Participants in this podcast include Prakash Sinha of Cisco Systems (PS) and Gian Trotta (GT).

GT: Welcome to another ebizQ First Look Podcast. I’m your host, ebizQ’s Gian Trotta. Web 2.0 apps can be bane or boom to enterprises that are seeking to boost agility and revenues. Done right, Web 2.0 apps can open new lines of business and new lines of communication to clients and customers. But, if you do it wrong, you’ll sow chaos inside, outside and throughout your enterprise. Cisco product manager Prakash Sinha is here today to follow up on some of the strategies he shared in the very popular webinar entitled: “How the Internet’s Largest Bookstore Mashes SOA and Web 2.0”. Welcome, Prakash, thanks for joining us.

PS: Thanks, Gian, happy to be here.

GT: Prakash, you’re on the record as stating that the network is the logical place to develop Web 2.0 applications and you’ve got some impressive examples from past history to back you up. Can you elaborate?

PS: Sure, if you step back and look at how the network, especially on the layer 4 through 7 was born, there are two dates that come to mind and stick to me. One was April of 1996, that’s when Netscape was born, went IPO and started this incline to revolution. They also created something that we take for granted now which is at the cell.

Another date that brings to mind is 2004 when Google went IPO. So Google completely changed the way, you know, the user facing applications are now being written. They introduced a technology called Ajax and if you look at what these technologies did was take all the applications from the data center and web enabled them and now, with the Google revolution, it’s actually creating these rich Internet applications based on these Ajax and JavaScript object innovation technologies.

As well as a mix of these technologies that are also coming together called mashups or remix applications. What they did was they created a lot of traffic on the network and because of that, to address the security implications that these implied, these type of technologies implied, as well as some of the performance implications on the user facing side and, you know, all of these infrastructure needed to be scaled and be secured and with remix and mashup applications, a mix of technologies are on the user side trying to consume, not just social network type of interaction but also enterprise web services.

So because of that, layer 4 through layer 7 is changing, the network is changing, becoming more application aware and application fluent so that’s the change that is going on in the network and definitely if you look at all the applications that are in the data center, of course you could go in and scale this infrastructure by replicating some of you data center applications but are not ---. And then you have to go into each one of these applications and change if you needed to secure them, each one of the applications need to be secured, right, now a network is one point that is standards based touches all of these applications so what better place and what better way to secure all of the applications by creating these security policies, scalability, addressing scalability issues in the network.

GT: OK, but I think Prakash, we’re seeing proof of this concept maybe at this time of the year, a few days before the holidays in online shopping sites that combine database access with Web 2.0 user reviews and outside functionalities like shipping charge calculators. Do you have any other case studies where this chaos and this potential has been managed well?

PS: Oh definitely and so, you know, when we actually had the webinar way back in October, we actually discussed a use case from one of our customers which is the largest bookstore in the world and they had these fulfillment channels that they wanted to reuse. Now we already know that, you know, that service oriented architecture is already going mainstream, it’s been about 6,7 years when all of the web-based service oriented architecture including Web services started taking off, so, you know, in our client and in our customers we see that they’re already adopting these technologies, mainly driven by reused and productivity implications as well as flexibility of IT infrastructure.

So, in case of, you know, the largest bookstore in the world, they actually have fulfillment channels they wanted to reuse that shared this cost of this infrastructure across their partners. Now, they have a very big competitor, a very well known competitor based out of the West Coast and it took their competitor about 6 years to actually create this infrastructure where they were creating new storefronts on their infrastructure.

The key driver was to reduce the cost of doing business, right, so in this case, you know, the internet bookstore actually started creating partner storefronts and they had a lot of challenges because they couldn’t control the security for their partners, they couldn’t control, you know, what kind of security models they used, couldn’t control what kind of technologies they used on the user side so what they did was they used some of our technologies to actually bridge or mediate between what was coming in through the web and what their back end systems could support and this is a great example of how to use IT for a flexible architecture, flexible infrastructure and still grow business by creating these new channels of revenue so their partner storefronts have a look and feel of their partner site and not the bookstore site which is creating or delivering the fulfillment for the user queries and user requests dor books and for other items like toys and other types of items like bikes, for example.

GT: Right, I think the one example was, one thing that came out of the partner innovation was one-click ordering that it was a case of where to facilitate, a feature to facilitate interaction with partners later succeeded as a consumer feature when exposed, it actually, I think, proved a great success but I’m wondering, can something like that really be architected far back into the business and development process?

PS: So, definitely, it can be. In fact, architecture is a key component of how you design how the user interacts with your systems, right. So, you know, one of the key, and one-click is a great example of that because it reduces the amount of clicks that a user does before they can order something so if you’re a user that has already used a storefront before, you can store some of the key properties of what the user likes and dislikes, you know and based on who the user is, you can customize the experience that the user gets and because of that, there’s less issues with hanging cards, people not, or if you’re addressing some of these performance implications that these rich clients bring on the network, you’ve already taken a step where you won’t have some of these lost businesses because user experience wasn’t good.

So, the key driver behind this is make it as easy as possible for the customer to order when they’re navigating a website but also reduce the latency of this interaction with the back end system and that’s one of the key things that you should actually look for when you’re architecting an infrastructure for user facing.

GT: Right, that’s a good subject and I hope that we can bring you back again in the future to talk about it but I wanted to return to another concept you’ve developed that’s equally interesting. It’s called, and if you could quickly summarize it, it’s called “5 Steps to Network Nirvana”. Would you be able to elaborate on that quickly for our audience?

PS: Definitely. So we actually talked about some of the key areas to look for when you’re creating an infrastructure that a Web 2.0 or user-rich internet user application is trying to deliver, right, and some of the key drivers at least from a business perspective and that is creating the infrastructure is, you know, how to deliver reusable and flexible infrastructure that can be used in many different ways and the key driver from a cost perspective is a better server utilization so the key things that we see from a network perspective for these Web 2.0 and SOA type of applications is how to deliver a flexible architecture, how to enforce consistent policies, how to deliver a simpler architecture because, as you know, most of these applications that are trying to access these services are simple clients or thin clients so they don’t have a lot of logic inside of them so definitely simpler architecture is the key area.

Then definitely because you’re trying to reduce costs, you want to improve the server utilization therefore less servers and then, of course, management is a key issue and you’ll hear a lot of things about governance for the new applications, the faster development is something that is, faster development and delivery of these new services because you only have so much time and you cannot quote everything into you’re applications so being able to deliver or increase the adoption rate by simplifying these application development, by creating policy-based offload is a key area.

So, if I were to summarize all of these points, the key things that from a network perspective I look for is how to virtualize service endpoint, how to secure these endpoints and how to deliver the scalable and highly reusable services, how to offload because reuse is the key goal and utilization and improving utilization is a key goal, how to accelerate and offload this processing because you want to reduce the latency for the user and then of course, management is a key goal as well so how to, you know, create some policy driven IT infrastructure so that it’s easier to manage.

GT: Right, I think that’s understood and I think that’s a sustainable and attainable plan.

PS: Absolutely.

GT: OK. Prakash, I want to thank you for a very informative and very actionable presentation. Where could listeners go on the Cisco site for more information?

PS: So, we have a website on Cisco.com, you can go to www.cisco.com/go/ace and we have a lot of white papers, thought leadership articles and product information so if you are interested, definitely, I would welcome you actually taking a look at our site.

GT: OK, great. Prakash, one last question before I go out and do some Holiday shopping both in person and online. What improvements are you looking forward to for the next selling season in e-commerce?

PS: So, definitely, what I’m looking for, I know for sure that this Web 2.0 rich Internet applications are becoming and will become much, much more pervasive. Service orientation is already, you know, going mainstream in the enterprise site. So, I would look for more technologies, much more secure product. Definitely, two key things I would look for at least from a network perspective is just as we saw with the advent of the thin client with Netscape, these rich clients have a lot of implications on the business side and the IT infrastructure side so what I’m looking for, you know from a network perspective two things: one is being able to secure, there’s lot of mandates coming in, especially for retail delivery. There you have to secure credit card transactions and things of that sort. I would seek to look for much more secure environment and the other area is being able to improve the utilization of servers by virtualization, being able to offload some of these compute intensive tasks that really belong in the network and away from applications and you know, being able to do this using some policy-based mechanisms.

GT: Right, that’s understood. I’d like to add for our listeners who have similar questions for Prakash that he’ll gladly answer any of them that you might post to do so, you just have to visit the landing page of this broadcast which is located at www.ebizQ.net/firstlook and for more cutting edge podcasts, webinars, virtual conferences, white papers, polls and more, the address, as always is www.ebizQ.net. I’m Gian Trotta, thanking you for your time and thanking our guest, Prakash Sinha of Cisco Systems again. Thank you, Prakash.

PS: Thank you.