re: Cisco Takes Hold of the EdgeI don't know... I think Optodoofus has a point. The ASR by itself isn't going to drive any other boxes to extinction (it doesn't scale very far, for one) -- but it seems obvious Cisco is trying to become the hub that houses all these applications.
Even if that's not a good idea, Cisco's marketing prowess and its position in the network might be enough to get it to happen.
It might not work out in the end, but I'd bet this is going to be a key theme for Cisco at the edge.
re: Cisco Takes Hold of the Edge"The only reason it has not been done this way to date is because these functions have always severely impacted the scale and performance of the edge router. If Cisco can solve this problem (and I'm not saying they can), this would significanlty change the economics for service providers"
It's the god box syndrome again. A box that does "everything" can be a "jack of all trades, master of none". You can get stuck with feature interaction bugs or limitations. New feature XYZ of BGP you'd really get benefit from might not be available with a software release that supports feature(s) ABC of the DPI code that you require. You end up making trade offs that you wouldn't have to make if these functions were in separate boxes. Putting all these features in the same box does save physical room, power supplies, and maybe some sheet metal. But are those savings worth it when compared to the limitations they can create on more operationally important functions you need to run and provide services with your network.
re: Cisco Takes Hold of the EdgeCisco is not a pioneer on adding more functions in an Edge Router, remember Redback (Which belongs to Ericsoon Currently) and Huawei have launched before their Edge Routers which integrate many functions in one box, mainly SBC, Firewall and DPI.
Huawei ME60 and Ericsson SmartEdge 1200 are Edge Routers that are offering this to the service providers NOW. Cisco 7600 has integrated some DPI Functions, Alcatel has a roadmap to add DPI in its Edge Routers 7750, and so on.
re: Cisco Takes Hold of the EdgeWell, the actual stuff was boring, so I diverted myself with other things. Like, on the home page of their ad, just below the Network Uber-user link, there are four tags (About the Event, Register for the Event, Tell a Friend, etc.).
Well, if you roll your mouse over to the extreme left of one of those tags, and position it just right, you hear what appears to be a geiger counter.
It's like one of those "play Stairway to Heaven backwards and you'll hear ..." or "the pixie dust in Lion King spells ..." kind of things.
I don't know what it means, but I'm just a little worried...
re: Cisco Takes Hold of the EdgeIt seems to be the trend to integrate more functions in the Edge of the Network, eventually the service providers will need to add SBC and DPI Functions in their Edge, so with the current technology they will need to decide if to use one or many boxes to do this.
The point is if Huawei and Ericsson is doing this, and now Cisco is also offering iT is because there is a potential market out there that will pay for it.
Now if this boxes can work properly is of course a must question, and the technicians will love to talk a lot of this topic, but we need to also remember that the products must comply with the market not only for the technology. The success of this products will depend very much in the markets demand and if they can keep up with the right performance.
Remember this is happening not only in the Edge, what about the transport network? we are seeing now Packet Transport Networks equipments which integrate optical functions like DWDM and Packet functions like PBT, T-MPLS , MPLS etc in one box. So this issue is not really so new , we are beginning to see it.
re: Cisco Takes Hold of the EdgeWell technically and operationally, if you want to introduce application smarts in an existing routing infrastructure, it still comes down to a introducing a new, separate box. Whether that's an ASR1000 or a stand-alone DPI platform doesn't change anything. Since Cisco positions other products for pure routing jobs, the question is whether the (modest) routing capabilities of the ASR turn out to be overhead or a differentiator with solutions from DPI/SBC specialists, as long as existing/additional routers are still needed.
re: Cisco Takes Hold of the EdgeIt is interesting to see a common pattern of trying to dismiss what is being proposed by the ASR...
For years the Edge was sort of a dilemma for Service Providers, and a lot of time was(is) spent on architectural discussions to try to minimize the compromises of the design / equipment capabilities.
I honestly believe we shouldn't be so quick in dismissing it, just as the CRS proved to be a "Juniper" killer, and it pretty much succeeded in keeping JNPR at bay, we might as well see the ASRs helping Cisco gain precious space in the battle for the edge, after a not so successful run in the Metro space due to entrenched point competition (AlcaLu, Huawei). This completes the picture with Home (Linksys), Unified Comms (enterprise and managed CPEs, ISR, IPT, Telepresence), Core (CRS), WiMax (Navini) and Video (Scientific Atlanta).
Smart move despite all of the "funny marketing"...
re: Cisco Takes Hold of the EdgeWith NX-OS and QFP it seems Cisco is seing a return on their investment when they acquired Procket. I ran across this in Network World (I know, poor source, but hey it made for interesting reading)
"The ESP, which is based on CiscoGÇÖs new QuantumFlow processor, allows services such as network security, deep packet inspection, firewall, quality of service, Network Based Application Recognition, broadband aggregation and session border control to reside in software and not require additional hardware support in the form of a service blade, Cisco says. QuantumFlow was developed with technology obtained from CiscoGÇÖs acquisition of Procket Networks in 2004."
2 cheers to my Procket mates still slogging it out inside of Cisco!
Even if that's not a good idea, Cisco's marketing prowess and its position in the network might be enough to get it to happen.
It might not work out in the end, but I'd bet this is going to be a key theme for Cisco at the edge.