This turned out to be another intensive year (more on that in my next post) and the above pictures capture our team’s presence at Carrier Network Virtualization, the last public event on the Lean NFV Ops program for 2015. I like to thank Christie Rice for her invitation to present at Intel Network Builders Summit in Palo Alto, and couldn’t be happier to hear that my presentation, Communication Networks Reloaded, was well received (see slides below) by the workshop participants. I would also like to thank Sue Morehouse with our Events Team and Andy Mast at the Cloud Innovation Center for all the work that went into getting our team organized for CNV15.
Photo album: https://flic.kr/s/aHskptDZXC
“In the network space there is a big opportunity for growth.” “The communications industry is moving to a cloud based infrastructure […] with standard high volume servers in a virtualized environment running in a dynamic, automated way.” – Diane Bryant, Senior Vice President & General Manager, Data Center Group. Intel Data Center Day. Santa Clara, August 27, 2015
“This is probably the most exciting time to be in network infrastructure.” “It is a market at an inflection point.” “Until now communications infrastructure has been built on purposely built fixed function appliances.” “Network service operators spend roughly $160B a year on equipment […] and run a $2.2T services business.” “Most of that has historically been a voice based business, but with more and more users and more and more devices connecting to the network you see an explosion of data traffic.” “This is creating an upside-down business problem where, sometime in the future, it will cost them more to deploy, operate and maintain those networks than the revenue generated.” “The answer to that problem statement is to adopt many of the principles of data center and IT infrastructure and cloud […] to take advantage of volume server economics and virtualization technologies, and to pool resources and assets across a number of workloads.” “By the way, Alcatel-Lucent is demonstrating [NFV, Networks Functions Virtualization] here this afternoon. You will see pretty sleek demos.” – Sandra Rivera, Vice President Data Center Group, General Manager Network Platforms Group. Intel Data Center Day. Santa Clara, August 27, 2015.
Note: the above is a collage of selected quotes, which have been edited for readability. Listen to the source webcast for the actual talking points and context.
Alcatel-Lucent Team (from left to right): Jose de Francisco, Bob Haddlelton, Ted East.
Last week our team went back to the Silicon Valley to join Intel’s Investors Day focusing on Data Center technologies in Santa Clara. The event’s agenda included presentations, a data center tour and several demonstrations. Here is the link to the webcast covering the executive talks.
The demonstration area showcased Intel’s projects and featured Alcatel-Lucent as a pioneer in the carrier cloud space and a member of Intel’s Network Builders ecosystem.
You can see Ted East running the Lean NFV Ops demo in the above pictures. We both ended up engaged in parallel conversations when addressing questions from different investors gathering at our demo station. By the way, our team was also present at both of Intel’s Network Builder and Partner Summits as well as at the prominent Developer Forum (IDF). All three events were held in San Francisco the previous week:
At last week’s Investors Day our team continued to validate interest in Lean NFV Ops practices, which call for:
- operational effectiveness tied to highly efficient utilization levels
- starting and remaining lean at any scale, this being key to dynamic networks.
As quoted above, the telecommunications industry is shifting gears by adopting general purpose computing in the form of x86 technology. We discussed NFV’s (Network Functions Virtualization) service agility subjected to real life RAS (Reliability, Availability, Serviceability) scenarios, as well as performance expectations and smart load placement in that environment.
We talked about the role of EPA (Enhanced Platform Awareness) and data plane acceleration in particular: DPDK (Data Plane Acceleration Kit) is a packet processing example leveraged by Alcatel-Lucent teams working on sophisticated network functions such as vEPC (Virtual Evolved Packet Core) and Rapport’s vIMS (Virtual Internet Protocol Multimedia Subsystem). Some of the discussions included Intel’s storage compression optimization as well.
Here is a quick outline of the solutions involved in the current version of the Lean NFV Ops’ end-to-end demonstration system:
- OSS (Operations Support System) – Motive Dynamic Operations
- SDN (Software Defined Networking) – Nuage Networks
- MANO (Management & Orchestration) – CloudBand Management System
- vIMS (Virtual IP Multimedia Subsystem) – Rapport
- vEPC (Virtual Enhanced Packet Core) – IP Mobile Core
- ANALYTICS – Bell Labs
- VIM (Virtual Infrastructure Manager) – OpenStack
- NFVI (Network Functions Virtualization Infrastructure) – CloudBand Nodes
- PARTNERS – CloudBand Ecosystem
Last but not least, this was another very rewarding event and there is a need for thanking Bob, Andy for demo support and Intel’s Jon, Lisa, Christie and Jeni for all of the help provided this past week.
Photo Album: Lean NFV Ops at Intel’s Network Builders, Developer Forum and Investor Day.
“In this panel our guests will discuss key issues facing orchestration providers, including the standardization efforts for descriptors and information models, intelligent workload placement options with Intel architecture, and operator priorities for NFV-O and MANO in 2016” – Intel Network Builders Summit. San Francisco, August 17, 1:30 pm.
From left to right: Jose de Francisco, Ted East, Bob Haddlenton, Valerie Noto. Alcatel-Lucent’s Lean NFV Ops Demo Station at IDF15.
The Lean NFV Ops Roadshow is back after a short summer break. Our team would like to thank Intel for inviting us to join Network Builders Summit, Partner Summit, and the prominent Developer Forum (IDF15) this past week in San Francisco. The above pictures were taken at Alcatel-Lucent’s station, which featured a live interactive demonstration. See my earlier post “Gearing for IDF15” for more information on that.
Additionally, you can see a new element: a Galileo powered retro Whack-a-Mole machine, WAM for short : )
Long story short, IDF15 attendees were able to have fun by testing our Lean NFV Ops system when playing this game, which is designed to create catastrophic failures.
On a serious note, the higher the game’s score the more critical the issues experienced. This exemplified zero touch healing and recovery executed in real time to sustain a High Availability (HA) environment. In these pictures you can see Intel’s Diane Bryant, SVP/GM Data Center Group, and Sandra Rivera, VP/GM Network Platforms Groups, putting our demo to the test.
I would also like to thank Renu Navale who moderated “The State of Management & Orchestration”, a panel discussion where I joined Ciena’s Recep Ozdag, Overture’s Scott Vandiver and Rift.io’s Tony Schoener in a fairly engaging discussion.
We discussed opportunities and challenges involving ETSI NFV’s Management & Orchestration (MANO) architecture, standardization efforts across the board, information models and enhanced platform awareness (EPA) among other topics.
There were 50+ people attending this session. Those providing feedback on this session mentioned that topics and viewpoints expressed there were followed with interest. My understanding is that the video will be available in a couple of weeks.
Regarding questions on the underlying fundamentals behind Lean NFV Ops: this link provides a brief paper on the topic just to begin with.
Generally speaking, “Lean” practices couple (a) effective quality management with (b) efficient end-to-end systems. That summarizes Lean’s paradigm across any industry we happen to look into, whether we are talking about manufacturing, services or software for that matter.
Our conversation on Lean NFV Ops has been well received since first presented this at Mobile World Congress earlier in the year. Proven interest has to do with addressing quality by discussing Reliability, Availability and Serviceability (RAS) and Quality of Experience (QoE) in the telecommunications context. These are critical success factors in an industry concerned with a carrier grade tradition facing which is now conflicting views when confronted with with compelling cloud economics.
I created the above chart on Lean NFV Ops to clearly show that we are optimizing for both Service Level (SLA) and Return on Asset (ROA) since High Availability (HA) and High Utilization levels shape dynamic network behaviors. Simply put, Lean NFV Ops is abut effectively delivering timely “service(s) of value” by operating at “resource efficient levels”. Making a return on the Communication Service Provider (CSP) investments (ROI) in NFV depends on achieving that. Otherwise, virtualization critics will keep reasonably doubting the need for any changes.
Equally important, sustaining a competitive advantage means that early implementations can be based on nimble architectures, which remain streamlined (and, therefore, stay lean vs. becoming bloated and over-engineered) as they rapidly change scale and/or scope. With that in mind, embracing Lean NFV Ops’ agile “continuous improvement” and making room for incremental innovation happen to be of the essence.
One more thought, Lean NFV Ops is not about a mechanical one to one implementation of known Lean principles. This conversation is centered on understanding the nature of the telecommunications sector so that we can come down to defining “how-to” guiding principles.
Lean \ˈlēn\ adjective: athletic, strong and healthy, absence of excess.
Circling back to IDF15, I am happy to share that we had good meetings with a number of partners throughout IDF15. Everyone’s feedback on Lean NFV Ops was very encouraging, most openly saying to be impressed by our end-to-end solution approach:
- Operations Support System (OSS) – Motive Dynamic Operations
- Software Defined Networking (SDN) – Nuage Networks
- Sophisticated Virtual Network Functions (VNFs) – Rapport
- Appliances deployed as VNFs – CloudBand Ecosystem
- All running on CloudBand 3.0, the platform discussed in this interview.
Tieto, a CloudBand Ecosystem Member, was also present at IDF’s Network Builders zone. Tieto’s team had onboarded a VNF on our platform and Daniel Nilson was kind enough to share his latest paper.
I would like to take this chance to give Ted East and our Cloud Innovation Center (CIC) team “kudos” to acknowledge the hard work going into all the timely system upgrades, which continue to make the Lean NFV Ops live demonstration an invaluable asset (WAM’s latest retrofit included : ) Last but not least, thanks to Andreas, Asaf, Debbie, Erez, Guy, Ken, Phil and Val for all the couching and guidance when preparing for IDF.
Click above to access this album on Flickr.