Egenera Blog

…or, Having Your Data Center Cake and Eating It Too

If unified computing and converged infrastructure are breakthrough IT management concepts, then I’m still mystified why more users don’t protest and ask for more universal, Open converged infrastructure platforms – in addition to the platform-specific ones hardware vendors offer.

In response, Egenera is taking another step towards a more advanced converged Infrastructure: Today we are illustrating, with Citrix Systems, that converged infrastructure CAN be certified tightly-compatible AND offer the benefit of openness.

The Perceived Trade-off
There are definitely pros and cons to tightly-integrated data center stacks.  Clearly, a tightly-integrated system such as the Acadia Vblock (the coalition of VMware, Cisco and EMC) promise users simplified product integration, alongside the simplified logical operation you get with converged infrastructure (i.e. Unified Computing).

However, others point-out the down-sides of this approach, namely the customer looses the opportunity to choose the computing, virtualization and storage platforms they may have previously preferred.

Note that this set of trade-offs will get worse: Additional platform vendors are now jumping into the fray… HP offers its pre-defined Matrix stack, and HDS has announced its own Unified Computing stack (leveraging their own servers and storage); more platform-specific stacks are sure to follow.

The benefits of converged infrastructure need-not come at the expense of choice

So we ask: What if… What if you could make this request –

• Give me a converged infrastructure system – A wire-once system that supported both physical and virtual servers, with inherent virtual I/O and converged networking so any server profile (template) could be instantiated on any server
• Give me unified management – Replace a dozen point-products (including SW provisioning, HA, DR and capacity elasticity) with only 1-2.
• Let it use standard hardware – Use standard x86 boxes from a variety of vendors, and use standard Ethernet ports and switches
• Let it use standard SANs from any major iSCSI or FC vendor. You choose
• Let it support the virtual server platform of your choice

What would that be worth to you?

The benefits of converged infrastructure are simple to communicate – it vastly simplifies management and configuration of IT resources by pooling compute, I/O, network and storage resources, and then logically defining them as server profiles/templates. Doing this on Heterogeneous resources, however, is the Holy grail.

The benefits of compatibility and choice
At face value, the announcement today that Egenera’s PAN Manager software is Citrix Ready is mundane. But a closer look at the implications which show you can marry
-    Verified integration and stability between a virtualization platform and
-    A converged infrastructure platform that is open for various hardware and storage options

To be sure, there has been an inherent trade-off between tight integration and simplicity of operation, and the desire for openness and heterogeneity. And different classes of users will always prefer different combinations of this trade-off.  But why not have both?

Or, The tale of the universal remote vs. integrated audio

There are 2 IT infrastructure management philosophies within in the vendor community you should know about…  And you need to know how they’ll impact control data center infrastructures, cloud infrastructures, and IT operations.

As I mentioned in an earlier blog about Band-Aids vs Cures, the two philosophies are essentially

• Automate what you’ve got: Use loosely-coupled automation, scripting and run-book-automation to orchestrate the piece-part components that otherwise consume management time and effort
• Virtualize/converge your infrastructure: Logically abstract the infrastructure (analogous to what hypervisors do for software) so that you can construct it in any form you need, with a minimal level of components and complexity

These each have implications into medium- and long-term data center efficiency (operational, capital). Let’s take a look at what they are:

The “Universal Remote” strategy:

This is known as loosely-coupled automation and/or Run Book automation. At first glance, this approach is pretty cool. Build a Service Manager and GUI, and then simply write drivers that speak to servers (i.e. DRAC, iLO, etc.), switches, SANs, etc.   Whatever you’ve got, this gizmo will orchestrate it… just like the Universal Remote we have at home.  Like your home entertainment center, it’s a *great* solution if you don’t want to make upsetting changes to your stereo, wiring, or components (still occasionally use that old Video Disc player?)

Sure, it’s a great investment-protection strategy. Except

1. You still have a rats-nest of connections and components
2. You’re still paying for the software and annual support $ for all of the in-place assets
3. Problem management is complicated by an additional  level of abstraction/indirection
4. You’ve done nothing to reduce/simplify the overall infrastructure, capital expense, energy consumption, etc.

The “Integrated Audio” strategy:

This approach is known either as Converged Infrastructure or Unified Computing, and is more analogous to the OS Virtualization world. It seeks to create a virtual infrastructure of logically defined and composed I/O, networking, switching, and storage connections. It’s a bit more like the all-in-one entertainment centers, or at least the systems that are predisposed to work with each other with a single control.

This approach has pros and cons. Obviously it can require a new build-out – but one with much more “rip” than “replace” because so many physical components are now virtualized. All I/O, networking and switching are logically composed. The advantages are

1. Never re-wire the data center again. There are minimal cables and components to begin with
2. The infrastructure is infinitely re-composable – and is not limited by the physical topology
3. Since most point-products (and associated SW tools) are eliminated, there are less components to manage, and fewer SW tools to buy, maintain, and pay support for.

So be on the look-out for more about converged Infrastructure and unified computing. As data center drive toward greater simplicity and fewer ‘moving parts’, this approach will continue to accelerate. And while the Universal Remote is pretty cool, don’t forget about the mess of spaghetti you still have behind the cabinet.

Today, as in nearly every day, I had a conversation with someone who suddenly latches-on to the concept of Converged Infrastructure – it’s elegant, simple, efficient, effective. They get excited about its versatility, and how virtualization-of-the-infrastructure is the analogous next step to virtualization-of-the-OS.

Then… they inevitably ask “But what are your differentiators?”.  And that’s the easy part! Because we’ve been at the converged infrastructure thing for nearly a decade longer than Cisco, HP or IBM. And we know the customer needs and objections well. So, how are we different? Let me count the ways:

1. We’re designed to be platform-neutral. That’s right – No hardware lock-in. Our IP is embedded in PAN Manager software, not in the hardware. And  PAN Manager is being ported to multiple server vendors. So you’ll be able to select the servers you want, yet manage them on a single fabric from a single converged infrastructure management console.
2. We base converged infrastructure on standard Ethernet. You read right – no expensive FCoE or InfiniBand. No vendor-specific Ethernet appliances or expensive switches. No CNAs or HBAs. Just standard on-board Ethernet ports, cables, switches. (But yes, we bridge to FC if you  want)
3. We’ve already integrated software provisioning, HA/failover, and Disaster Recovery. That means you don’t need to buy or integrate multiple products or third-party solutions. With HP you’ll find you need to page across a dozen products to get the functions you need. To get there using Cisco you’ll need to add Ionix, clustering software, EMC Powerpath and more. Consider the dollars you’ll save on licensing, maintenance, training and operations with Egenera.

Sometimes simplicity, elegance and logical design is all you really need to make a good IT choice.

Compared to 12 months ago, the term “converged infrastructure” is now practically a commonplace term.

All major vendors — Cisco, HP, IBM — are offering their corporate version, their interpretation, their pre-configured systems.  And the press & analysts are eating-up the marketing for all of these proprietary solutions.

Meanwhile, in the adjacent cloud computing space, another different evolution is happening. No sooner do multiple vendors begin to offer public clouds then the community demands open interoperability between them.

So, where is the Open, multi-platform equivalent in the converged infrastructure space? A model that rejects hardware-specificity? My set of requirements would start with:

• designed to be hardware-independent
• operates on standard Ethernet with standard Ethernet switches
• is software-based
• unifies heterogeneous hardware on a single fabric
• integrates high-level services such as failover and disaster recovery
• rejects HW vendor-specific lock-in

I’m happy to say that this is Egenera’s mission.

And we’ve seen the light that others have missed.  10 years ago we too went down the tightly-integrated converged infrastructure hardware path, with the production-grade BladeFrame — with 10,000’s of blades in use on trading floors and data centers around the world.  And the light we saw was that the next logical step was to bring converged infrastructure technology into the software world — so we un-bundled our PAN Manager software, creating an opportunity for a hardware-independent Converged Infrastructure system.

Yet, we now see major vendors headed down a 10 year-old path of designing special-purpose hardware once again….  Meanwhile we like to think we’re enlightened, now pursuing the multi-platform software path. You use the hardware and networking you already prefer, while attaining the cost-efficiency and simplicity you desire.

Now, ask the other vendors about their plans to offer an Open converged infrastructure management system, and let me know what you find.

Egenera announced a new version of Dell PAN System and PAN Manager Software today that greatly expands the scalability of PAN Manager on a number of fronts. PAN Manager can now manage more blades on a single converged fabric , more VMs, and more servers. It also increases the IO throughput, utilizing 10Gbps Ethernet uplinks, 8Gbps Fiber Channel and 20Gbps between blade chassis.

As more applications go virtual, this increased scale and throughput will help customers consolidate and manage their applications. Of course, with PAN Manager, you can also manage your physical servers just as easily.

Egenera continues to deliver a simplified solution that enables customers to quickly deploy and manage their applications in a secure, highly available fashion. We also provide one of the only Converged Infrastructure products on the market today. Why does that matter? Well, it’s the convergence that enables you to reduce or eliminate hardware such as NICs and HBAs. That results in lower costs and also reduces the chances of failure.

Converge, Unify, Simplify.  That is a message that you’ll be hearing a lot from Egenera. We are the only heterogeneous vendor on the market today that provides a Converged Infrastructure on multiple hardware platforms. Today’s announcement of the next version of Dell PAN System is a major step forward in our strategy to deliver PAN everywhere.

This blog entry is about the need to simplify data centers, not to mask or automate the stuff that’s there already. And it’s about Converged Infrastructure: what it is and why you should know about it. And finally, it’s about why some of today’s infrastructure management products – which simply manipulate in-place complexity – are dangerous Band-Aids masking your real operation and cost issues.

The two philosophies
There are two major philosophies/architectures for how to manage production infrastructure. The first is to control in-place assets with code or run-book processes. The other abstracts the topology, then re-assembles it in real-time. One’s a Band-Aid. The other is a cure for complexity.

There is an analogy in the software world: OS virtualization vs. run-book server provisioning.  While automated run-book approaches work, you’re still limited by physical software. In contrast, virtualization’s abstraction has revolutionized and simplified the software world.

Here’s a bit more detail:

• In-place automation technology masks underlying topologies and complexity by manipulating the in-place infrastructure assets.  While this approach has had merit (it was also pursued for years by Cassatt and Scalent), it is ultimately constrained by in-place physical topology, and doesn’t reduce or simplify physical infrastructure or capital investments. This approach has yet to be proven in the production data center.
• Converged Infrastructure (CI) technology (also known as Unified Computing).  This approach first abstracts/virtualizes I/O, networking, and storage connectivity so it can be re-configured at will. In doing so, it reduces and simplifies underlying physical infrastructure (the need for NICs, HBAs, cables, switches, etc.), reducing both capital and operational costs yet still increasing agility/flexibility.  (More elaboration on these two approaches is available on the Fountainhead Blog)

Here’s an example of implementing each approach: With in-place automation, the management engine is constrained by physical data center topology.  If a physical server needs to be re-purposed but does not physically contain the proper quantity of NICs or  HBAs, or is not physically connected to required switches or IP load balancers, it cannot be used . Similarly, in the case of Disaster Recovery, if the recovery environment topology is not a close physical match to the production environment, then it cannot be used.  In contrast, Converged Infrastructure technology permits IT management to infinitely reconfigure physical servers, I/O, switches and IP load balancing on-the-fly. This allows individual servers – or indeed entire environments – to be configured to exactly to match specific topology. This also allows 100% sharable pools of servers, and 100% sharable failover (Disaster Recovery) environments.

How vendor approaches are playing out – It’s all about CI
The reasonable question to ask is: Which vendors are choosing to offer which approach?

As it turns out, nearly all vendors have elected to go with CI.  Cisco is most notable with their UCS, followed by HP with their BladeCenter Matrix, IBM with CloudBurst and open fabric manager, and finally (humbly) Egenera with PAN Manager.  All use forms of CI to abstract I/O, network and switching. Interestingly, Dell has chosen to break with the pack by acquiring in-place automation technology for their Infrastructure Manager product, based on in-place automation…. although they continue to OEM PAN Manager from Egenera.

Why has CI/unified computing essentially become the de-facto industry direction?  Because of its perfect complementarity with virtualization, and its ability to solve the data center infrastructure complexity problem.  Analogous to how hypervisors virtualize and simplify the data center software domain, CI effectively virtualizes and simplifies the data center I/O and network infrastructure domain. When used in concert, OS virtualization and CI abstract both software and infrastructure, letting IT operations configure and deploy entire server profiles at will, and with minimum time/cost/components.

Because of the broad CI following by most vendors, we should expect to see a degree of standards and, potentially, even interoperability.  And for these reasons, IT shops that go with CI will also experience a lower degree of technology risk simply because of its breadth of adoption. Furthermore, we’ll begin to see accelerated maturity of CI products enabling even more IT management functions.

Where Egenera’s placing its bets
For the past 10 years, Egenera has backed the Converged Infrastructure approach. Initially implemented as a bundled HW/SW combination on our BladeFrame, it’s now also available in a software-only form on Dell blades (and stay tuned for other platforms).

We believe that Converged Infrastructure and use of fabric-based networking is the future of the data center – and the way of simplicity.   And we believe that those who take another path will wake up one day to find themselves woefully behind the industry.

Egenera – as well as much of the IT industry – is standing at an exciting point in its evolution.

When Egenera was founded in 2000, nearly nobody had heard of virtual I/O or converged networking. In fact, blade servers were hardly part of the vernacular. But the company had a vision of simplified data center management, more “universal” HA and DR, and a more integrated approach to re-purposing servers.

Where we are

Today, with the maturing OS virtualization market and the newly developing unified computing domain, industry terminology has changed but our vision hasn’t.  We’re still focused on how unified computing, converged networking, virtual I/O and unified provisioning can radically simplify how availability and scale are easily delivered.  And using these concepts, virtual (or physical!) Infrastructure-as-a-Service (IaaS) platforms for cloud computing are simple to create and manage.

The inflection point the industry is going through is one where the management of workloads and infrastructure is changing. The change is away from physical point products, and toward a disruptive approach levering fully abstracted OS, I/O and networks, with unified, integrated services. Just look at the number of  products being replaced by VM management systems. And similarly, Egenera software is replacing the need for dozens of infrastructure management point-products with a unified, simplified approach.

The inflection point Egenera is going through is slightly different, but designed to help accelerate the disruptive changes happening in the industry. When we began the company, there was no hardware to support unified fabric computing, no virtual I/O, and no volume blade servers. But today, there are… so we are re-making the company to take advantage of two important assets: our PAN Manager Software, deployed in over 1,400 locations globally, and the advent of volume blade servers that are becoming pervasive in the data center.

Where we’re headed

One central tenet in Clay Christensen’s The Innovator’s Dilemma is that companies often need to re-invent themselves if they intend to excel in a changing market – even going so far as to introduce seemingly competitive products to their traditional base.  Which is why we’re taking our PAN Manager software and making it available on third-party platforms in addition to our battle-tested BladeFrame platform. This is not meant to imply that we are de-emphasizing BladeFrame in any way; rather, we are offering choice.

In so doing, we’re also changing the format of our company. We’re emphasizing our software. We’re emphasizing working with more hardware partners. Our management changes also reflect this, and our staffing does too… in fact, we’re hiring software engineers and field personnel in many important areas.

And our results so far?  Well, you’ve probably already read about new customers who’ve discovered the value of Egenera software coupled with Dell hardware. And you’ll be hearing shortly about how we’re beating some very large new entrants into the converged data center management space.  You’re also going to be hearing a great deal about the value of unified management in concert with unified computing architectures – radically simplifying how availability, disaster recovery, scale and IaaS are managed.

We’re betting on the fact that unified computing and converged data center management will evolve to perfectly complement the advent of OS virtualization.  And so, we’re organizing today to be the only independent provider of unified computing across multiple platforms, including our own. Our mission continues to be to bring radical simplification to creating highly available, highly agile data centers.  And stay tuned regarding our new products and platforms.

Tags: Add new tag

I had to chuckle a bit when I read a recent claim that Cisco’s UCS was the “first” unified computing system certified on Oracle 10g and 11g RAC.  That’s just plain false.

So stop the presses: To set the record straight, it was actually Egenera (using PAN Manager software) that was first… announcing 9i support back in 2003… even before Oracle had the “Certification/Validation” program in place!  Then we have the Validated 10g RAC solution last year, and the  Validated 11g RAC solution in May of this year. Check out the Egenera/Oracle history and Documentation and white papers.

What, you say? Egenera does Unified Computing?  Yes! And we’ve been doing so since 2001. Although we refer to our PAN (Processing Area Network) software by a different name, “Unified Computing” is characterized by use of stateless blades, abstracted/virtualized I/O, and converged/virtualized networking. In this manner, servers (both physical and virtual) and infrastructure can be instantly re-purposed in the cases of changing demand or equipment failure.

Now, for some perspective

Now, why does all of this matter? In a word, Maturity.

Obviously, being “first” is good for a bit of chest-thumping. But in this business, what matters to customers regarding  being first is (a) that you adhere to solid technology vision, and (b) that you’ve been evolving the product longer than others.

Regarding vision, Egenera has had our technology vision of the Processing Area Network, of stateless virtual I/O and of virtual networking (including load balancing) since 2001.  This was well before even OS virtualization was mainstream.

Regarding product evolution, Egenera’s software is currently in its fifth-generation version, and on two (soon to be more) hardware platforms. It’s all about maturity-of-solution, and of years listening to customer requirements. Some illustrations:

• Simplicity: First-and-foremost, our approach is simpler than anything else we’ve seen on the market. That includes the fact that it’s simple-to-use, and that it obviates the need for nearly a dozen IT operations point-products.
• Functionality: Our system is more than just about converged networking.  It bundles virtual I/O and converged/virtual networking with automated services like High Availability, Disaster Recovery, pre-configured server profiles, multi-tennancy, chargeback facilities and more. Plus it has been scaled to support hundreds of blades across dozens of locations.
• Platform support: Our unified computing approach is based on PAN Manager software, which first operated our  BladeFrame products, and now the same software operates the Dell PAN System
• Endorsement:  Finally,  our approach is in use by hundreds of companies in thousands of worldwide locations.

Net-net: Let our vision and maturity speak for itself.  And if/when others claim a “first”, dig a little deeper to check for the facts.

Everybody’s talking about IT infrastructure/operations simplification, but only a few are taking concrete action to do something about it.

So consider two views of simplification: (1) simplification at the “modular building-block” scale, and (2) simplification at the management/operations scale.

Egenera’s customers have been driving us on both of these issues, hence our announcement of a number of recent wins — across a variety of industries — which endorse our chosen approach. Follows are some concrete areas where we’ve focused our energies:

On simpler modular building blocks:

Working alongside Dell, we recently announced the Dell PAN Datacenter-in-a-Box. While I admit the name may not be as elegant as the product, it takes the most frequent configuration needs, and bundles them all with a mission-critical level of availability.  It’s essentially a standard blade enclosure with standard networking HW and standard SAN storage… complete with embedded virtualization (if you choose to use), fast server provisioning, and N+1 high-availability for any blade running any workload. And it takes < 1 day to install and have up-and-running.

So if you’re installing, say, an ERP system, here’s what you should think: I’ll run my scale-out DBs native, I’ll virtualize app servers and other services (using the VM of your choice), and I’ll wrap the entire environment (Including VLANs, embedded Load Balancing etc etc) with an N+1failover safety-net.  And you only have to buy a single item.

Or, for example, if you’re rolling-out a new VDI environment, you’ll want to run controllers, presentation servers and virtualized O/S instances all within a highly-available environment with local storage and HA. Again, a perfect use for a building-block approach to adding new services to your datacenter.

Elegant idea you say? But isn’t this just the equivalent of throwing a bunch of piece-parts together with a bow? Nope. Read on….

On Simplification of management & operations

So the next challenge is to simplify the more granular aspect of datacenter management - all of those pesky point-products you have to integrate, page between, and keep updated.  The diagram here illustrates what I’m talking about… as many as 13 products are needed if you’re in a reasonably sophisticated org managing both physical and virtual servers in a production environment.

The approach customers tasked Egenera with is to simplify operations and to make architecture more elegant.  So we sought to understand why these 13 products have been traditionally needed, and then how we can re-design providing these in a simplified, holistic, and unified manner.  The solution was our PAN Manager software that allows configuration and deployment of all of these functions in about 6 steps.  That’s essentially 1 product with 1 window to provide these 13 functions.

Don’t believe things could really be this simple?  Just read about our customers, the value they’ve derived, and what they have to say. They’re running production data centers in every industry, worldwide.

Lots of buzz in the air about Cisco entering the server market with the Unified Computing System – using the words “Revolutionary”, “Breakthrough”, etc.   Now, to have Cisco enter the server market is certainly revolutionary… but neither their technology nor their business value is particulary new. But their marketing sure turned-up the volume on it all.

Cisco has sat by the sidelines for a number of years, and watched Egenera in particular — as well as IBM, HP and even a few smaller players – enter the market with ‘converged’ networking and repurposeable servers.  As long as 7 years ago, Egenera was selling our BladeFrame, which is essentially what UCS is.   Stateless x86 servers, converged network backplane. Low component counts, low complexity. Extraordinarily low TCO compared with tradtional servers, I/O and networking.

7 years of maturity in the making

So what does 7 years (and hundreds of customers w/thousands of  installs) get you? Well, a few things. First, on the Hardware side, Egenera knows how to move to standard, lower-cost, volume hardware, demonstrated by teaming with Dell. And, Egenera knows how to make the entire *converged* system work over standard Ethernet – to keep costs, simplicity and risk to minimum.

On the Software side, Egenera has also figured out how to make the “special sauce” (our PAN Manager) run on other hardware too.  That means that all of the experience we have with running BladeFrames now comes as a mature 7th-generation software package.  That’s a lot of development.

So big deal. What does that mean?  Well, if you  look at Cisco’s solution (or HP’s for that matter), you find a very rich set of controls for the converged network, I/O and servers. Yawn.  (Egenera did all that years ago). Cisco then partnered with VMware for virtualization, and BMC for higher-level management like SW provisioning and failover. Good for them.  In the mean time, Egenera’s software had already embedded all of these functions within our “single pane of glass” GUI. That’s complete integration of a VM environment as well as HA and DR/COOP functionality – rather than working across multiple vendor GUIs.

What else you should know about Egenera

With this level of product maturity comes a few more things you should think about:

• Mission-critical production references – Yep, we got ‘em. In just about every market, in just about every geography.
• Refined, simple-to-use operation—the GUI is simple to use – and environments  can be set up in 6 easy steps.
• TCO studies, uptime statistics – Got those too. In fact, we can show you statistics from some of the largest users & hosting companies that show “five 9’s” of availability month-to-month.
• HW and SW certifications – Oh yeah… and by being around for so long, we’ve certified tons of applications and O/Ss on our system, so you don’t have to worry
• Integratability – And finally, we “play nice” in your sandbox. With a full Web Services interface (say you want to build a self-service xSP portal) as well as a monitoring API so you can use us with your favorite accounting/chargeback system.

And finally, let’s talk about Value. A bunch is coming out about Cisco’s Pricing.  Then there’s the analysis.  But either way, you’ll find that from a price/performance perspective, the joint Dell/Egenera solution stacks-up against the Cisco approach.

Ask us and you’ll see.