In another shock move for most of us (except people in the UK, who are now well used to price increases and are currently trying to get their heads around how the hell British Gas can justify a 35% hike the day before Centrica, their parent company, announced having made 1 billion pounds profit), Citrix have decided to increase their prices on all product lines in EMEA, effective 1st September. We've got another few months' grace on Subscription Advantage but they will also increase from 1st Jan next year.
Get your Citrix orders in before September guys!
Thursday 31 July 2008
Citrix hand Ingram Micro their US revenues on a plate
Citrix took the unusual step this week of canning 3 of their 4 distributors in the US. Only Ingram Micro remain and they will now be processing Citrix orders from all of the 1,900 partners in the States. Apparently 75-80% of the Citrix business in America goes through the channel and the US does (I think) around half of the total number for Citrix globally. Now I only got a C in GCSE maths but, by my reckoning, that equates to around about half a billion dollars going through one single distributor!
I can't for the life of me fathom out why Citrix would want to do this. For two main reasons.
Firstly, there are, oddly enough, one or two resellers around who don't actually want to be forced to buy from just one distributor. Credit is always an issue for the smaller resellers and, even for the bigger organisations for whom credit is not a hurdle, resellers want to know they are getting a good deal. Monopolies are never good for customers and if Ingrams can pretty much charge whatever they want, resellers may lose confidence very quickly. It also allows for unfair sales situations where, for example, the disti can supply a favoured partner with lower pricing than another, and neither does it allow a reseller to price-check an item. (Note: I am writing very much on behalf of the resellers - as a disti myself, the latter situation has led to as much pain as it has joy over the years!)
The second reason is VMWare. A battle royal has long since commenced, both in the server virtualisation arena as well as the desktop. The last thing Citrix need to do at this stage is jeopardise this by losing resellers, regardless of how few that may turn out to be. If a reseller doesn't like Citrix's decision but has already invested in this area of technology, which product do you think are they now going to push?
Wake up Citrix - you are now in a C.O.M.P.E.T.I.T.I.V.E. market, your own monopoly days are long gone!
Having said all that, Ingram Micro must be chuckling all the way to the bank about now. At a (probably rather conservative) average of 5% profit for the disti, they would have pocketed approximately $25 million on last year's figures. Twenty. Five. Million. Dollars. Margin. On one (quite small) vendor.
I am assured it won't happen in this country. I have a really good relationship with the guys over at Ingram who were taken on last year to push sales of Access Essentials (we don't sell it so there is no overlap with what Ingram do), but I may well now be extra, extra nice to them just in case it does!
I can't for the life of me fathom out why Citrix would want to do this. For two main reasons.
Firstly, there are, oddly enough, one or two resellers around who don't actually want to be forced to buy from just one distributor. Credit is always an issue for the smaller resellers and, even for the bigger organisations for whom credit is not a hurdle, resellers want to know they are getting a good deal. Monopolies are never good for customers and if Ingrams can pretty much charge whatever they want, resellers may lose confidence very quickly. It also allows for unfair sales situations where, for example, the disti can supply a favoured partner with lower pricing than another, and neither does it allow a reseller to price-check an item. (Note: I am writing very much on behalf of the resellers - as a disti myself, the latter situation has led to as much pain as it has joy over the years!)
The second reason is VMWare. A battle royal has long since commenced, both in the server virtualisation arena as well as the desktop. The last thing Citrix need to do at this stage is jeopardise this by losing resellers, regardless of how few that may turn out to be. If a reseller doesn't like Citrix's decision but has already invested in this area of technology, which product do you think are they now going to push?
Wake up Citrix - you are now in a C.O.M.P.E.T.I.T.I.V.E. market, your own monopoly days are long gone!
Having said all that, Ingram Micro must be chuckling all the way to the bank about now. At a (probably rather conservative) average of 5% profit for the disti, they would have pocketed approximately $25 million on last year's figures. Twenty. Five. Million. Dollars. Margin. On one (quite small) vendor.
I am assured it won't happen in this country. I have a really good relationship with the guys over at Ingram who were taken on last year to push sales of Access Essentials (we don't sell it so there is no overlap with what Ingram do), but I may well now be extra, extra nice to them just in case it does!
Thursday 24 July 2008
Buy your Riverbed, F5 and Expand shares now!
I've noticed that the point products in the Application Delivery Controller (ADC) market are rapidly being snapped up, typically by vendors who a) do a whole lot more than just accelerating traffic and b) see ADC as just a small, but critical, part of what they do (or, more accurately, should be doing).
As I see it:
1) Applications, file transfer protocols and web traffic are demanding more and more bandwidth.
2) The Internet backbones aren't quite keeping pace just yet, particularly in developing nations.
3) Latency will always be a problem.
For these reasons, WAN acceleration solutions and ADCs are increasing in importance to CTOs. Not necessarily so much as a single project perhaps (although these do still happen of course), but more as a kind of "tick in the box" type request. Enterprise-sized organisations are incorporating these requirements into larger projects and so the main IT vendors that are being expected to provide the solutions for those larger projects feel compelled into having an offering in this space.
The proof is in the pudding:
Juniper buys Peribit in 2005
Cisco buys Fineground in 2005
Citrix buys NetScaler in 2005
Citrix buys Orbital Data in 2006
Blue Coat buys Packeteer in 2008
Brocade buys Foundry Networks in 2008
I don't claim this is an exhaustive list, I'm sure there are others I've forgotten, but all of these acquirers had one thing in common: they all came from different backgrounds but obviously felt the need to back up their core product lines with some sort of WAN acceleration technology.
Which leads us to the tantalising question. Who's next?
By taking a glimpse at the Gartner Magic Quadrant for this area of technology, it doesn't really take a rocket scientist to work out who candidate number 1 might be. Riverbed sticks out like a sore thumb. It concentrates on WAN acceleration and nothing else, it's the market leader by quite some margin and, for the reasons mentioned above, there must be several pairs of lips being well and truly licked at the moment. Talk about primed to be plucked.
Who else? Well, my money's on Expand and F5. Expand is in a similar situation to Riverbed but would be a lot cheaper. F5 have a very sound and complete product portfolio, an exceptionally loyal channel and some very, very good technology indeed. Perhaps VMWare might be a candidate to buy them? VMWare have pots of cash, I don't expect F5 would set them back too ridiculous an amount and it would really annoy Citrix!
As I see it:
1) Applications, file transfer protocols and web traffic are demanding more and more bandwidth.
2) The Internet backbones aren't quite keeping pace just yet, particularly in developing nations.
3) Latency will always be a problem.
For these reasons, WAN acceleration solutions and ADCs are increasing in importance to CTOs. Not necessarily so much as a single project perhaps (although these do still happen of course), but more as a kind of "tick in the box" type request. Enterprise-sized organisations are incorporating these requirements into larger projects and so the main IT vendors that are being expected to provide the solutions for those larger projects feel compelled into having an offering in this space.
The proof is in the pudding:
Juniper buys Peribit in 2005
Cisco buys Fineground in 2005
Citrix buys NetScaler in 2005
Citrix buys Orbital Data in 2006
Blue Coat buys Packeteer in 2008
Brocade buys Foundry Networks in 2008
I don't claim this is an exhaustive list, I'm sure there are others I've forgotten, but all of these acquirers had one thing in common: they all came from different backgrounds but obviously felt the need to back up their core product lines with some sort of WAN acceleration technology.
Which leads us to the tantalising question. Who's next?
By taking a glimpse at the Gartner Magic Quadrant for this area of technology, it doesn't really take a rocket scientist to work out who candidate number 1 might be. Riverbed sticks out like a sore thumb. It concentrates on WAN acceleration and nothing else, it's the market leader by quite some margin and, for the reasons mentioned above, there must be several pairs of lips being well and truly licked at the moment. Talk about primed to be plucked.
Who else? Well, my money's on Expand and F5. Expand is in a similar situation to Riverbed but would be a lot cheaper. F5 have a very sound and complete product portfolio, an exceptionally loyal channel and some very, very good technology indeed. Perhaps VMWare might be a candidate to buy them? VMWare have pots of cash, I don't expect F5 would set them back too ridiculous an amount and it would really annoy Citrix!
Citrix announce Q2 08 results
Citrix released financial results for the second quarter of this year yesterday. I couldn't listen in to the call myself but a contact of mine did and he has given me permission to publicise his synopsis of what was said. Unfortunately I can't name him in person due to compliance issues surrounding his role but I'm sure he will accept my thanks in absentia for the below. (I am going to a wedding in Germany tomorrow so don't have the time to add my own comments today but will do so next week some time.)
Europe was by far the strongest region for the company, and grew revenue 22% versus the year-ago quarter, which outpaced the total 17% growth.
The company did see signs of a weaker economy in Asia Pacific with deals slipping out of the quarter and the same held true for the US.
One of the more important discussions that would relate to your geography is the high likelihood for price increases in Europe to offset the foreign exchange fluctuations. Management didn't provide a whole lot of detail regarding local pricing initiatives, but said details will likely be announced before the end of the quarter. I would be curious to see if this information would help close deals in the pipeline sooner, or push them out indefinitely...
The company's Application Virtualization group was up 7% year-over-year, but there was a slight decline in license revenue offset by solid renewals in subscription advantage contracts. Platinum continued to sell well and was 30% of total Application Virtualization license revenue for the quarter. Sales pipelines were said to be running at unprecedented levels, and they saw a record number of million-dollar deals in the quarter. Going forward the company expects XenApp license growth to be flat for the second-half of the year, which I think is a reflection of a cautious spending environment and also difficult year-over-year comparisons that were driven by price increases in Q2 of 2007. This was previously a segment that management thought would be growing in mid-to-high single digits.
Expectations for XenServer were cut in half. Management reduced expectations from $50M to $25M in 2008 saying that a slower ramp from OEM distribution and slowing economy forced their hand. In 2009, they expect revenues to at least double, so probably around $50M. Management sounded encouraged about the future opportunity, saying that of the 3,100 authorized XenServer partners, 500 contributed to deals in the quarter. They signed 800 new customers, but the majority are still in pilot mode from the sound of it. In Q3, they will be delivering the next release of XenServer focused around 4 key upgrades: 1) high availability; 2) disaster recovery; 3) P to V migration tools; and 4) XenCenter for configuration management.
XenDesktop has had a fairly successful launch - Citrix had 10,000 downloads in the first 30 days, and they already have 2 customers with more than 1,000 seats each. Sounds like many of the beta customers have committed to going live in the second-half.
Application Networking was a huge surprise as it beat expectations across the board. The new NetScaler MPX box accounted for just over 10% of the products shipped, and Enterprise (versus internets) were the biggest buyer. The company also signed a big deal for WANScaler worth over $5 million which helped the division out tremendously.
Europe was by far the strongest region for the company, and grew revenue 22% versus the year-ago quarter, which outpaced the total 17% growth.
The company did see signs of a weaker economy in Asia Pacific with deals slipping out of the quarter and the same held true for the US.
One of the more important discussions that would relate to your geography is the high likelihood for price increases in Europe to offset the foreign exchange fluctuations. Management didn't provide a whole lot of detail regarding local pricing initiatives, but said details will likely be announced before the end of the quarter. I would be curious to see if this information would help close deals in the pipeline sooner, or push them out indefinitely...
The company's Application Virtualization group was up 7% year-over-year, but there was a slight decline in license revenue offset by solid renewals in subscription advantage contracts. Platinum continued to sell well and was 30% of total Application Virtualization license revenue for the quarter. Sales pipelines were said to be running at unprecedented levels, and they saw a record number of million-dollar deals in the quarter. Going forward the company expects XenApp license growth to be flat for the second-half of the year, which I think is a reflection of a cautious spending environment and also difficult year-over-year comparisons that were driven by price increases in Q2 of 2007. This was previously a segment that management thought would be growing in mid-to-high single digits.
Expectations for XenServer were cut in half. Management reduced expectations from $50M to $25M in 2008 saying that a slower ramp from OEM distribution and slowing economy forced their hand. In 2009, they expect revenues to at least double, so probably around $50M. Management sounded encouraged about the future opportunity, saying that of the 3,100 authorized XenServer partners, 500 contributed to deals in the quarter. They signed 800 new customers, but the majority are still in pilot mode from the sound of it. In Q3, they will be delivering the next release of XenServer focused around 4 key upgrades: 1) high availability; 2) disaster recovery; 3) P to V migration tools; and 4) XenCenter for configuration management.
XenDesktop has had a fairly successful launch - Citrix had 10,000 downloads in the first 30 days, and they already have 2 customers with more than 1,000 seats each. Sounds like many of the beta customers have committed to going live in the second-half.
Application Networking was a huge surprise as it beat expectations across the board. The new NetScaler MPX box accounted for just over 10% of the products shipped, and Enterprise (versus internets) were the biggest buyer. The company also signed a big deal for WANScaler worth over $5 million which helped the division out tremendously.
Tuesday 22 July 2008
XenServer in a box. Literally.
One of our partners, 360is, are probably the foremost XenServer reseller in the UK at the moment. They are certainly the most pro-active and are leading the way in terms of advocating the product over VMWare throughout the end user community, not without success either, I hasten to add. The deserve their success too; they understand it better than anyone and have put a lot of time and effort into the product, both when it was XenSource and since the Citrix acquisition too.
(Now you may think I'm a bit stupid advertising our best partner to all and sundry, but the reality is, my lovely competitors are already constantly on the phone, offering them the world on a plate, so 360is are unfortunately far from the hidden gems they used to be. Still, we've only got Citrix to thank for that I guess...)
Anyway, I went to an event that they held yesterday which I found very interesting. Rob, the MD, asked the audience how many of them were using virtualisation and I don't think there was anyone there who wasn't. Bearing in mind that when the same question was asked of audiences 2 years ago, and the number of hands represented only about 10%, that's not bad. How prevalent virtualisation is within their data centres is another matter entirely, but the virtualisation message is undeniably filtering through to smaller companies now, as well as the large enterprises.
The most interesting thing for me, though, was their self-designed hardware device, specifically created to run XenServer. This is a 2U box that is kitted out with a motherboard you'd typically find in a mid-range system, a beefy amount of RAM (128 GB) and 16 core CPUs, but no on-board storage or disks. They include loads of free services such as installation and next business day on-site support and if, like one of their customers was, you were considering buying new hardware (from the usual well-known brands) specifically to implement a virtualised environment, you'd probably want to check this out first in my opinion. According to 360is' figures, you can save an awful lot of money.
I thought of it as a thin client for the data-centre, just with one big exception. Instead of a thin client costing about the same as a PC and not working nearly as well, this thing'll cost you less than a standard server and will actually do the job better!
(Now you may think I'm a bit stupid advertising our best partner to all and sundry, but the reality is, my lovely competitors are already constantly on the phone, offering them the world on a plate, so 360is are unfortunately far from the hidden gems they used to be. Still, we've only got Citrix to thank for that I guess...)
Anyway, I went to an event that they held yesterday which I found very interesting. Rob, the MD, asked the audience how many of them were using virtualisation and I don't think there was anyone there who wasn't. Bearing in mind that when the same question was asked of audiences 2 years ago, and the number of hands represented only about 10%, that's not bad. How prevalent virtualisation is within their data centres is another matter entirely, but the virtualisation message is undeniably filtering through to smaller companies now, as well as the large enterprises.
The most interesting thing for me, though, was their self-designed hardware device, specifically created to run XenServer. This is a 2U box that is kitted out with a motherboard you'd typically find in a mid-range system, a beefy amount of RAM (128 GB) and 16 core CPUs, but no on-board storage or disks. They include loads of free services such as installation and next business day on-site support and if, like one of their customers was, you were considering buying new hardware (from the usual well-known brands) specifically to implement a virtualised environment, you'd probably want to check this out first in my opinion. According to 360is' figures, you can save an awful lot of money.
I thought of it as a thin client for the data-centre, just with one big exception. Instead of a thin client costing about the same as a PC and not working nearly as well, this thing'll cost you less than a standard server and will actually do the job better!
Thursday 10 July 2008
Citrix vs. VMware vs. Qumranet
I don't expect for one moment that the people who read my blog (and, by the way, we have now reached the monumental landmark of over 200 unique visitors and 600 page views, which I am incredibly pleased with considering I've only been doing it 2 months, so thanks very much for visiting!) are not also loyal devotees to Brian Madden's website but just in case there are one or two unaware of Brian's work, I thought I'd bring this to your attention.
There is an Israeli company called Qumranet who have a remote display protocol called Spice. Brian regards this as being "one of the most amazing things I've seen in this industry in a long time". The remote display protocol (Citrix = ICA, Microsoft = RDP which is also used by VMware) is one of the most critical components of a good user experience in Virtual Destop environments, but possibly yet more important is the hypervisor that the whole thing runs on. Qumranet uses KVM as its hypervisor, Citrix and VMWare obviously use their own. So Brian and his pal Gabe Knuth are going to do a bake-off.
I love bake-offs and, in my opinion, vendors are far too reticent about participating in them. If you're confident about your technology being as good as you keep telling us it is, go ahead and prove it...! Anyway, this one will stack Qumranet's "Solid ICE" VDI product up against Citrix XenDesktop (using XenServer as the VDI host) and VMware's VDI solution.
As Brian freely admits, this test will be paid for by Qumranet but anyone who has met Brian would not doubt his technical independence and integrity, regardless of financial compensation. So, to avoid any complaints once the results are released, he wants your input on how you think the tests should be designed and carried out. Feel free to offer your 2 cents' worth here.
I, personally, cannot wait for the results. I would like to make an educated prediction at the winner but I am not really technically capable enough. So I'll hazard a not-very-educated guess, bearing in mind the hypervisor is being tested and not the protocol, so in other words weight and origin of hypervisor, basic virtualisation technique and amount of VMs on a machine will probably have more significant effects than just remoting capability:
1. Citrix
2. Qumranet
3. VMWare
We did a bake-off here at COMPUTERLINKS recently with ADCs (Application Delivery Controllers). We tested Juniper, F5, Packeteer, Blue Coat and Citrix. Don't tell anyone but Citrix WANScaler actually came out the best - and by quite some margin too. But then, I would say that, wouldn't I?
There is an Israeli company called Qumranet who have a remote display protocol called Spice. Brian regards this as being "one of the most amazing things I've seen in this industry in a long time". The remote display protocol (Citrix = ICA, Microsoft = RDP which is also used by VMware) is one of the most critical components of a good user experience in Virtual Destop environments, but possibly yet more important is the hypervisor that the whole thing runs on. Qumranet uses KVM as its hypervisor, Citrix and VMWare obviously use their own. So Brian and his pal Gabe Knuth are going to do a bake-off.
I love bake-offs and, in my opinion, vendors are far too reticent about participating in them. If you're confident about your technology being as good as you keep telling us it is, go ahead and prove it...! Anyway, this one will stack Qumranet's "Solid ICE" VDI product up against Citrix XenDesktop (using XenServer as the VDI host) and VMware's VDI solution.
As Brian freely admits, this test will be paid for by Qumranet but anyone who has met Brian would not doubt his technical independence and integrity, regardless of financial compensation. So, to avoid any complaints once the results are released, he wants your input on how you think the tests should be designed and carried out. Feel free to offer your 2 cents' worth here.
I, personally, cannot wait for the results. I would like to make an educated prediction at the winner but I am not really technically capable enough. So I'll hazard a not-very-educated guess, bearing in mind the hypervisor is being tested and not the protocol, so in other words weight and origin of hypervisor, basic virtualisation technique and amount of VMs on a machine will probably have more significant effects than just remoting capability:
1. Citrix
2. Qumranet
3. VMWare
We did a bake-off here at COMPUTERLINKS recently with ADCs (Application Delivery Controllers). We tested Juniper, F5, Packeteer, Blue Coat and Citrix. Don't tell anyone but Citrix WANScaler actually came out the best - and by quite some margin too. But then, I would say that, wouldn't I?
Subscribe to:
Posts (Atom)