Ivar, I disagree with you on couple of points.
1. When a company is buying a thin client solution, the look at performance and cost. They know that if the performance of the solution will not be close to what end-users are used to get with PCs, they will get an ear-full and solution will never succeed. So they definitely talk about protocols and performance. One of the main reasons VDI did not take off until lately is because RDP protocol is just not good enough for a large VDI deployment. Now that we have PCoIP, HDX, RemoteFX (soon), etc. companies are starting to look at VDI. 2. Cost. You said that revenue from a SunRay is $400. I am not sure if SunRays cost more in EU, but in US, depending on the model of course, they range between $200 and $400 per unit. Now, again, customers look at a full cost of the solution. That means, servers that host virtual desktops, storage, thin clients, etc. With SunRay solution, you need an additional "server" - SunRay Server. Thank God for virtualization! So now we can say that SunRay Server is just a VM running alongside your virtual desktops and that makes it even better because traffic is within the physical server. Before virtualization VERY few people even considered SunRay because of that "need for additional physical server" in their datacenter. Their thought was "why do we need to put another box into our already crowded datacenter if I can use Wyse clients and just throw Wyse Device Manager on one of my existing Windows servers"? So that was extra cost. Now, Sun came out with Sun VDI. Packaged VDI solution. Great idea! Works well out of the box. One issue. It does not scale as initially advertised. We were doing an implementation at one of our customers and we were told by Sun (after customer already purchased all required pieces - servers, storage, etc.) that we will need 3(!!!!) additional servers! Apparently there was a disconnect between engineering at Sun and Field SEs. Turned out that Sun VDI did not quite work for larger environments because of the need to support MySQL integrated database. So we had to add 3 physical (!) servers to the solution to host Core VDI servers. Again, goes back to cost. 3. Interoperability. I agree, SunRays offer a great deal of flexibility and can connect to a number of backend platforms. So problem is that so can Wyse and Dells of the world. Wyse clients can connect to Citrix, VMware, Terminal Servers, Hyper-V. The way I see it, the advantage that SunRays have, for now, is the "set it and forget it" ability. SunRays work for a very long time. Once you setup your environment, you do not need to do much of administration, only basic maintenance. However, with all of the competitors making their zero-clients too, I think this advantage will disappear soon. So I believe Oracle needs to act now. Not 6 months from now, but NOW. At least on the marketing side. At least with the release of SunRay 3. They need to get SunRay name out there. Otherwise, I fear, it will be too late. Again, I LOVE SunRays. I use SunRays. I pitch SunRays. I implement SunRays. I would like to do it for many years to come! I would also like to hear some Oracle guys' thoughts on this. I understand that you cannot disclose anything. But I would just like to hear your own opinion and not the opinions or plans of Oracle/Sun. Please feel free to contact me if you have any questions. Thank you! ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Vitaly Tsipris Systems Engineer -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Ivar Janmaat Sent: Friday, March 26, 2010 3:56 PM To: SunRay-Users mailing list Subject: Re: [SunRay-Users] Microsoft RemoteFX Hello Vitaly, I agree that Oracle should act now. We loose precious time because of the take over. I work closely with the Dutch Sun branch but because no forward looking statements can be made by Sun (Oracle will take over control of the Dutch branch in July) it is not possible at this time to make strategic sales plans and request product RFEs. This situation is now in effect for about a year. It has hold back Dutch Sun Ray sales considerably. I hope marketing and sales funding will be available after 1 july to recover the lost sales. For instance a proper Sun Ray 3 launch event is badly needed with a proper follow up plan. The question is: Why should Oracle invest in the Sun Ray VDI infrastructure? The answer is simple: Because it can be one of the best selling parts of Oracle's solution stack In order to be the best selling product it needs customers who want to buy it. So what makes it so special that it is the best choice for a customer to buy? I can assure you that people who decide on buying 500+ Sun Rays don't talk about PCoIP, ALP or VDPAU an Xvideo options. Those people expect "it" to work so they don't get complaints from the people who work with it. Because of the fact that Oracle wants an integrated stack, I believe more effort will be made to make the Sun Ray product work out of the box. This is needed, but not essential for selling Sun Rays. What is special about the Sun Ray is the fact that it is a three tier solution. No other product can deliver something remotely similar. The three tiers (Sun Ray - SR server - Windows/Mac/Unix) makes it possible to hookup different virtualization technologies without changing anything on the desktop. If you have 10.000+ desktop this is a mayor advantage! So you can migrate from Citrix to Windows Terminal Server to VMware to Virtualbox to Hyper-V to Verde 2.0 to ... what ever and you don't need to redeploy workstations. Or you can use all available virtualization technologies at the same time because some are better in some areas and others are better for other areas. I can not predict the shakeout which will happen in virtualization technologies in the next 10 year. Nor can the customer. So if you are a smart customer you want to buy a desktop which will support them all. That is why I would like Oracle to provide uttsc like clients for all virtualization technologies. If this can be established then the Sun Ray can be a standard in desktop/workstations. If this is the goal, and you can align marketing, sales and product development to get this done the reward will be tremendous. Total pc sales in 2010 is estimated at 336 million units. Half are mobile so I am not sure how Oracle will market the mobile Sun Ray. So about 168 million are desktops. Let assume 1/3 are business desktops. This makes 56 million desktops which are replaced every year. So what should we aim for? I would suggest a modest 10% for a start. So lets sell 5.6 million Sun Rays a year. At an average revenu of $400 per Sun Ray this would amount to 2.24 Billion dollars annually. Why should Oracle let this money go to Dell, Wyse, etc if it wants and can sell the Sun Ray as part of the whole solution stack to their customers! What can go wrong? Oracle can decide to only work on a Sun Ray connection to Virtualbox and Oracle VM. Then Oracle will place itself on the same level as other thin client solutions like Vmware/Wyse, Redhat RHEV, Microsoft/Wyse, etc. If this happens I too believe the Sun Ray will become a niche product because you have no compelling arguments for top management to buy Sun Rays. At best you may win a few proof of concepts designed by the IT department of some small company. I am an optimistic guy so I hope I can convince Oracle to follow the right path ;-) Ivar Vitaly Tsipris schreef: > Ivar, > > You make some good points. However, my feeling is that because for the > past 10 years people only heard of Wyse, Dell and HP, they will not > think that these vendors do not have the history of product development. > Add to it that Dell and HP are a platform of choice for a lot of > customers in their datacenters for servers and sometimes even storage, I > think Oracle/Sun has a steep hill to climb. > I am also concerned about some "reversed opinions" that customers may > have. What I mean by that is a situation where customers keep hearing > about PCoIP and HDX these days and do not realize that PCoIP is very > much like ALP protocol that Sun has had for years. So when we will have > a conversation with them about their VDI project, they will be thinking > "ALP... oh yeah, VMware has had it for a while now" as opposed to > "ALP... Sun had it for while and VMware is just now validating the > advantages of such protocol". > Bottom line is that I think Oracle/Sun need to throw their hat in the > rink as soon as possible. Otherwise, I fear they will be an "also runs" > instead of a leading VDI/thin client solution vendor. > Again, I could be totally wrong here. I am just expressing my concerns > and eager to hear people's opinions. > > Please feel free to contact me if you have any questions. > Thank you! > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > Vitaly Tsipris > Systems Engineer > > _______________________________________________ SunRay-Users mailing list [email protected] http://www.filibeto.org/mailman/listinfo/sunray-users _______________________________________________ SunRay-Users mailing list [email protected] http://www.filibeto.org/mailman/listinfo/sunray-users
