IBM LinuxONE; what’s in a name?

So the new IBM LinuxONE has now been officially launched. And not to put too fine a point on it, the Lustratus opinion is that it is pretty much the best Linux server around. In fact, to really stiEmperor_300x230ck my neck out, the LinuxONE could become the premier Linux server of choice in the next 5 years. As long as IBM doesn’t trip over its own feet to snatch defeat from the jaws of victory…

Let’s just take a moment to reflect on what IBM’s got. The LinuxONE currently comes in two sizes, the full-scale enterprise Linux server (Emperor) and an entry level server (Rockhopper). Cunning use of penguins to stress the link to Linux ūüėČ . LinuxONE offers a range (if two is a range) of Linux servers with outstanding reliability, security and non-disruptive scalability coupled with probably the best data and transaction handling facilities in the world. Bold words, but there is proof (see later).

But the LinuxONE also offers the openness and productivity support expected in the Linux world. Customers can choose between Red Hat, SuSE and Ubuntu environments, a range of hypervisors such as KVM and PR/SM, familiar languages such as Python, Perl, Ruby, Rails and Node.js, various databases like Oracle, DB2, MongoDB, MariaDB. In addition, LinuxONE adopts open technologies extensively, including Openstack, Docker, Chef and Puppet.  Even the financiang for the LinuxONE is more aligned with Linux and Cloud expectations, with a usage-based fixed monthly charge or even a rental option being offered. The LinuxONE is even the basis of an IBM community cloud being rolled out now.

So how can anything go wrong? And anyway, how can I make those claims about reliability, security and so on? Well of course, the secret is that the IBM LinuxONE is based on the IBM mainframe, arguably the most proven server the world has ever known for reliability, availability, data and I/O handling, transaction processing and enterprise serving. To this base, IBM has been able to build on its extensive experience over the last few years of running Linux workloads and serving Linux needs with z/Linux, providing the ideal launchpad for delivering the ultimate Linux servers. Fortunately IBM has not tried to resist the march of open technologies, taking the opportunity to bring open, non-IBM and IBM offerings together with the aim of delivering the premier Linux server environment.

The ‘but’ is that IBM cannot manage to tear itself away from its pride in the mainframe. Rightly, IBM is very proud of its mainframe technology and its long history of success under the most demanding environments. Perfectly understandable. And so I suppose it is only natural that IBM would want to refer in all its marketing literature to the fact that the LinuxONE is an enterprise Linux mainframe, and to stress that it IS a mainframe, albeit with significant Linux and open technology support added. But from the outside, this makes no sense. let’s split the world up into three camps; mainframe fans, those who do not know about mainframes and the mainframe ‘haters’. Perhaps ‘haters’ is a bit strong, but there is absolutely no doubt that there are a significant number of companies across the world who for various reasons see ‘mainframe’ as almost a derogatory word; old-fashioned, expensive, etc.. So how will the three markets react to the LinuxONE? IBM mainframe fans don’t need to be told it is a mainframe; they know, and they will also usually have an IBM rep who will be pointing it out with great frequency! The uninitiated who know nothing of mainframes would not see any plus or minus from being told the LinuxONE is a mainframe; they will simply want to look at what the LinuxONE can do for them, what tools and environments it supports etc.. But the third category can only see the ‘mainframe’ word as negative.

I can almost hear some people pointing out that this is a silly argument. That anyone who starts to look at the LinuxONE and who knows anything will quickly work out it is essentially an IBM mainframe. But I would submit that is not the point. Reaction to the mainframe word is to put the third group off from taking a closer look. Once they do look, as long as the server has the tools and offers the capabilities they need, and they can carry it forwards in their company without overtly exposing the ‘mainframe’ word, the strength of the LinuxONE offering will carry it through.

So I make this plea to IBM. Please, please, remove ‘mainframe’ from all the literature. Replace it with ‘server’ or ‘Linux server’ or enterprise Linux server’ or whatever. LinuxONE should be associated with being the best, most reliable, most productive, most scalable, most effective and safest range of Linux servers in the world, not with being a Linux-enabled mainframe.

Was Visions Solutions right to acquire Double-Take?


vision1
On May 17th Vision Solutions announced that it was acquiring Double-Take – bringing together its own IBM Power HA/Disaster Recovery business and Double-Take’s strength in the same market with Windows and to a lesser extent Linux. But did it do the right thing?

On the face of it, this was a natural move for Vision. Its strength was in providing availability solutions based on IBM’s 64-bit Power servers. This market is a strong and inflential market, but is nowhere near the size of the Windows market. Also, since there are few players in the more specialist IBM Power market, Vision already holds a reasonable share. For growth, it was essential for Vision to do something. It had choices – it was already partnering with Double-Take to supply Windows to its IBM base, and had other partners for Linux. But by choosing to acquire Double-Take, it is definitely buying a Windows availability provider who also does Linux rather than the other way around.

My concern is that my instincts tell me the Linux market is going to be much more interested in availability than the Windows one. UNIX in general is a more secure and robust operating system than Windows, and therefore people using Linux may have greater expectations of availability, provided through file mirrors, backups, network switchovers and disaster recovery scenarios. In contrast, any user of Windows knows that the system is always going to have its availability problems – just think how many times you have to reboot when you use your own Windows laptop or desktop. The question is therefore; how strong and widespread is the demand for availability on Windows?

At some level, there is demand. For example, a lot of people would like their office data backed up so the can get it back quickly if there is a problem. But beyond this basic data level, it seems to me demand may be limited. Indeed, this seems to be backed up by the fact that in the past year or so Double-Take has suffered slowing growth and market pressures on its finances. Maybe Vision would have been wiser to acquire a Linux availability provider, like SteelEye Software for instance.

Linux v z/OS on IBM mainframes

Five or ten years ago, this sort of question would have been unthinkable, but now mainframe users are increasingly facing a choice between whether to use Linux on System z or z/OS to host new mainframe workloads.

These new workloads may be the result of a consolidation project, or simply taking advantage of flexible architectures like SOA to utilize spare mainframe capacity, but the decision is not an obvious one in either case.

On the one hand, long-time mainframe guys will say that z/OS has grown up with the mainframe and therefore must be the best choice. But IBM has done a lot to its version of Linux for the mainframe, and Linux bigots will be quick to point out that the license costs will be cheaper and there are strong advantages in standardizing on a portable and flexible operating system enterprise-wide. Worst of all, given the polarized nature of IT in general, the decision makers find it hard to get unbiased advice on such a divisive question.

In the end, the answer to the question of whether z/OS or¬†Linux on System z is better¬†is not surprising – “it depends”.¬†This subject is discussed in much more detail in a¬†free Lustratus report, “Choosing the right¬†SOA platform¬†on IBM System z”, available from the Lustratus web store. While this paper focuses particularly on developing or moving SOA workloads¬†onto System z,¬†the analysis applies to any new mainframe workload. Summarizing the arguments¬†in the paper, the major differences that affect the decision are that Linux is designed to offer a¬†common¬†environment across many platforms, and is¬†thus less attuned to individual¬†platform capabilities by definition, and that¬†whereas Linux has been¬†designed¬†for the ‘server’ model where it is used to operating one¬†type of workload, z/OS has been built to handle multiple subsystems from the start.

The common environment aspect of Linux offers flexibility, helps to drive license costs down and leverages widely available skills. The multi-system capabilities of z/OS combined with its close linkage to the System z platform offer the greatest exploitation of System z facilities. But as always the devil is in the details.

Steve