource less sensitive operations to qualified public cloud providers.

We’re excited at AMD because our products are particularly well positioned for the cloud. True, we’ve been a processor of choice for major cloud vendors and date center operators for years.  And now we are poised for significant growth because our products offer an optimal mix of the features that matter for the cloud: performance, virtualization optimization, energy efficiency, security and value.  We also offer a broad portfolio of platforms to fit and scale for cloud implementations of any size.  And it just so happens that public sector IT managers tell us that is exactly what they are looking for as they begin their migration to the cloud.

For more information on AMD and cloud computing, read the white paper: Understanding Architectural Priorities Inside Government Clouds.

Attend this webinar and hear industry experts from AMD and HP discuss best practices and common mistakes to avoid when adopting cloud computing and other virtual technologies into your infrastructure: http://gcn.com/webcasts/2011/02/amd-building-a-cloud-computing-infrastructure-022211.aspx?tc=page0

Steve Kester is the Director of U.S. Government Affairs at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

* State of Public Sector Cloud Computing, May, 2010 – Office of the U.S. Chief Information Officer -  http://www.cio.gov/pages.cfm/page/STate-of-Public-Sector-Cloud-Computing

Parallels, the host of the summit, is a very interesting virtualization company.  Founded in 1999, Parallels offers an innovative range of products that support all major PC platforms and operating systems covering both consumer and commercial segments. AMD’s intersection with Parallels has been on two fronts – servers and graphics. And if you stop by and visit us at booth #112 during the show you can check out our multi-core processor technology designed specifically for cloud environments, the AMD Opteron™ 4100 series processors – the perfect complement for the Parallels Virtuozzo Containers product. These processors feature energy-efficient innovations that make them one of the world’s lowest energy consuming x86 server processors per core.

And when optimal performance and VM density are the goals, the AMD Opteron 6100 Series processor delivers up to 12 cores per processor for an outstanding virtualization option. In fact, the AMD Opteron 6100 Series processor recently received an InfoWorld 2011 Technology of the Year award with its massive virtualization capabilities specifically noted.

You will also be able to see graphics virtualization technology from Parallels running with our AMD FirePro™ professional graphics products. Graphics virtualization is going to be another key discussion in the world of cloud computing this year.  AMD has partnered with Parallels to give high-end professional users the ability to finally get graphics acceleration in a virtual machine with Parallels^® Workstation 4.0 Extreme.  In other words, you can now get dedicated graphics within a VM with acceleration for all your professional engineering or digital content creation applications. In addition, as with all ATI and AMD FirePro graphics cards, this professional application is optimized and certified to provide one of the best and most reliable experiences demanded by users. With multiple cards in a single system you can achieve dedicated graphics acceleration with multiple virtual machines thereby helping to increase the efficient use of your hardware assets.

We’ll also be showing off AMD’s new ATI FirePro™ RG220 graphics card with Parallels Workstation Extreme. The ATI FirePro RG220 is a full performance graphics accelerator plus IP transmission card.  This combination enables remote graphics output with the PCoIP protocol over any LAN or WAN to a zero thin client device. So for those VDI sessions with complex graphics requirements you can now provide a dedicated, graphics accelerated, VM in the datacenter that is available to any user across the network.

I will be delivering a presentation at the summit on Thursday at 1:30 p.m. that will dig a little deeper into AMD FirePro™ and AMD Opteron™ processor server technology for the cloud.  And, as an added incentive, I’ll be giving away an ATI FirePro V5800 Professional Graphics Card immediately after the presentation.

And if that isn’t enough, while you’re at the booth don’t forget to register to win an HP ProBook 6555b laptop which we’ll also be giving away at the end of the show.

Hope to see you there.

Tim Mueting is a Product Marketing Manager at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

One of the many things I love about my job here at AMD is that I get to meet and talk with a very diverse group of industry leaders who are tasked with analyzing, planning and implementing the latest and greatest in IT solutions and technologies.  I’ve watched virtualization as both a technology and a strategy evolve and transform how datacenters and even desktops are managed.  I’ve watched the dramatic change in how we evaluate x86 server platforms in terms of performance, power efficiency and overall value and how virtualization has played a key role in this transformation.

The advancement in virtualization technology has now led us to cloud computing.  Recently, I was fortunate to contribute, along with Margaret Lewis, in the writing of the AMD sponsored Cloud Computing Clusters for Dummies book.  In the book we discuss cloud computing in general, what some of the unique server requirements are for cloud computing, and how AMD is driving the adoption of cloud computing as an IT infrastructure.

There is no question now that cloud computing is real – though I wonder sometimes if the real question isn’t “just what is cloud computing”?  It seems to me that the definition of the cloud depends on who you talk to.

Today it is difficult to open a trade journal, read a blog, or even watch TV without finding some reference to “the cloud”.  Mothers are creating, editing and sharing family pictures via “the cloud”.  Business men and women are collaborating on documents, spreadsheets and presentations all through “the cloud”.

We defined “the cloud” in our Cloud Computing Clusters for Dummies book as “the next stage in the evolution of the Internet;  the means through which everything – from computing power to computing infrastructure and applications, from business processes to personal collaboration – can be delivered as a service wherever and whenever needed”.  The cloud is a set of approaches that can help organizations quickly and effectively add and subtract resources in almost real time.  Cloud Computing is about both the business model and the technology.

Some would lead us to believe that in the near future everything will be in the cloud -that the cloud is in fact a replacement for today’s traditional data center.  Prominent industry leaders have predicted that in the future the traditional data center will be replaced by a small number of very large cloud-based datacenters.

Still others are more cautious and are voicing concerns about security, privacy, and regulatory compliance.

Traditionally, Cloud Computing is said to have two distinct models.  The public cloud, like those you hear about from Amazon, Google, Microsoft and others where services are offered on a per use basis and the underlying infrastructure is shared amongst all customers.  And the private cloud, the cloud infrastructure that sits behind the firewall but provides cloud-like services to its select set of internal customers.  More recently we have been hearing talk about a third model – a hybrid model, or the Hybrid Cloud.

According an Information Week interview of HP’s Russ Daniels, CTO for cloud computing and VP of Cloud Strategy, HP’s take is that “virtually every enterprise will operate in hybrid mode, with some of its operations on-premises and some in the cloud”.  Daniels went on to say that contrary to some theories put forth, cloud computing is not a replacement for the data center.

This seems to be a much more practical approach to me.  Yes, some small businesses may find it practical and cost-effective to move all of their IT processing to the cloud, and certainly all of us will use cloud-based services (think social media and email) at an increasing rate.

But the enterprise is a much more complex environment.

I found this definition from TechTarget’s website SearchCloudComputing.com:

A hybrid cloud is a cloud computing environment in which an organization provides and manages some resources in-house and has others provided externally. For example, an organization might use a public cloud service, such as Amazon Simple Storage Service (Amazon S3) for archived data but continue to maintain in-house storage for operational customer data. Ideally, the hybrid approach allows a business to take advantage of the scalability and cost-effectiveness that a public cloud computing environment offers without exposing mission-critical applications and data to third-party vulnerabilities.

Both Microsoft CEO Steve Ballmer and Paul Maritz, president and CEO of VMware appear to agree on this concept.  In a recent interview with ITworld, Mr. Ballmer was quoted as saying “I think the new solutions that get pioneered in the cloud will also get retrofitted so they can be run on-premises by customers, because no customer is going to have everything in the cloud. People are going to have these kinds of mixed environments”.

Similarly, Mr. Maritz recently noted, “The new service based on VMware vCloud Datacenter enables enterprises to not only improve efficiency internally, but also be in a position to tap into external resources in a non-disruptive way when it makes business sense to do so.”

The cloud is indeed real, and my guess is that in the future most if not all enterprises will realize the benefits of some form of hybrid model.

I’m interested in hearing your thoughts on cloud computing and the hybrid cloud model specifically.  Are you moving any of your enterprise applications to the cloud?

Tim Mueting is a Product Marketing Manager at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

Enter the AMD Opteron™ 4000 Series platform, formerly codenamed ‘Lisbon.’ Designed specifically for cloud environments, the 4000 Series gave datacenter managers the option for powerful and highly scalable solutions with the lowest power consumption in AMD’s history. With the right solutions on the market, we believe 2011 will be the year of the private cloud in large enterprises as well as SMBs. And, while we expect our 4000 Series platform to continue to excel here, given the outstanding virtualization capabilities in our 12-core AMD Opteron 6000 Series platforms, we fully expect to see 12-cores of processing muscle in a number of private cloud environments. For example, HPC in the cloud is a topic that is gaining steam, which is an area AMD Opteron 6000 Series platform is ideally suited for.

We saw a huge explosion of data in 2010, and that is only set to grow in 2011. Cloud computing will continue to be an integral part of the way we manage this information and make IT as efficient as possible.

www.youtube.com/watch?v=k8SmkneuE9A

Don’t forget to come back tomorrow for another clue and your chance to win a Toshiba Satellite A505D-S6987 laptop. See the complete contest rules here.

Julie Lass is a Senior Public Relations Manager at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

We decided to explore the knowledge base from the DCS group and build the same approach with customers looking for something between an off-the-shelf solution and a completely custom solution.  The journey began with customer conversations.  We listened to the customer feedback and realized that they were looking for a variety of features but they agreed on the need to maximize the performance per watt per dollar.

The result of these conversations is the newly released Dell PowerEdge C6105 server.  The C6105 is a 4-node shared infrastructure server in a 2U form factor. It is loaded with the AMD Opteron™ 4100 Series processors, which have remarkable power efficiency at around 6-watts per core.   

We have delivered an extremely efficient server with:

• A lower cost of ownership,
• 48 cores, 280 watts in a 2U for enhanced virtualization
• Extremely efficient hot-plug redundant power supplies
• Individually serviceable nodes to increase uptime
• 1.5x more storage than comparable products

The PowerEdge C6105 is geared toward scale-out cloud, Web 2.0, hosting, and HPC applications where core count and power efficiency are more important than raw compute power. It’s also great for power, space, weight and budget constrained data center environments such as co-located services (multiple services in one location) and large public organizations such as universities, and government agencies.

Take a look at this video to learn more about this server’s anatomy.

www.youtube.com/watch?v=5Bd-cIPvWas

Armando Acosta is Data Center Solutions Product Manager at Dell.  His postings are his own opinions and may not represent AMD’s or Dell’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD and Dell are not responsible for the contents of such linked sites and no endorsement is implied.

Visit the PowerEdge C6105 page on Dell.com.

The answer: Yes. And it has been done by The New York Times in 2007. How? Apache Hadoop, a software framework designed to support data-intensive applications.

AMD, along with companies like HP, IBM, and Yahoo!, was a sponsor of the 2nd annual Hadoop World conference last month, which was attended by a Who’s Who of the big data and cloud computing worlds. Companies ranging from financial institutions to large cloud providers to social network sites were there, united by one common goal: how to most efficiently and effectively manage today’s explosion of data and information.

For the uninitiated, Hadoop is the open source community’s answer to Big Data challenges. Administered by the Apache Software Foundation, Hadoop is designed to enable advanced analytics over large amounts of data in cloud environments. Exactly how much data are we talking about? Hadoop is designed to allow applications to work with thousands of nodes and petabytes of data – or one quadrillion bytes.

In my mind, Hadoop is a natural extension of both the shift to the Internet as the major data transport for web and cloud computing environments and the almost unfathomable proliferation of data. To support this trend on a hardware level, we’ve designed the AMD Opteron™ 4000 and 6000 series platforms to specifically address the rigors of these environments. Check out our latest whitepaper here on what AMD is doing in the realm of “big data.”

The Hadoop project also has support in high places, boasting an impressive array of collaborators including Yahoo!, Facebook, and Google. Major services firms such as NTT and IBM are building global practices around the Hadoop ecosystem.

Venture backed Cloudera, where Hadoop founder Doug Cutting now works, recently released an enterprise-ready commercial distribution solution for Hadoop. According to Ed Albanese at Cloudera: “We continue to innovate and expand the definition of what a Hadoop-based platform is, offering new and improved ways for enterprises to make sense of the extraordinary amounts of data they generate. The need for businesses to store, manage and analyze data quickly and cost-effectively is greater than ever in light of the exponential growth of information both online and off.”

For our part, AMD has long been a supporter of the open source community, and we’re happy to be involved in a project that represents a meeting of the minds of global software and cloud community. In fact, we are Gold Level sponsors of the Apache Software Foundation, which administers the open-source Hadoop project: http://apache.org/. Have a look at my blog from this past summer for more information on our involvement with Apache.

John Fritz is Director, Strategic Software Alliances at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

This was a good tie in to my discussion earlier in the day about cloud computing.  Companies see the opportunity for some major cost savings, but it will require some unconventional thinking.

Over a dinner of incredible Lebanese food, I was able to reflect with the team on some of the things I heard this afternoon about the future of the cloud and what it means to this part of the world.

As my first trip to the region I was amazed by not only the wealth of the region, but also by the technology.  I often use the story told to me by one of our US customers in the oil business that high performance computing (HPC) technology is so critical to the industry because “it costs about a million dollars to put a hole in the ground, so you want to make sure you pick the right spot to drill.”

Armed with that perspective, I hit Dubai.  Yes, those in the petroleum industry have ample access to technology, but that is not to say that they are without challenge in the region.  One of the most amazing things that I heard was that bandwidth and infrastructure are still not where they need to be just yet.  So, while the idea of a private cloud (inside your firewall) is perfectly acceptable, the idea of a public cloud is less appealing in the region based on things that many take for granted. 

This got me thinking about some of the discussion from the GITEX Cloud Confex that I was involved in yesterday.

I believe that most people were at the session to find out more about cloud computing because they have heard a lot, but still don’t know how all of this will impact their businesses. There is obviously still a lot left to work out, and much of the learning is happening in real time.

Some of the key cloud considerations that we all need to address include:

Security – I believe this is the #1 concern for people when addressing cloud computing.  But it was interesting how the conversations progressed. People are afraid to trust their data to a third party out of fear of security. However, when you peel back that onion, what you find is that these external organizations (like Amazon, Salesforce.com or Google) might actually have higher levels of security than you have. When your business is built around supplying a secure infrastructure, you get pretty good at these things.

Data Location – This was more of a concern the week before in Europe, but you need to think about where your data is going to live.  Or possibly live in the future.  Are you governed by legal requirements in your business or your country that put limitations on where data is stored?  It might be easy on the front end for your vendor to tell you where the data lives, but what about 6 months or a year? Be sure to ask.

Portability – The key to cloud technology is that it is portable.  Build a private cloud, expand it out to the public realm.  Build a public cloud and bring it in house to better meet your organizations’ needs. Build a hybrid cloud. They key on portability (as people found with outsourcing) is that with the right data structures and processes in place, things can work. But having the wrong processes in place might force you into bringing the whole service back into your data center.  You have to be prepared to do that seamlessly.

SLAs – Make sure you hammer out the SLAs up front.  Make sure you and your vendors are all on the same page when it comes to  SLAs.  Make sure you both have the same definition of simple terms like “access”, “downtime” and “response”.  It is far better to know these things up front.

Skill Sets – The idea that the cloud is outside of your firewall does not mean that you don’t need people with those skills inside of your firewall as well. The best way to help your partner understand what you want is by actually being able to articulate what it is that you want.

Continuity – Make sure you know how to keep your business and applications going, no matter what the environment.

As with any good panel there were advocates as well as skeptics. Be sure to listen to both sides as you plan out how cloud will impact your business in the future.

What are your experiences? Leave a comment and let us know your thoughts.

John Fruehe is the Director of Product Marketing for Server/Workstation products at AMD.  His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

The cloud appears to be an example of hype translating to sales.  According to IDC’s Worldwide Enterprise Server Cloud Computing 2010 – 2014 Forecast, spending on cloud computing products and services is expected to grow at a compound annual rate of 25% per year, and will generate $56 billion in annual revenue in 2014.

The federal government has also recently directed all agencies to start planning now to incorporate cloud computing.  The directive is part of the ongoing effort to streamline IT operations to meet a mandated goal of $3 billion in cost saving by the end of fiscal 2012.

The federal government diving in headfirst would appear to be a solid endorsement of cloud computing, but in some circles it’s also raising questions about cloud security.  Depending on who you talk to, securing the cloud can mean just about anything: detractors worry about external threats, internal threats, collocating of data and applications, privacy concerns… the list goes on.  One thing we can (I think) all agree on is that security is a real concern, and will have a direct impact on the success of cloud computing.

Whether we are talking about public clouds, private clouds, community clouds, or even hybrid clouds, providing a secure infrastructure to protect our data and guard against unauthorized access and malicious intrusions must be an essential component of any cloud computing implementation.  For all of us at AMD, true end-to-end security requires the complete integration of security components within all layers of the IT stack – starting with the hardware platform itself.  AMD offers a competitive range of hardware features that helps to ensure security on the hardware platform.

To that end, in 2004 AMD introduced a feature called Enhanced Virus Protection which sets portions of system memory aside as “data only.”  Any code resident in these areas may not be executed, only read from or written to.  AMD’s Enhanced Virus Protection, working in concert with OS support, acts as a preventative measure, localizing the virus where it will be short-lived, non-contagious, and eventually flushed entirely from system memory.

Beginning in 2006, AMD started shipping AMD Virtualization™ technology which included key extensions to enable the establishment of a “root of trust” within an initially untrusted operating mode. This technology enables pre-authentication of a hypervisor or virtual machine image before users are able to decrypt and load them.

In order to create a “root of trust” the processor is first reinitialized, which establishes a known secure execution environment where a software component called the secure loader (SL) is designed to execute in a way that cannot be tampered with.  The secure loader executable is then copied to an external device, such as a Trusted Platform Module (TPM) for verification.  This process uses unique bus transactions intended to prevent the operation from being emulated by software that could not be detected.

Most recently, AMD introduced AMD Virtualization™ technology for I/O Virtualization in all of our G34 and C32 platforms.  This technology consists of an I/O Memory Management Unit (IOMMU) embedded in the chipset which isolates virtual machine memory and protects against malicious attacks from system peripherals.  I’ve written at length on virtualization technology and how it has transformed the operational aspects of today’s datacenters, increased the utilization of existing resources and changed the way we evaluate performance, capacity and value of our hardware resources.  I believe that as virtualization continues to play a larger role in enterprise application and cloud computing, protecting systems at the I/O level becomes increasingly important.

Establishing trust and transparency across all the architectural layers in the IT environment, particularly a shared cloud environment, is critical to achieving end-to-end platform integrity.  As a founding member of the Trusted Computing Group (TCG), AMD has dedicated itself to the development and promotion of open, vendor-neutral, industry standards for trusted computing across multiple platforms.

Now that everyone’s talking about the cloud, let’s get everyone focused on innovation in cloud security – an effort from which we’ll all benefit.

This post also appeared on VMblog.com

Tim Mueting is a Product Marketing Manager at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

Not only did we solicit your participation in our online contest, we also met a number of you at VMworld in San Francisco.   A few of you were passionate enough about the subject to jump on camera to share your opinions and predictions. With VMworld Copenhagen this week, we thought it was a great time to share your video responses.

www.youtube.com/watch?v=9Rosv1KzT0k

Much like our online respondents, VMworld attendees defined the cloud as an end to the constraints of computing as we’ve come to know it.  They were excited about the possibilities of combining platforms, operating systems and applications that can benefit from the cloud computing model.  As we expected, respondents are using the cloud both in a personal and professional capacity – in more ways than we could even begin to count.

www.youtube.com/watch?v=O7KmtAuWeFg

Then we asked– how do you define the ideal cloud infrastructure? Over and over again we heard: the ideal cloud infrastructure is scalable and reliable.  We agree, and our recently launched AMD Opteron 4000 Series Platform is designed to address exactly those requirements.

So, how did our respondents do, and is there anything they missed?  If you’re at VMWorld in Copenhagen, we’d also love to hear from you.  Is the cloud computing hype in full force on another continent this week?

Tracey Blanton is a Product Marketing Manager at AMD.  Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.