This report series documents the results of a Cisco IP video infrastructure, applications, and data center test. Earlier this year, following months of talks with Cisco Systems Inc. (Nasdaq: CSCO), Light Reading commissioned the European Advanced Networking Test Center AG (EANTC) to conduct an independent test of a premium network solution to facilitate advanced IP video services for service providers, enterprises, and broadcasters alike.
It's an understatement, but we'll say it anyway: Carriers are concerned whether their service delivery networks will scale to the forecast exponential growth of video services.
In today’s competitive environment, both telcos and cable companies see the future as one IP-based network that helps to provide more flexible services. But those networks, however operationally complex, need to be manageable. And these companies need to support advanced IP video services, specifically meeting broadcaster, consumer, and enterprise requirements for HD video distribution.
So it's no wonder service providers are so tough on telecom equipment makers. And it's no wonder that some telecom equipment makers are choosing to lead the discussion on IP video, rather than waiting for their customers to ask for tomorrow's network today.
The zettabyte diet Cisco Systems Inc. (Nasdaq: CSCO) has been spreading the zettabyte-era gospel since mid 2008. If one network-based application could really drive traffic towards the zetta mark (1 sextillion bytes), IP video would be everyone’s first bet. And these days the term “IP video” has moved on from a basic walled-garden IPTV idea to a fairly open concept.
Cisco's Medianets
As it has shown in the past, Cisco understands the implications of bandwidth explosion both in the residential and in the enterprise world. In the latter, Cisco is already considered a fixture but it has not been resting on its laurels. Cisco has been busy adding enterprise-focused applications such as telepresence, video surveillance over IP, and digital signage to its arsenal. Cisco has described its IP, media-aware infrastructure initiative using the term "medianet" since December 2008. The company’s CEO, John Chambers, put his full force behind the story. (See Cisco's Video Blitz and How Cisco Does It.)
So Cisco has a point in analyzing the market and coming up with a marketing strategy that matches the needs of service providers. But how much of the story can it actually deliver today? Was there a chance Cisco would be ready to submit equipment and support for a test of Medianet?
It’s OfficialThis test was big. Mega, even.
The Megatest
Actually, Cisco welcomed the idea – quickly realizing, however, that this effort would be monumental compared to any equipment test ever published in a magazine in our business before. A six-month program was installed; fifteen business units got involved; Cisco’s senior vice president for central development organization backed the project; and it quickly adopted the name “Megatest” until the team converged on the more realistic name of “Über Megatest."
In an unprecedented testing campaign, European Advanced Networking Test Center AG (EANTC) was commissioned by Light Reading to investigate all of a medianet’s aspects: We started from Cisco’s network infrastructure supporting enterprise and residential customers as well as a purpose-built network for broadcasters. We then moved on to the comfort of the end-user’s home and looked at the applications Cisco says it can bring to the market. And we closed with a focused investigation into Cisco’s service provider data center solutions, which we will cover in a separate special report next week.
The tests covered the following main areas:
High availability with sub-second failover time for all network services
In-line video quality monitoring
Massive scalability of IP video services
Storage area network solutions and virtualization
In addition, we received extensive demonstrations from Cisco showing both residential and enterprise applications. These will be documented in the next article in this series, coming later this week.
Results snapshot
In short, Cisco’s medianet solution showed excellent results:
8,188 multicast groups were replicated across 240 egress ports in a point of presence (PoP), showing that Cisco could serve 1.96 million IP video subscribers in a single metro PoP.
Accurate in-line video monitoring was demonstrated for video distribution and contribution over IP.
Sub-50 millisecond failover and recovery times were shown for video distribution and secondary distribution networks using, for the first time in a public test of Cisco equipment, point-to-multipoint RSVP-TE.
No video quality degradation in the face of realistic packet loss in the network
Excellent quality of service (QoS) enforcement in Cisco’s new ASR 9010 router for both fabric oversubscription and head-of-line blocking
Hitless control plane failover for converged network
In the world’s first test of this breadth and width, Light Reading and EANTC have found that IP infrastructure from Cisco really does deliver. This is the first in a three-part report series wherein Light Reading and EANTC will share these groundbreaking results.
In this report, we cover Cisco's medianet service delivery network. Here's a hyperlinked table of contents:
— Carsten Rossenhövel is Managing Director of the European Advanced Networking Test Center AG (EANTC), an independent test lab in Berlin. EANTC offers vendor-neutral network test facilities for manufacturers, service providers, and enterprises. He heads EANTC's manufacturer testing, certification group, and interoperability test events. Carsten has over 15 years of experience in data networks and testing. His areas of expertise include Multiprotocol Label Switching (MPLS), Carrier Ethernet, Triple Play, and Mobile Backhaul.
— Jambi Ganbar is a Project Manager at EANTC. He is responsible for the execution of projects in the areas of Triple Play, Carrier Ethernet, Mobile Backhaul, and EANTC's interoperability events.
Prior to EANTC, Jambi worked as a network engineer for MCI's vBNS and on the research staff of caida.org.
— Jonathan Morin is a Senior Test Engineer at EANTC, focusing both on proof-of-concept and interop test scenarios involving core and aggregation technologies. Jonathan previously worked for the UNH-IOL.
Your assessment of the "test" seems pretty accurate. If the purpose were to validate anything, the test clearly did not deliver. So this was just a marketing spiel intended to create the perception of a validated solution in which the new Nexus and ASR found a home. I guess it's the next best thing to do if you don't have any meaningful deployment references to speak off >:)
Agreed, competition needs to be encouraged. However, I can see it more like a challenge/virtual RFP, where pre-set transport conditions (ie up/down bandwidth and QoS goals) are to be met by vendors. Alternative designs can be evaluated based on cost and energy footprint.
PS. Frankly, I am still puzzled at the purpose of EANTC testing of Cisco Service Delivery - it looks like a giant and complex network with no particular purpose - no performance goals, no energy goals, no management/opex goals etc. A lot of effort with limited benefit to the end users.
It would be nice to see other big vendors such as Juniper or Alcatel step up to the plate, organize a nice big test, and demonstrate to the lightreading community what they can do in this space as well, because they offer (or in the context of testing "claim to offer") very legitimate and competitive high-end transport solutions for IP video of all kinds.
Network-level efficiency is not necessarily equal to sum of it's components. Lots of lean boxes can actually draw more power that fewer complex devices.
There is some work on the network-level efficiency rating (NCR) being currently scoped. However, such measure would probably benefit service providers, not equipment manufacturers.
The obvious case in point would be the "Cisco's IP Video Service Delivery Network" itself.
With due respect to Cisco's effort, this test is clearly sponsored by them and features virtually each and every product Cisco sells in the carrier space, as well as 3(!) different operating systems from Cisco. Such exotic setup cannot be energy efficient as it multiplies layers and support infrastructure way beyond the necessity.
In the network design case that would look more like a real thing and less like a dealer's showroom, there would be fewer layers and devices; new products (read ASR9000) would take the lead and energy hogs (read 7600/6500) will not be even qualified.
It would be very interesting to see a challenge like this (a network with fixed functionality and minimum bandwidth requirements) answered by multiple vendors (say, ALU, JNPR and Cisco) and independent integrators who can mix-and-match products for best efficiency; notwithstanding the obvious political issues around such a test, it could certainaly evangelize an ICT future that is both greener and more practical than the usual single-vendor dog-and-pony show.
Your note is appreciated and certainly a good idea for future tests.
Serious tests of energy efficiency require a controlled environment and a standardized efficiency metric. The Alliance for Telecommunications Industry Solutions (ATIS) has defined environmental guidelines and test methodology in ATIS-0600015.2009. It defines a Telecommunications Energy Efficiency Ratio (TEER) which is a handy metric to compare energy efficiency. However measuring TEER comes with some rigid methodology. The bottom line is that one cannot simply add a power meter to the test because that would not prove anything :-)
The focus of this test was different (video aware network services) so we did not strive to control the environment and run the ATIS set of tests, which require different hardware configurations to some extent (like a chassis fully loaded with a specific type of line card).
Other than that, the biggest question from my point of view is how one would measure end-to-end service efficiency. State of the art is just to measure and compare a single device's efficiency -- but what if one vendor requires a lot of (lean) boxes to provide an end-to-end solution, whereas another vendor can do the same with one (heavy) system?
We did. Light Reading paid EANTC to perform the test.
Of course, because of the amount of gear involved and the cost of simulating this kind of network, Cisco had to agree to be tested. They signed off on some things, but not on others, as the test report states.
But EANTC worked for us on the deal and we maintained editorial control of the tests that were approved throughout the reporting process.
Nice Test! I have a question regarding the "Green FootPrint" of these large scale tests. What would be the power usage for both the network and test gear? Any plans to add the green factor in future large tests?
The blogs and comments are the opinions only of the writers and do not reflect the views of Light Reading. They are no substitute for your own research and should not be relied upon for trading or any other purpose.
To save this item to your list of favorite Light Reading content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
This test was big. Mega, even.
Digg
Del.icio.us
Reddit
