Reference post below by Shamus McGillicuddy
It still boggles my mind why there is such a fascination with large bridged networks rather than relying on the proven ability of IP to manage path selection. Spanning Tree doesn’t have the features to ensure optimal path selection. Maybe it’s that the data center is often designed by people with a strong background in computers rather than by network engineers. I’ve seen many cases where data centers have traffic going over the wrong path causing congestion because they can’t get Spanning Tree to place it on a more optimal path. Then add this trend to run layer 2 over the WAN with VPLS. Sure, you don’t have to deal with IP addressing and route distribution, but the tradeoff is a large geographically separated collision domain with little control over path selection and less ability to troubleshoot and monitor it. IP routing is a solution that shouldn’t be overlooked. It was designed specifically for this reason, and it’s easier to spell. SDN may prove to be a great solution, but it’s too young yet.
Excellent insight. New technologies and methods will provide more challenges for network security. That’s job security if you can keep pace.
While 802.11ac may be of interest to those looking to enable laptop and mobile users high speed access, that’s just at the access tier of the LAN. SDN has more potential to change the architecture dramatically, and that not withstanding adequate means to measure performance and monitor security in that environment.
Yes, visibility into the cloud has to take a more prominent role. That will require innovative approaches. Are the three big NMS providers able to move fast enough to address this need? I’m looking to startups for the new approaches. And what of Open Source products, which have come a long way? Why invest 3/4 million into COTS and then not develop the customizations and integration to make it do everything you need in your environment? A better approach is to use Open Source and invest the money saved into human resources to configure and integrate the tools – the added benefit is a top-notch support team to keep it in pace with the network changes.
Added complexity has its costs. Measuring the performance of a dynamically changing topology, the performance of the SDN system itself, and added complexity in network security are just a few challenges. Software-Defined Networking certainly has potential, but I’m still waiting to see if this can realize a ROI and performance improvement given the additional complexity. I don’t think everyone is ready to jump on this bandwagon just yet.
Original is reposted below:
What does 2013 have in store for the networking industry? We asked five top industry analysts to predict on networking trends for this year. Click on the links below to find out what will happen in data center networking, network security, campus LANs, network management and software-defined networking.
Data center networks will continue to wrestle with the limitations of spanning tree protocol in 2013, but enterprises that move to alternatives like network fabrics will find roadblocks to scalability. Meanwhile, enterprises will use Ethernet exchanges to build hybrid cloud environments and cutting edge micro-electromechanical systems (MEMS)-based photonic switches will start to make some noise in the data center. Erica Hanselman, research director at London-based 451 Research, shares his predictions for how the data center networking industry will shake out in 2013.
In 2013, network security vendors need to develop third-party ecosystems that help enterprises correlate data among the various components of their security architecture. Also, network security pros will need to sort through the software-defined networking (SDN) hype to figure out how secure these new technologies are. Meanwhile, enterprises will accelerate their adoption of next-generation firewalls and advanced threat protection systems. We asked Greg Young, research vice president at Stamford, Conn.-based Gartner
Inc., to share his views on the changes we’ll see in network security this year.
Campus networking has lacked innovation for a few years, but 2013 may switch things up a bit. While wireless LAN vendors will be pushing faster 802.11ac networks this year, the industry may also see some architectural changes that could finally deliver true unified wireless and wired campus LANs. We asked Andre Kindness, senior analyst at Forrester Research, to share his views on
the changes we’ll see in campus LANs this year.
Emerging virtual overlay network technology will force network management vendors to develop tools to monitor these new environments in 2013. Meanwhile, enterprises will demand better visibility into their public cloud resources and virtual desktop infrastructure deployments. Enterprise Management Associates Research Director Jim Frey shares these and other predictions for
how the network management market will evolve this year.
What’s in store for software-defined networking? IDC analyst Brad Casemore predicts adoption will grow among service providers and cloud providers; vendors will battle each other in Layer 4-7 network services and SDN controllers; and OpenFlow may evolve, but very slowly. In the longer term, IDC projects that the SDN market will reach $3.7 billion by 2016. Here’s more of what Casemore had to say about the SDN market in 2013.
Change Management is an important function in most organizations. It carries more weight than many of the other ITIL functions because it’s the biggest pain point. It’s a well established fact that upwards of 80% of all outages are self-inflicted. IT managers are constantly getting heat over deployment that didn’t go exactly as planned. When you boil that down to lost productivity or missed business opportunities it amounts to a sizable amount of money. These are just some of the reason Change Management gets so much well deserved attention.
So, you establish a Change Advisory Board. There is a lot of preparation and documentation that has to go into any change before it’s presented to the board for approval. Each change is categorized, analyzed, scrutinized, until everyone involved is thoroughly mesmerized. The time required to get a change approve may also have increased five-fold. The process is controlled through some rather expensive management software, well documented, well planned, and hopefully well executed.
The question is: After expending all this effort into the Change Management process, expending the resources in additional planning and documentation, and spending all the time in meetings, and prolonging the time required to get a task accomplished, did CM reduce service disruptions and save more money than was invested in the process? Let’s face it; if not, then throw the whole thing out and go back to shootin’ from the hip.
The stakeholders on the board probably didn’t review the detailed documentation that has been prepared. There are probably only a few people in the entire organization who will ever read it. The stakeholders only have a few important questions: why is this change necessary, what’s the impact, who or what will be affected, what are the risks, are they adequately mitigated, and is there a viable back-out process. There are probably a few key people in each business unit who could review the implementation details and provide their respective stakeholder with a recommendation and/or list of concerns and remediations.
Is the CAB keeping any metrics? Are you aware of how many changes of each category are being implemented? Were they on schedule? Were the impacts more or less than expected? Is there a way to relate an incidents as a result of a change to the change in your incident reporting system? Is all this management making an improvement, or have you just spent more resources managing with no real gain? When you make a change to the process, does it streamline the process and/or improve the results?
Change Management is good. CM in the context of ITIL framework is excellent … but we must always keep focused on the end objective – becoming more efficient and effective. CM for the sake of CM is a common ill and needs to be tempered with CS (common sense).