Transcript for HP's UC Reference Architectures
Marty Parker: Hello everyone, this is Marty Parker of UCStrategies and today we have a very interesting podcast that I have the privilege of moderating on the topic of Hewlett Packard’s Unified Communication and Collaboration, or UC&C program. And within that, an interesting new addition to that program called UC Reference Architectures. UC Reference Architectures are a very interesting tool, which are going to make the HP UC&C solutions much easier for their systems integration partners to present to customers. It’s going to take a lot of load off the SI partners on the backside. It’s going to make it less costly to introduce and deploy. It’s going to make it quicker to deploy. There are all sorts of benefits.
Alan MacLeod is with us from Hewlett Packard, the director of UC&C Initiatives, as well as Jan Kelly, of UC&C Product Strategy. We’re going to hear from them about the initiative and then we're going to ask them questions as we proceed through the podcast. I’ll just say one more thing before I call on Alan, which is that we’ve been saying here at UCStrategies, really since the outset of UC, that it was going to be the systems integration or solution integration programs that were going to make this market. We think we’re at that stage in this market now. As we’ve discussed in prior podcasts, the products are getting pretty well rounded from many different venders. Now the customers are going to be looking more at the question of how does this fit with what I own, how easy it is to deploy, what are the economics; those kinds of questions. We think that is going to be something that the system integrators will have to answer because they’ll be the people in the room with the customer at the time.
So Alan, why don’t you open this up by giving us a description of the UC&C program and the UC reference architectures. Then we’ll let our UCStrategies experts pose questions to you and Jan.
Alan MacLeod: Great, thanks Marty. HP has been focusing on the UC&C industry now for approximately four to five years. We work very closely with Microsoft as well as some other venders to bring solutions to the market based on both their technologies as well as HP’s technologies. The primary focus we’ve had previously has been to work with our own global systems integrator departments inside HP – our technology services team – as well as our hosting and outsourcing groups – the enterprise services team – to bring these solutions to customers. In that, we’ve developed a lot of skills and knowledge on the UC&C market space. We’ve developed a lot of understanding of how to make Microsoft’s products work exceedingly well for customers, and address a lot of customers’ requirements.
So as we’re moving forward, we’ve developed a strategy working with our partners to bring the combined Microsoft-HP solutions to customers. What we’ve taken from our learnings from our hosting, outsourcing, and services teams, we’ve brought those back into our product portfolios to actually develop these reference architectures. So the reference architectures are the combinations of many years of knowledge and work and experience we’ve had inside HP. What we’ve tried to do is piece them together to allow partners to get an accelerated start on where they would be with a full UC&C stack.
I think as UCStrategies has made clear over the last number of years, UC&C is a very complex solution and a single vender is severely challenged to actually bring a complete solution to a customer that addresses all their requirements. Then the complexity in that multi-vendor combination becomes a big challenge. So our RA’s – our reference architectures – are designed to simplify that process. Jan if I can ask you, would you mind giving a very brief description of what a customer or partner would find inside those reference architectures, please?
Jan Kelly: What the partner or customer would find is the reference architectures are basically an HP configuration that takes a look at combining our experience, our relationship with Microsoft, and our test lab. We’ve come up with kind of a prescriptive set of hardware, which includes service and storage that is optimized for a certain Lync set of users or number of users, and a certain workload based on what those users are doing. In addition, HP extends that architecture to include the network, the end-user network that goes all the way to the desk – which is a key component of Lync – and gives also the prescriptive architectures for what kind of network to deploy. Again, both of these are optimized for Lync.
One thing to add is these are coming out as a series. There’s a whole family of reference architectures; it’s not a single one. They range from the smallest configuration, 250-2,500 users, all the way up to some of the large multi-site deployments of 80,000 users. So we can pretty much cover any size that a customer would require.
Marty Parker: That sounds pretty powerful. I’ve actually had a look at the 2,500-user reference architecture and one thing that impressed me was the amount of testing data that are reported out from your test labs to assure the customers that the performance thresholds can be met.
Alan MacLeod: I think the only other point to add is that as many customers know, and many partners know, HP has a very broad portfolio of products. We, as part of our UC solution, can extend all the way from obviously the device in your hand or the device on your desktop, all the back through the network into the server, into the storage that sits in the data center and everything in between. In that, we take away another layer of complexity, which is, most of the solution comes directly from HP itself.
Marty Parker: I’m going to start by calling on Kevin Kieller because Kevin is a member of the UCStrategies community who is very directly involved in systems integration work in Canada. Kevin, how does this resonate with you? What would you say about this, and what questions might you have for Alan and Jan?
Kevin Kieller: Thanks, Marty. From a systems integration perspective, I think this is a great initiative. Certainly small deployments of Lync can be two servers, but as you look at a mid-size and enterprise architecture and certainly, as Jan mentioned, if you look at up to 80,000 seats or when you get over 10,000 seats, it does get to be very complicated. Systems integrators have to sort through all that complexity. At the end of the day they're in front of the customer and they're making the promises in terms of getting all these pieces working together.
The fact that HP has taken their multi-year expertise, and I’m sure some good lessons learned and done some lab testing, some stress testing, these are things that systems integrators know they should do. Quite frankly, it’s hard to find the time to do that when you're dealing with individual customer deployments. So I think it’s a great initiative.
I have a two-part question. What percentage of the design work would you suggest that the reference architecture saves the system integrator doing? The second part is, if the systems integrator isn’t doing that design work – which might have been billable for them – what do you see their role and what are the additional revenue opportunities, because they’re kind of getting this accelerated start by using your reference architecture?
Alan MacLeod: Thanks, Kevin. I would like to jump on that one. We think we’re probably taking the partners to 80 percent of the solution. In that, as you referenced, we do extensive backend testing, we do extensive sizing/scoping of the various workloads and the way they're used, right down to actually building out individual SKUs-like components from the portfolios to actually complete the solution. So we feel for the partners what we’ve done is taken away a lot of that high risk, high cost work that sits typically back in their own offices and their own labs that normally isn’t tightly associated to a customer. It’s kind of the knowledge that you get from experience.
What we haven’t done for the partners and where we think we can’t do all this work for the partners without becoming too limiting for our customers, is the fine tuning for the customer solution – the bit where the customer knowledge comes in of how the customer wants to use it, how their own network is architected, how their users are architected, what their users actually want. That work has to be done by somebody who is right in front of the customer with knowledge, with experience of the customer, and that’s where I think our partners come in.
So to that end, we feel that what we’ve done is taken away a lot of risk for the partners. We’ve enabled them to go with confidence to customers with what they know is a solution that can achieve the goals and aspects of what the customer is going to look for without actually prescriptively telling them this is what the customer must have. So we’ve let them have the flexibility to choose what they want. It’s not exactly the Model T Ford analogy; you can choose other than black. But a lot of the work is done for you up front.
Marty Parker: Well that’s great. That’s a good question, Kevin. Let me move on to ask Joseph Williams, who has a deep channel experience here with the UCStrategies team, what’s your perspective on this? Do you think this is moving the ball forward, Joe, and what questions would you ask?
Joseph Williams: I think this is. Reference architectures have a lot of value and I think we all recognize that value. My question to Alan and Jan would be, there’s a set of reference architectures obviously that Microsoft distributes and that SBC venders distribute and things like that. What differentiates your reference architectures from those? And then, what value – given your answer – what value would that provide to the enterprise individual who is trying to figure out what bets to make here?
Alan MacLeod: When we look at typically a reference architecture coming out from one of our partner venders, an SBC is a partner vender and Microsoft obviously is a partner vender. What we see them producing is a fairly generic assumption based on the underlying infrastructure and very specific assumptions based on their own individual products. If you think about it, that makes a lot of sense. It’s hard for a company, for example the software vender, to say what every single component of hardware is going to do in a given setup. So what we’re trying to do and what we have done is we work with those other partners. So we do work with Microsoft, we do work with our SBC venders to fine tune our solution. We actually then are able to instantiate our own individual components as well. So if you can imagine, we add on top of the existing references that you see from the other venders, and again Microsoft is a great example because they’re obviously the core part of our software interface.
So, from that aspect, if I was an enterprise customer looking for a solution and a reference to start building from, then what HP has done with the RA’s, with our reference architectures, is take the other components they're going to consider as their solution and bring them all together. So from our perspective, I think what we would suggest that our strength is, we’re bringing a much more holistic description of everything you're going to need. And, to some extent, if it completely matches your requirements it’s going to give you pretty close to a kit list of products to pull out as well. But again, that leaves that opportunity or there still is that space for making sure it’s fine tuned to the customer’s requirements. That again is something that’s hard for even HP working with somebody like Microsoft where we control a lot of the hardware and a lot of the software to be some prescriptive that we know exactly what a customer wants. So there still is a little bit more work to do. But, we get them one more step further down that path to reaching their goals.
Joseph Williams: So the HP reference architectures have the advantage then of reducing that the uncertainty that enterprise faces?
Alan MacLeod: Exactly. That’s exactly right. In fact far better worded than my answers.
Marty Parker: Thank you for the question and the answer. Let me call on Roberta Fox. She, like myself, she and her firm spend most of their time with the end customer. So Roberta, what do you hear when you hear this presentation and these questions? What’s the customer going to think? Are they going to perceive this value or are they going to have questions?
Roberta Fox: Yeah, a good question, Marty. I’ll put it in context. I’m curious, as a former HP professional, consulting professional, who was exposed to and trained in all the great high quality engineering processes and methodologies. My question for Alan and Jan is how are you taking this expertise across the broad range of UC portfolio elements, which is servers and networks and mobile devices, to work with Lync or Lync applications to make sure it’s easier for the channels but also easier for the customers to make sure that they’re going to be successful? Does that mean things are pre-staged or pre-configured? I’m just trying to understand how does it all fit together bringing the products and the process to work with the integrators?
Alan MacLeod: I think probably the best place to start with that is there is no replacement for experience and knowledge. So what we’re not proposing with our RA’s is that a partner can come along with no experience with Microsoft, no experience with networking, no experience with servers and take our reference architecture and turn up a successful solution for a customer. That isn’t something that we’ve got to the point to be able to achieve for people. What we do have, to your comments on working with the technology consulting organizations inside HP, we have been able to leverage their experience of what works in architectural design and we have been able to experience what works in product architecture and an actual implementation and bring that into the RAs.
So, combining that with a knowledgeable partner, and again we’re looking to enable the knowledgeable partners because unfortunately voice is complex and as fabulously simplistic in some ways as UC has made many things, it hasn’t taken away all of the complexity. So we are looking to partner with smart partners or knowledgeable partners. I think for enterprise customers when they look at that, before if you think prior to us having our channel partnership program in place, the customer had a choice if they wanted to use HP hardware to work with HP or do a self-deploy. Now, to some extent, they have the flexibility to pull in all of the HP partners. So they can still work with HP. They can work with a partner of their own choice who is relevantly skilled and who has the right relationships with HP and Microsoft. To that end, I think for the enterprises, we’ve given them back some choice, but we’ve given them back choice in a way that I think actually helps them rather than choice that just becomes confusion.
Roberta Fox: That’s great. And my only other suggestion is to help the channels with some of that excellent process, structure, and mindset for deployment. So, for the larger organizations for the Lync deployments, just make it a cookie cutter and roll it out and get it working so people can use it.
Marty Parker: Roberta, I will echo your point. One of the things I would like to see the systems integrators, with HP’s backing, be able to do is to convert the savings that Kevin talked about already, the savings in the design cycle, and convert those more to customization of UC to the customers’ workflows and business processes. I don't know if you’d agree with that Roberta, but I think the customer could spend the same amount of money and get much more direct benefit by moving the money from one category to another.
Roberta Fox: Absolutely, or they could tell you tie in to Art’s points in preference for tying it into the business.
Marty Parker: Yes, right. So Clark Richter, let me call on you and ask again, I know you've got a channel perspective and a lot of other perspectives, but from your perspective what would you like to ask here?
Clark Richter: Thanks, Marty. My question is a little more tactical in nature. I’m curious in terms of beyond the reference architecture, is there a specific solution and how it’s fulfilled? Is there a SKU, is it a bundle available through distribution and things of that nature? And then the second question would be, how is it supported? Is there a single point of contact in terms of supporting the solution?
Alan MacLeod: HP as you know, is a very heavy channel-focused company. We make the vast majority of our products, if not all of our products, available through the channel. And a large portion of our revenue comes from our channel partners. So in that respect, looking at this whole UCC solution, all of the HP components are available through our channel partners. Now, there are some criteria for skills. For example, we do have specific networking partners. We do have specific server partners. But we have many partners that cover both. So, our two-tiered distribution to our major distributers into our major channel partners and into our smaller channel partners, all of them have access to the components that make up the solution.
The reference architecture information is publicly available. Anyone can go and pull it down and have a look at the information to understand exactly how things work. In that respect anyone can actually use and take that forward.
HP again has had a long, extensive history with Microsoft stretching back 30 years. In that, we have a lot of experience in supporting Microsoft products. We do sell support to our customers directly for the ability to support with Microsoft, oftentimes in conjunction with Microsoft, oftentimes it’s the front end to Microsoft, and oftentimes where the customer has a direct Microsoft support relationship, we’re able to work with that customer to leverage that support program, to give them the best possible support. That all depends on how you interface to HP.
So across the entire solution, HP has the ability to support all the Microsoft components. And of course, all the HP components are a part of our support programs. Many of the third party components we’re recommending are also part of our support programs. So for any given customer looking to deploy this, there’s a conversation that can absolutely be had with HP on whether we're the right party to give you support across the entire end-to-end solution. That, to be honest, is one of the unique capabilities we think we bring in HP. Not only do we support all of our own products obviously, but we often offer the first line of support for many of our partner products. As far as the Microsoft UC and HP UC solutions coming together, we can offer that frontline support for everything for the customer in that respect.
Marty Parker: Clark, I thought that was a great question. The customer, in many cases Alan, is looking at, on their current systems... they’re looking at SKU types of numbers for support. They’ve got this much for hardware, this much for software, this much for refresh. They see three line items and there’s their support plan. Are you going to offer as part of HP UC&C services through your SI partners, are you going to offer them such a simple SKU-based maintenance program?
Alan MacLeod: For most of the HP products, absolutely, that’s exactly how it works today. For some of the more specialist software components, because we typically have to do a sizing on the actual offering to the customer, it’s a little harder for us to offer it directly as a cookie cutter SKU through the individual partners. So that is an objective and a goal, but it’s not one we're at at the moment.
Marty Parker: Phil Edholm, you've got an architecture background in this industry and you would know the phrase “reference architecture.” What do you think of this initiative and what questions would you have?
Phil Edholm: I think it’s absolutely an outstanding effort to essentially take the complexity out, and especially for channel partners who don’t have the resources to be able to develop these by themselves and generate the assurance. Obviously it’s a huge benefit because in the end it enables people to combine best of breed components from HP, Microsoft, and others into a solution.
I think the one area though that is incredibly interesting with this is the fact of the network. HP is actually in a very unique position here because if you look at obviously the other major network players like a Cisco, they’re not actively promoting Lync and driving Lync as part of their reference architectures, similar with Avaya. And Juniper obviously doesn’t have the strong associations on the Lync side, and the servers and the other components. So HP is in a unique position there. However, it does bring up a couple of questions. First is, what do you see the advantages, from a networking perspective, of having an HP complete reference architecture including the network? And then, I think the reverse side of that question is obviously for a reasonable percentage of the marketplace they have an existing investment either in Cisco or in Juniper. What are the both challenges, but also support that they would get of choosing the reference architecture? So, kind of two questions that would be asked of the network – what’s the value of having HP and then how well does this all integrate if you don’t have HP?
Alan MacLeod: The first part of the question, I think probably I want to start with a really glib response if you don’t mind, which is, there’s no such thing as UC without a network. The reason I make a statement quite that direct is unlike a lot of the other traffic we push across a typical corporate network where the end-users sit, UC does require a lot of its traffic to get through when it needs to arrive – real-time traffic. As we add, go beyond voice and move onto video, the bandwidth required goes up as well. So in that respect, a network becomes a really important part of a UC solution. It’s not just a data center deployment.
I think, to the first part of the question where that unique advantage comes into HP, is when you look at it from the Lync/Microsoft perspective, a lot of Microsoft’s traditional partners are very, very strong in the data center play, and are very strong in the servers and the storage, or they’re very strong in the PC play. That makes sense. That’s why Microsoft has been a very strong company.
With HP, we’re also strong in those areas but the unique part is we connect them together. That, I think, is where we have a unique benefit to customers looking for a solution. Not only partnering with Microsoft where Microsoft is very, very good – on the clients, on the desktop, and in the handheld device as well as in the data center. We also plug all that stuff together and we do it in a way that’s appropriate. So HP’s unique value there is one of experience at all ends of this communication, or experience of all ends of every single device that’s involved.
Obviously with our networking products, we have a portfolio to match anybody’s in the market. It’s exceedingly strong. We have global availability, global distribution. We are one of the clear leaders in the networking market space. So, many of the things that you would need to make a UC solution successful, Quality of Service (QoS), the ability to control what traffic goes where, etc., is all inherently built into the HPN solutions today.
Again, looking at it from a forward-looking perspective, as we’ve seen, HP Networking (HPN) is one of the leading companies in the software-defined networking side of things. Amongst a lot of companies are folks on SDN (software defined network) or the data center where it’s obvious value comes into play. HPN has done that as well, but they’ve also focused on the access side. So, in that respect, we’re looking for SDN as being the next step forward in how to make our networks really support our traffic most effectively. We actually gave a demonstration in conjunction with Microsoft – which actually the video is now available on YouTube – of Lync being able to control how a network treated Lync traffic, but control from the data center and not control from the client. So in that, it becomes the trusted environment and we're able to make sure trusted traffic goes through. So that’s the immediate value I think for somebody like HPN.
Phil Edholm: There were really two concepts in that demonstration that I thought were very powerful, but they actually are very separate. One was the use of SDN to manage the flows in the network. But the second was actually, could be potentially independent of SDN, which was this API protocol that Microsoft created, and you actually implemented in that demonstration, which essentially the Lync server is passing off ordered pairs of IP addresses and ports saying this address and this port wants to have a real-time connection through the network with this address and this port. The benefit and value of that is eliminating actually what’s one of the biggest bugaboos in networks of trying to manage real-time data, which is needed to do deep packet inspection and guessing at the ports at the edge of the network to try to ascertain what is this traffic, what QoS flow should it get, what QoS handling should it get into, what class of service, etc. Do you have any immediate plans to implement that within the HP portfolio even without SDN? It seems like an incredibly powerful integration with Lync.
Alan MacLeod: We have the ability to do the traditional DPI-based (deep packet inspection) classification already, I think as you alluded to. We are leveraging our SDN solution today in other areas. For example, we have a security solution out there, which allows us basically to see rogue traffic and actually physically put a block directly on a port. Again, to your point, that takes away a lot of the need to do heavy DPI-based workload. We can just spot that traffic at the access edge and filter it out.
In terms of our plans to work with Microsoft, we clearly have a strategic relationship with Microsoft on UC. Clearly we are talking about and looking at all of these areas. What I can’t give you is a straight answer that says yes or no we're going to do something. But a thinking person could reach their own conclusions.
So just coming back to the second part of the question on the multivendor environment, today there’s now very few environments where it’s totally single vendor network implementation. So we’re very used to going into a mixed environment and interoperating with those other products out there from say Cisco or Juniper or other vendors. The good thing about networking is that open standards has become pretty much the way forward and it does allow that interoperability. Again, as I mentioned, HP has a lot of experience in that area. We’ve grown our networking business. We didn’t explode in at the volume levels we’re at now.
Marty Parker: That’s a very interesting discussion here. Thank you for the questions, Phil, about the network components, and Alan, for the answers. But let me push it one step further. I thought your point that there’s no UC without a network is right on target, not flip, but right target. And the point I want to make is that the networks we see today aren’t just the LANs and the WANs. It’s the wireless networks. And, our expert on wireless, Michael Finneran, is here with us. Michael, what do you think about the reference architecture, vis-à-vis the way we’re all living on the wireless networks and what does that mean to us?
Michael Finneran: Thank you Marty. I’ve been champing at the bit here. Excellent presentation Alan, but as Marty mentioned, one of the biggest elements in the UC deployments today is mobility. Microsoft now has a full suite of Lync mobile clients. Does the RA address any mobile capabilities – both cellular and of course, closer to home for HP, Wi-Fi?
Alan MacLeod: This initial version of the RA doesn't actually get as far as touching on the Wi-Fi components. But, let me continue to answer if you don’t mind. As you've seen one of the stated directions, one of the stated major focuses that HP has is around mobility. With mobility we mean mobility in the truer sense, not mobile in the sense of just the handheld devices. In that, there’s a lot of work that’s going on inside HP and we’re now taking out to our customers where we’re taking their business challenges, many of which are driven by bring your end device, and the users connecting directly to corporate networks with their own personal devices. How do we actually build those into the network and give… or build them into a solution and give them fair and reasonable access without creating enormous complexities for both the IT department as well as the end user? Of course the IT department will live with the complexity, the end user will not. The end user will go and do something else and it’s the something else we don’t want them to go and do.
So the RA’s today do not immediately address the wireless components. Wireless is part of our phase two plan. In that, Wi-Fi is a major part of what we're doing. Again for us, UC&C is so tightly coupled to mobility. We see the need for real-time communications increasing inside customers. We see the need for bring your own device and support for mobile devices significantly increasing inside customers – nothing unique there I don't think. The two of them are just hand-in-hand. So for us, addressing those two problems together is critical. And, enabling our partners to also address those problems for our joint customers is also critical.
Michael Finneran: Thank you, Alan.
Marty Parker: Great. Let me wrap this back around one more time to the customers by calling on my good friend Steve Leaden. Steve, what do you think a customer is going to be asking about or looking for when they see an HP reference architecture being presented by an SI?
Steve Leaden: You know Marty, thanks and Alan, thanks for joining us. I think it’s a very interesting time to be a part of the UCC landscape. I’m very pleased to see HP really stepping in here with both feet ready to go, if you will, with this reference architecture. It’s a very interesting time also to be in this space, especially being aligned with Microsoft Lync. The reason being that more and more we're seeing, especially amongst the enterprise community, individual enterprises asking for Microsoft to be vetted as part of the telephony and part of the UCC upgrade process.
What is the strategy here relative to service level agreements, to network assessments, who are your hardware partners going to be in the gateways and the endpoints and things along the lines of also just a complete kind of implementation cycle to fully support the client – especially again, as you're promoting Microsoft Lync here?
Alan MacLeod: That’s a multifaceted question, but it seemed quite simple when you first asked it.
Steve Leaden: Yeah, sorry.
Alan MacLeod: That’s fine.
Marty Parker: He works with the customers and they don’t usually keep their questions into one facet anyway.
Steve Leaden: Exactly, yes. Thank you.
Alan MacLeod: Yeah, they’ve got a real business to manage. Let me break down the answer into multiple sections if you don’t mind. From an SLA prospective, there are two components I think we can look at from SLA. One is the SLA against the service and the other is the SLA against the components, the products themselves. So, the easy one to answer is against the products themselves. HP offers warranty on virtually all of our products to some level depending on what product we’re talking about. So for example, many of the networking products come with a lifetime warranty. So in that respect of an SLA, we do offer the traditional things you could expect to see from a vender—we warranty our products. Then secondly, we do offer support. So customers are able to go purchase support from us, with that comes some level of SLA as well.
So, if you're just a typical end-user or normal end-user who is looking to buy your own products, implement them on your own infrastructure, either directly yourself, using HP as the SI, using another SI – one of our partner SI’s, then in that respect, HP’s ongoing SLA to you is going to be around the warranty of the products, as well as anything else that you may have got as part of the deal with our SI organization or through your partner.
The other aspect to the SLA side of things, I think where it becomes much more relevant to the question that you're asking is, with our enterprise services team, they do have a full hosted UC solution. That includes the messaging, the collaboration, and the voice. It’s built on the Microsoft platform, so it’s built in Exchange and SharePoint to Lync. In there, those customers do, as part of their service, get an SLA and it does talk about availability of the service, not just of the products themselves.
So in that, I think HP touches SLA’s in many different ways and enables partners to offer SLA’s if partners want to be the hoster, enables partners to offer SLA’s if they want to be the support organization, and enables HP to offer those SLA’s behind it, as well as HP as a services company that’s actually hosting an outsource company to offer SLA’s directly to the customer themselves.
Steve Leaden: Great thanks. And then, just my follow-up question would be, what’s your sweet spot targeted for both premise-based solution as well as the hosted base? Do you have a particular lion-sized target market that you're looking at or ranges of?
Alan MacLeod: Yeah, basically if you own a computer you’re one of our target market. For the on-premise solution where we're working with our partners, as Jan alluded to, we built RA’s from about 250 users up. So the initial release of the reference architectures will go from 250-2,500 users. So to us that takes us from where we think the realistic low end is, up to what we consider to be the mid-market. Again as Jan alluded to at the beginning, we have traditionally built RA’s from many other solutions up to about 80,000 users. Typically, I think, when we cross into 80,000 we start to see people thinking very much in a bespoke way. It’s very much how they need their solution architected. At that point, the value of the RA becomes lower.
By contrast, in terms of the customers we have and the customers we support – we can use HP IT as a great example – we go to many, many, hundreds of thousands of users. So, for us the sky is the limit. If you're in business and you have people in your employ, you're probably a target customer for us at the very high end. So where it becomes interesting is when do we cross over with partners and when we don’t cross over with partners. So we have an ability within our own consulting organization to deal with customers of a certain size. Typically it’s that mid-tier and up that we can directly deal with. That varies by country and varies by region on what that size point is. With our partners, we can go all the way down to that 250-point reference. With many of our global SI’s we can go all the way up to the absolute largest possible size. So really, we do address all of the market as we possibly can.
Some of the limitations on the low end, I wouldn’t say are limitations. It’s more a case of with the market going so cloud oriented in many ways, when you get below 250 the decision on whether you go to the cloud may be more…there may be more impetus to move to a cloud-oriented solution than maybe keeping on premise solution. That you can see, I think, in the Lync architecture kind of builds a little bit around that story. So, that’s where we have that soft, lower edge.
Marty Parker: Thank you for that, and Steve, great questions. I’d just come back and build on what Steve introduced there. Alan, you said that perhaps when it gets into the larger size, it’s more bespoke or customized. I think having a reference architecture when you walk into a large account is equally as important. It gives them a message that there is a best practice or at least a proven practice and they don’t need to reinvent the wheel. In fact, they shouldn’t try to reinvent the wheel. Sometimes large enterprises have people that are very good at inventing wheels and it’s not always the best practice. So, I think a reference architecture has its value all the way up the line.
Art Rosenberg, anything else you want to comment on here or question you would like to ask?
Art Rosenberg: Actually, it was covered to some extent between the cloud and mobility and the one area that always requires customization and I’d like to know to what extent the reference architecture deals with that – one way or the other. That is what I call mobile customer services. The contact center can become virtual, the customers are mobile, and putting that all together, is there anything in the reference architecture that offers good solutions for managing that kind of customization – the mobile apps, for example?
Alan MacLeod: This first foray into the reference architectures doesn’t get as far as you're describing, Art. It really just enables those skilled SI’s to build on that platform and add their value, bringing in the customized contact center, bringing in those mobile applications.
Art Rosenberg: Okay, are you saying that once you get into those customized applications they will need other tools that could obviously integrate with what you are?
Alan MacLeod: Exactly, yes. Many of those other tools are going to be existing partners, for example Microsoft. In that they can build in the capabilities from those companies. Again, the RA is built on a lot of the standard products you see from HP, so based around our standard service technologies, based on our storage technology, and our networking technology. They do offer a very open, standard platform to build on. So in that, we don’t add incremental limitations on what partners could be brought in to add those incremental tools to the story.
Art Rosenberg: That sounds good. Because you're creating some very strong layering here so that they don’t have to provide all the layers. They just add onto the top of it.
Alan MacLeod: Yeah, I think traditionally, I mean, my background has been in networking for many, many years – too many than I want to admit on this call. In that, I’ve often seen venders go to the appliance approach and really shrink something down. In many places that does work well, but I think once you hit UC – and I think UCStrategies has been very strong in this message – it’s not easy to make this a shrink wrap. It’s not easy for one person to do everything. So in that, what we’ve tried to do with the RA’s, Art to your point, is leave that layering opportunity there for the skilled partners to come in and add their unique value and customize the solution for the customers. What we’ve tried to do is get as close to simplifying the complexity in the back ends to enable them to do that and leave them the correct opportunities to go to market with their customers.
Marty Parker: Well, thank you Art for that. I’m going to wrap it up and say thank you very much to Alan MacLeod - director of UC&C Initiatives, Jan Kelly of UC&C Product Strategy and Readiness from Hewlett Packard, HP. They bring to this conversation a great depth of customer-oriented design, customer-oriented support. I think most of us know they have a huge professional services business ranging from networking across computing into email and now UC. I think this is going to be a real boon to the industry as HP puts its full weight behind this supported by the reference architecture program. So with that, thanks to everyone and we’ll talk to you next week.