Architectures and Solutions All types of solutions as they relate to industries, technologies, and partnerships.

A world without communication

A few weeks ago California was rocked by something so small in magnitude but sent a wake-up call that hopefully will be heard around the country. Three fiber optic cables were cut below a manhole cover alone a city street. Three cables that then disabled every cell phone, land line, internet connection for over 80,000 residents in the South Bay of silicon valley. Even more alarming was the fact that emergency public agencies including fire department, police department, even hospitals were completely cut-off from the outside world and even their sister agencies. These cities learned first-hand what life is without communication. Signs were placed on the streets informing residents that if they had a medical emergency to drive themselves to the hospital. Police officers were to be flagged down in the street and in some cases curfews were debated if communication could not be restored for general safety. Police departments could not even run finger prints or check data bases on suspects they did detain. All this from cutting 3 cables snaking underneath public streets. When company’s talk about building systems and not just about combining individual technologies, this is hopefully in the back of their mind. When building a system it is easier to look for single points of failure, to look for failovers not across each individual technology but for entire systems. This is why architects take so long with their designs, not because they are slow because hopefully they are being thorough. The consequences of not being so can be devastating.

Read More.

Posted by Matthew Stein at 09:11AM PST

Application Delivery Services or Bigger Pipes

As I think about application delivery services in the network, I wonder if they’re really needed with the increased speeds networks themselves provide today. Does squeezing out a bit more throughput through cache, compression, content distribution, content-based routing, protocol optimization and XML processing really matter when users are seeing fiber to the house? The answer must be yes, or else there wouldn’t be a market for these services. I realize now that the primary need comes from the divide between developers and network administrators. When an application doesn’t perform as expected the developers say the network needs to provide more bandwidth and the network people say the application code isn’t optimized for running over the network. It’s always been both sides blaming the other and the people affected are the users of the application who are subjugated to a lower quality of experience.

Then I realized that while users have expectations of experience, so do developers. Developers are under tight deadlines as companies look to be more agile and more distributed in a global economy. Basic features that would optimize code receive lower priority.  The assumption is these requirements will be handled elsewhere. The “elsewhere” could be the web server or web client, most have caching built into them, but the network is best positioned to support and provide these services to developers, if they will plan accordingly and work with the network team. There’s a convergence of applications and the network. The architects must specify configuration parameters that indicate to developers and provisioners when these services are activated. They may also specify best-practice formatting conventions that have to be observed. This will ensure the application delivery services are available in the network and are being used reducing the time to development and improving the quality of experience.

Posted by John Gaudin at 02:11PM PST

JohnG…sends

—
Enterprise Architecture, Sr. Manager
Cisco Systems, inc.

Are Transparent Network-based Services Used in Application Development?

Today’s infrastructure architecture is becoming more sophisticated and regarded by many as one of the main pillars of information technology. The IT infrastructure consists of the foundational building blocks on which applications and business processes run, it provides generic services that can be used by multiple applications. The network is a key infrastructure element that provides such services called network-based services. In other circles they may also be called infrastructure and/or SOI services. Network-based services may be decomposed further to atomic or composite, in many cases it’s just a matter of taxonomy. For example; “application acceleration” is a composite service that may be comprised of “cache, compression, protocol optimization and content-based routing”, which may be further decomposed into very specific functions such as static or dynamic cache. Transparent network-based services require no direct interaction with an application but enable functionality for the application. Just because these services exist, doesn’t mean they’re going to be used. Consider not only the before mentioned acceleration services, but also security services such as; encryption, day 0 mitigation, intrusion detention and prevention, and anomaly detection. Think of communication services like multimedia bridging, session control, session records, and topology management. There are also virtualization services that do load balancing, VLAN, VPN, and VSAN. These examples just start to demonstrate the capabilities provided by the network.

The challenge is while application developers may know these services exist, they still develop the same functionality into their applications because the governance is not in place nor is the communication (or dare I say collaboration) between the developers and network administrators. Looking at application acceleration- headers must be properly formed and the network made aware for the service to properly operate. Network-based services are becoming more sophisticated and as applications continue to be highly distributed, they will require a library of centralized and standardized services to ensure compatibility with each other. In the end a solid enterprise architecture practice is needed that documents a catalog of network-based services and a change in culture that brings together the application and infrastructure teams to best execute the IT strategy that delivers the business vision.

Posted by John Gaudin at 12:35AM PST

JohnG…sends

—
Enterprise Architecture, Sr. Manager
Cisco Systems, inc.

Network-based Services as Architecture Artifacts

Good Enterprise Architecture helps ensure that business strategy and IT investments are aligned.  IT resources that are documented and/or modeled to contribute to an architectural description become artifacts.  As organizations expand and affiliate, their enterprise architecture becomes more complex.  Forward-looking enterprises want architectures that will meet the current and future needs of their organization with a catalog of artifacts that are agile and adaptable in supporting business capabilities.

As technology continues to grow new resources for architecture artifacts are becoming available on what seems a daily or even hourly basis and some come from unexpected sources. The network has gradually become an important provider of architectural services. Classic transport networks have been optimized to increase throughput, availability, and configurability and are increasingly application-aware. But many enterprise architects still think of the network layer as infrastructure for transport and remain unaware of the new application-related, value-add services that today’s network can provide.

When I think of network-based services that are application centric, I think of two types.  The first type are exposed services, or those services with a well-defined and well-documented set of API’s that create new functionality in applications through a request/response method.  Examples include presence and location information.  Information the network is able and ready to give in real-time that an application can then use to streamline a business process.  The second type are transparent services, or those services that don’t require an explicit call from an application, but are implicit through best practices and configuration options.  Examples include application acceleration and virtualization services.  Services that in a properly architected enterprise perform as expected.

By properly cataloging and documenting network-based services, an enterprise architect will have new artifacts to simplify and standardize many of the capabilities that need to be provided by IT to support business processes.  Explore the network-based services that are offered today.  You may be surprised at what Cisco and the network have to offer.

Posted by John Gaudin at 05:20PM PST

JohnG…sends

—
Enterprise Architecture, Sr. Manager
Cisco Systems, inc.

Wanted: A Collaboration Architecture for the Enterprise

You would never think we find ourselves in the middle of a downturn if you hear the enthusiasm with which people promote and discuss everything related to collaboration in the enterprise. If you have doubts about it, check out the recap of last week’s Voiceon in Orlando here.

There you have it: collaboration is en vogue. Only 2 years ago, the vision was Unified Communications (UC), now it is all about collaboration architecture. And moreover, collaboration is one of those topics that everybody can relate to – at this point in time, we have all enjoyed a variety of technologies and tools that allow us to reach out to people we work with in better, richer ways. We have been visually vowed by TelePresence, we have connected with people we know on social networks, we have contributed to wikis, we have written blogs,  we have laughed at the latest youtube video a colleague distributed, we have joined web conferences, we have shared online workspaces – you name it. So the collaboration discussion is often about what worked and what didn’t work for us, and about consuming the next collaboration service.

And there is merit in discussing the human communication aspect of collaboration – of course it is a key part of the discussion. But let us not kid ourselves – enterprises are not looking into collaboration because they want employees to have more fun and richer experiences. You can rest assured CEOs, CFOs and CIOs are sitting together in one room thoughtfully scratching their chins, going “Look here, if I align all this stuff optimally with my business processes, I can squeeze massive productivity gains out of my resources (aka employees)”. Because yes, by extending the reach of the enterprise with innovative tools employees can more effectively reach other employees, customers or partners and drive business processes forward from anywhere, and do so much faster.

Read More.

Posted by Paul Liesenberg at 09:27AM PST

Is QoS reemerging from the dead?

Next week at Voicecon, I will be presenting in a track about getting enterprise networks ready for voice and video communications. Which prompts 2 thoughts: First of all, I started my career as a software engineer for Class 5 voice switches, and when I made the switch to data communications I made a career out of talking about QoS (for ATM, I am aging myself now) when it came to those emerging network supporting voice services. And the second thought is that, instead of feeling finally vindicated about the fact that finally and indisputably QoS matters, I feel like we are now facing a whole new challenge: to make QoS seamless. Because that is the only way it will become relevant.
I’ll say something controversial: the QoS naysayers were right all along. Why? We do have very powerful end to end tools to offer QoS across the network. We have RSVP, differentiated services, we have CAC (call admission control), and we have the signaling protocols to invoke all of these. So when it comes to insular network thinking, we have had the QoS problem solved for a long time, end to end across the network.
What I like to call the horizontal QoS dimension.

Read More.

Posted by Paul Liesenberg at 07:58AM PST

My First Virtual Trade Show

In light of today’s restricted travel budgets it’s no surpirse that there’s an increase in the use of virtual tradeshows.  On the 11th of March I managed my first virtual tradeshow booth for Cisco- an enterprise architecture show hosted by IDG, Infoworld and Computerworld.  As best as I can tell the show is free to attend by simply registering and supplying some professional information.  With that you’re able to view the virtual booths (above is an image of the Cisco booth), view collateral from each of the vendors, download collateral, and view the web-based presentations of the main hall speakers.

At the Cisco booth we offered sections on:

• Cloud
• Collaboration
• Security
• Service Oriented Architecture (SOA)

If you’re interested in taking a look at the Cisco booth and viewing the collateral click the following link to be taken to the landing page for the Enterprise Architecture Virtual Conference

I’d appreciate your feedback on the booth, the collateral, and the videos.  If you’re interested in my thoughts of the virtual tradeshow then read the full blog.

Read More.

Posted by John Gaudin at 11:27PM PST

JohnG…sends

—
Enterprise Architecture, Sr. Manager
Cisco Systems, inc.

What happens when your medical records are available for sale at Wal-Mart?

I wanted to move beyond cloud computing and yet the news keeps bringing us back to this subject.  Perhaps it is because the golden promise of lean IT with the side of business agility and responsive remains too enticing to pass.  Perhaps it is simply that such a capability has been so ingrained in our personal applications that we are just as eager to bring it to our corporate lives as well.  This week it was identified that Google has another leak that allowed a segment of the documents shared on its site to be seen by those without rights or permissions.  While my shopping list that I have on Google probably does not need Top Secret clearance and thus does not frighten me if lost; my health records that Google is offering to store for me might.  But as much as I want to dwell on this subject, something far scarier is here and the implications are for all of us if we engage in the digital age or not.

Today an announcement came out that Sam Club (under the corporate brand of Wal-Mart) will now be selling electronic healthcare record systems with clinics of any size.  Truly we are hitting a level of mainstream when the store that sells my family household supplies and toiletries is going to sell an electronic healthcare system a mere three times the cost of a 65" plasma TV.  As an avid believer in electronic conversions and the power of a digital age, I also fear a transition that happens to an industry not yet ready.  A private clinic with a handful of doctors while excited to adopt technology, do they have the knowledge base to provide intrusion prevention and firewalls to block outside hackers.  This package from Wal-Mart consists of tablets; will these clinics understand wireless security and enable it correctly? 

Read More.

Posted by Matthew Stein at 08:46AM PST

The new face of healthcare

There is no doubt that Healthcare is changing. Providers are responding to an ever increasing demand and changing demographics resulting in new innovative forms of Healthcare delivery. A changing emphasis on prevention rather than cure as well as the move to delivering care closer to the home means the boundary between the Health and Social Care environments is continually blurring. Combined with the challenges of an ageing population and the differing needs and demands of a new generation, the approach to Healthcare delivery must respond effectively.

All of the above means there has never been a stronger link between technology and the business of administering healthcare. Aside from general access to care records and medical images, innovation through the use of collaboration, digital media and networked medical devices supports objectives such as partnering across the wider health community, health awareness and promotion and virtualised clinical resources. The use of location technology offers efficiencies on behalf of the patient by releasing more time to care. Hence, healthcare delivery is increasingly dependent on technology.

Read More.

Posted by Mike Badham at 08:21AM PST

How different is public vs private?

As I interviewed James Urquhart last week diving into definitions of cloud computing and then compare that with the interview this week with Craig Huegen who is director of Cisco’s Data Center I am stuck by the similarity in concepts between public and private clouds.  At one point there was a significant distinction between on-premise and off-premise applications and the infrastructure that was needed to support each one.  Thus the delta between an off-site cloud based service rented by a company as compared to information kept in-house.  Yet now with more companies moving their data centers to remote locations to take advantage of inexpensive and plentiful land and power, how much different is a public and a private cloud?

Read More.

Posted by Matthew Stein at 09:47AM PST

We felt the same pain as you.

As the success of the Internet took hold, the IT infrastructure teams in many enterprises struggled with managing the growth of data center resources in an organized fashion.  Older architecture paradigms and models certainly didn’t help, as architectures were generally built in one direction, from business process architecture to application architecture to infrastructure architecture, resulting in a significant amount of custom infrastructure deployed for a wide variety of applications.

Cisco IT certainly wasn’t immune to this phenomenon.  Despite identifying the need for an infrastructure architecture virtual team and driving principles and standards, the architecture was still too opportunistic—individual application platforms were deployed on common infrastructure deployment if a standard closely matched, but in many cases the deployment architectures were tweaked enough to create a lack of consistency.

In addition, Cisco’s continued strategy of moving into adjacent markets saw a need to support new business models within IT systems and architectures:  the acquisitions of Linksys and WebEx brought some significant change to Cisco’s traditional IT focus of supporting an innovative manufacturing company.

Read More.

Posted by Craig Huegen at 03:17PM PST

What happens when clouds rain?

Earlier this week Google experienced another embarrassing outage, this time a crash caused their email program to be unavailable to an unknown number (although apparently extensive percentage) of users.  This brings up an interesting question, what happens when the cloud goes down?

It has become a common belief that while internal applications and especially those that live on personal computers occasionally go down or crash, those web based applications that live off premise never fail.  We saw one of the most startling wake-up calls (no pun intended) come when Skype went down not for a few hours but for almost a week in August of 2007.  Small companies that leveraged Skype for their primary phone system for internal calls had the choice of trying to quickly sign up for some type of internal phone service or close their doors.  Now as we live in a world where we don’t just host services but now critical data as well, an outage truly can cause a business to halt business until it returns.Clearly the technology used to guard and connect to these clouds is taking on a new level of importance, as well as the companies that do the hosting. 

Maybe this gives us pause before we jump full fledge into external clouds to consider the potential when we host data off-site.  Who is the company, and how do they build the infrastructure to support their service.

Posted by Matthew Stein at 04:34PM PST

Do Clouds really reducing challenges to IT or make new ones?

An interesting discussion ensued on Twitter one weekend between myself and George Reese of Valtira.  George—who late last year posted some thought provoking thoughts on O’Reilly Broadcast about cloud security, and has recently completed a book on cloud computing—argued strongly that the benefits gained from moving to the cloud outweighed any additional costs that may ensue.  In fact, in one tweet he noted:

“IT is a barrier to getting things done for most businesses; the Cloud reduces or eliminates that barrier.”

I reacted strongly to that statement; I don’t buy that IT is that bad in all cases (though some certainly is), nor do I buy that simply eliminating a barrier to getting something done makes it worth while.  Besides, the barrier being removed isn’t strictly financial, it is corporate IT policy.  I can build a kick butt home entertainment system for my house for $50,000; that doesn’t mean it’s the right thing to do.

Read More.

Posted by James Urquhart at 10:45AM PST

Reducing retail operational expenses does not need to be painful

Store payroll is the #1 controllable expense that all retailers have as a lever to control operational cost management.  However for store management, it’s not as easy as just sending people home to lower costs ("hey there is a great ballgame going on, do you want to help the store out and head home")... there are consequences.  There is a limit to reducing store staffing, and trust me you automatically know when a store is too lean on the floor coverage ("Hello... is there anyone around to help me?")  Been there, experienced that.

There is a way to do more with less. The key is to stretch out your existing staff so they can cover more areas in the store without physically standing in that department.  This could range from having service assistance buttons in the department that send a request directly to the employee or creating a "collaboration session" between the customer and a remote expert. In any case you can dramatically reduce the floor staffing needs (and reduce payroll) without scarifying the customer’s experience.

Read More.

Posted by Curtis Foster at 05:50PM PST

Confusing Cause and Coincidence

A cause is something that makes something else happen.  Out of two events, it is the event that happens first.  To determine the cause, ask “Why did it happen?”.

An effect is what happens as a result of a cause.  Of two related events, it is the event that happens second.  To determine the effect, ask “What happened?”.

One can logically deduce an effect based on the cause and can describe it using connecting words in the following format “cause connecting word effect”.  Examples of connecting words are:

An example of a cause/effect statement: “It is raining therefore the ground is wet.”.  However, one cannot flip the order of the cause/effect statement and have it still make sense “The ground is wet therefore it is raining” may not be true.  The ground could be wet for a number of reasons, it could have rained earlier, but not currently, sprinklers could be running, a fire hydrant could be open, somebody could have spilt a glass of water, or the water table may be rising.  Any number of things could cause the ground to be wet. 

As I was reading a blog from a colleague Matthew Stein titled Does the customer really know best? I was reminded of how often we confuse cause with coincidence and how difficult our job is to properly model cause/effect scenarios and how difficult it is to determine the right cause based on an effect.

Let’s examine in more detail how difficult it is to determine cause and effect and how easy it is to confuse cause with coincidence…

Read More.

Posted by John Gaudin at 09:00PM PST

JohnG…sends

—
Enterprise Architecture, Sr. Manager
Cisco Systems, inc.