collective unconscious of the smarter planet

Positive press for OpenStack -

PayPal and eBay are yanking VMware software from some 80,000 servers and replacing it with the free and open-source alternative known as OpenStack, Boris Renski, OpenStack Foundation board member told Business Insider.

Renski is also a cofounder of Mirantis, an OpenStack consultant company backed by Dell and Intel. Mirantis worked with PayPal on the project, he says.

PayPal used a set of tools from Mirantis called Fuel, a PayPal spokesperson told us.

Today Mirantis released Fuel for free to the public. (In geek speak: it was released under the open source Apache 2.0 license). It’s a collection of scripts and software that helps companies deploy OpenStack.

Initially, PayPal is replacing VMware on about 10,000 computer servers. Those servers will go live this summer, Renski said. “The grand vision for project is, over time, they will replace all of their virtual infrastructure with OpenStack, not just PayPal, but PayPal and eBay, together,” Renski said. That’s about 80,000 servers across their data centers, he said.

PayPal has been a big supporter of Open Stack for a while. But this project is still dangerous territory for VMware, as PayPal could become an example of how other enterprise can replace VMware with OpenStack, too.

To be sure, enterprises are not dying to get rid of VMware, even to save money on software license fees. Most of them really love VMware’s software because it’s a reliable way to run lots of different applications on the same computer server.

But cloud wars are coming between different so-called “cloud operating” systems: VMware is up against OpenStack (backed by IBM, HP, Rackspace, Red Hat, others) and CloudStack (another open source project, backed by Citrix). 

Think of this like the desktop operating system wars: Windows vs. Macs vs. Linux.

The ultimate goal for most enterprises is something called a “hybrid” cloud. This means that a company will run some of its apps in its own data center (a “private cloud”) and some of them in a public cloud. They can easily move apps between the two.

More importantly, they can use multiple public clouds. If one provider has problems, they can  quickly switch to another.

Enterprises will have to choose their cloud operating system and the stakes are high. Enterprises will spend $80 billion on cloud computing by 2016, Gartner estimates.

So where does the biggest cloud player, Amazon, fit into all of this? Amazon is the reason for it all. These cloud operating systems rose to compete with it. The idea is to reassure enterprises that they if they choose an alternative to Amazon, they will have lots of cloud computing vendors to choose from, not just one.

The performance costs of virtualization - “virtualization overhead” —

One of the dirty little secrets of virtualisation is the performance cost: operating systems running inside a virtual machine are slower than those running natively on the same hardware, sometimes by quite some margin.

This is termed virtualisation overhead, and with current whole-system virtualisation, it’s a given. It always happens. The question is, how much.

Back in the days of ESX Server 3, VMware itself admitted that integer performance suffered an overhead of up to six percent and more complex CPU operations up to 18 per cent. It claimed that Xen 3 was about twice as bad.

These days, things are not so serious, and the performance differential between the main hypervisor vendors has mostly evened out.

Stiff competition

Even so, bear in mind that this was on a single host with a single virtual machine. Contention for shared resources is also a significant issue, especially when it comes to disk storage where virtual machines often share a single drive or array.

Most current virtualisation on x86 is whole-system virtualisation: each virtual machine is a complete emulated PC containing a complete PC operating system. The virtual machine’s “disks” are actually files in a file system managed by a different operating system.

All the componentry of that nice uniform hardware platform that means you can move virtual machines from host to host – network cards, motherboard chipset, graphics adaptor and so on – is not nice fast hardware, it is software emulations running as part of the hypervisor.

Hardware extensions

Gradually, x86 chips are acquiring hardware extensions to assist in this emulation. The first generation of hardware virtualisation assist extensions was Intel’s VT, introduced in some of the last models of Pentium 4, the 662 and 672, in 2005. AMD-V followed with the Socket AM2 Athlons in 2006.

This hardware virtualisation merely allowed hypervisors to create a “Ring minus-one” – essentially, trapping Ring 0 (kernel-mode) code and running it though a software emulator. The CPU-intensive process of mapping virtual machines’ memory to the host’s physical memory still had to be done in software.

This changed in 2007 with the arrival of AMD’s second wave of hardware virtualisation, Rapid Virtualization Indexing (RVI), in the Barcelona generation of Athlons and Opterons. RVI provides the hypervisor with shadow page tables in hardware.

Page tables hold the map that translates addresses in an operating system’s memory layout to physical RAM addresses. But from inside a virtual machine, these emulated physical addresses are actually blocks of the host’s RAM. RVI’s second level of indirection accelerates the translation of memory addresses inside virtual machines to real physical memory addresses.

This makes little difference to a virtual machine’s pure CPU performance, but significantly enhances memory-intensive workloads, to the tune of 42 to 48 per cent.

Intel’s equivalent is called Extended Page Table and appeared with the Nehalem-family Core i3, i5 and i7 processors in late 2008.

For the moment, this is all hardware can do to help. The remaining techniques are a matter of software and system configuration.

In praise of paravirtualisation

Emulation is expensive so another good way to boost performance is to avoid it. From the virtual machine perspective, one way to do this is to modify the guest operating system or its drivers to be aware that they are running in a hypervisor.

For example, a guest operating system can be provided with special drivers that talk directly to the virtual network connecting the virtual machines to the host machine, rather than emulating a physical network card.

Similar methods can be applied to storage (such as SCSI and iSCSI), graphics, input devices and even memory management.

Microsoft calls this Enlightened I/O for its Hyper-V Server; support is built into Vista SP1, Windows Server 2008 and later, and drivers are available for Windows Server 2003, SUSE Linux Enterprise Server 10 SP3 and Red Hat Enterprise Linux 5.2 to 5.5.

VMware had a similar approach, the Virtual Machine Interface, which allowed Linux guests to communicate with the hypervisor, but this has now been outpaced by hardware virtualisation.

VMware also offers the vmxnet virtual NIC, as well as enhanced vmxnet2 and vmxnet3 drivers that offer TCP Offload Engine acceleration to virtual machines on suitably equipped hosts.

For Xen, there are PV drivers and for KVM, Virtio, which both offer analagous functionality.

Just passing through

The final, and in some ways most drastic, step is to avoid the emulation overhead by directly connecting virtual machines to physical hardware.

The simplest and theoretically cleanest way of doing this is by offloading storage, for instance, to a SAN; a virtual machine accessing a SAN in principle suffers no more slowdown than a physical server would.

As in the case of a physical server accessing storage over the network, though, this ideally means dedicating network interfaces to storage – which may mean adding multiple network interface cards to the host and configuring dedicated routes between virtual machines and networked storage devices.

Hyper-V also supports pass-through disks, where a virtual machine can directly control a dedicated LUN of a storage device on the host machine.

Windows Server 2008 R2 adds a new feature, Cluster Shared Volumes, which allows multiple hosts to share access to a single storage LUN, adding a degree of scalability to pass-through disks.

VMware currently takes the prize in this department, though, with its ability to directly dedicate not only SCSI controllers, but as of ESX 4, entire physical PCI and PCIe devices to a specific virtual machine.

The VMDirectPath feature allows one or two PCI cards in the host machine to be connected to the operating sysem running in a specific virtual machine rather than being managed by the hypervisor itself – from a simple USB controller to a dedicated physical NIC or storage device.

A slightly more modest optimisation is to place the swapfiles of Windows virtual machines directly onto the host’s vmfs storage.

There is always a price to pay, though, and this one is a biggie. There is a significant drawback in attaching dedicated devices to virtual machines, whether they are just disk partitions on the host server or physical interface cards and any attached devices.

Although such techniques can deliver pretty much full native performance, they hinder some of the key advantages of virtualisation: the ability to snapshot virtual machines for backup purposes, duplicate them and migrate them from one host to another.

At best, virtual machines accessing external, non-virtual resources often need to be shut down before migration, and snapshots must also duplicate any external resources – thus removing the scalability and fault-tolerance of virtualisation.

Differences of opinion

Virtualisation is now a key part of the x86 platform and it is not going to go away again. Further hardware advances will continue to improve speed and reduce overhead but there’s a long way to go before x86 servers can match the performance and scalability features of systems that have been doing virtualisation for decades, such as IBM’s System z mainframes.

On the other hand, paravirtualisation is also important. There are significant gains from having guest operating systems that know they are guests and can request services from the host server or its assistance in performing demanding operations.

Some of the possible improvements will probably remain limited by competitive demands, such as the different virtual machine formats of all the main hypervisor vendors and their totally different driver architectures.

Historically, such issues have receded either when everyone becomes compatible with Microsoft’s formats or the functionality moves into hardware.

That is still some way off for x86 virtualisation but it is a rapidly developing area, so watch this space.

Some interesting statistics on virtualization from a vendor study.

Server virtualization is perhaps not as pervasive as many believe, and customers are not as locked into any particular hypervisor as many companies peddling this magic software layer might hope.

This info comes from the latest V-Index survey from Veeam Software, a maker of add-on management tools for VMware’s ESXi hypervisor, which is conducted on a quarterly basis in the US, UK, France, and Germany.

The survey only talks to fairly large companies – those with 1,000 or more employees. About a third of the companies surveyed had more than 3,000 employees. Veeam survey not merely its own customer base, but rather server shops at large, which also have other hypervisors.

In the September V-Index, 86.5 per cent of the 578 organizations that participated in the poll had some sort of server virtualization in their data centers. And across all enterprises, including those who did not have server virtualization at all, an average of 38.9 per cent of servers were virtualized, and they had an average of 701 servers in their data centers.

From the looks of the data, only x86 servers and hypervisors were part of the survey. This is a serious shortcoming, considering how much Unix and proprietary gear is still out there among large enterprises.

Just for fun, the V-Index survey asks companies how many physical servers and virtual machines they have, and then asks separately what consolidation ratio they are attaining on their machines. With the former, you can calculate the actual consolidation ratio, and the latter is – at least according to Veeam – the perceived consolidation ratio, which usually turns out to be higher than the actual consolidation ratio in the V-Index survey.

Across the four geographic regions and all the companies surveyed, the perceived consolidation ratio was 9.8 virtual machines per physical machine. But if you do the math and calculate the actual penetration ration, companies are actually squeezing only 5.1 virtual machines per host on average.

It could just be that some IT managers garbled their responses and have screwed up the data, but perhaps Veeam is on to something.

The penetration of various hypervisors on x86-based servers depends on whether virtualization is being used to run virtual desktop infrastructure (VDI) or more traditional server workloads.

On traditional server stuff, VMware still rules, with 67.6 per cent of those companies that have hypervisors saying that one or another flavor of ESX Server or ESXi is their primary hypervisor, with only 14.4 per cent going for XenServer from Citrix Systems, and 16.4 per cent opting for Hyper-V from Microsoft. Red Hat’s KVM is tossed into the Others category, which accounted for a meager 1.6 per cent.

When you shift to talk about hypervisors running on servers to specifically stream VDI desktops, then XenServer is cited by 24.9 per cent of shops as their primary hypervisor, Hyper-V by 20.3 per cent, and ESX by 54.2 per cent.

Now here’s the interesting bit: 38 per cent of companies using virtualization for traditional workloads say they are planning to change their hypervisor during the next year.

The cost of the current hypervisor platform was cited as the main reason for the jump by 58.9 per cent of the jumpers, with nearly half saying that they didn’t like their current vendor’s licensing model, and they did like the features offered with alternative suppliers or that the alternatives had matured enough that they could contemplate making a shift. While VMware was not mentioned specifically by name as the one that companies were thinking about ditching, KVM, Hyper-V, and XenServer have been in catch-up mode for a while, and have largely caught up.

But of course, VMware is focused on cloud fabrics and infrastructure and platform clouds, areas in which Citrix, Microsoft, and Red Hat are playing catch-up.

Some interesting thoughts from the VMworld conference on what enterprise IT can learn from web-based companies.

Ragavan got plenty of information across in his 55 or so minutes – and the key narrative was striking because it was so explicit about what the enterprise can learn from the Web, in terms of automation and management in the cloud era.

Rather than making claims that web company IT workloads are somehow different, and easier, than enterprise IT, Ragavan laid out a clear vision of why the web works.

The most obvious strategic imperative we can learn from the web is simplification. Web companies generally support only a small range of system images for developers to target, which is one of the reasons a company like twitter can support a global user population in the hundreds of millions with a support staff that amounts to a handful of people.


Enterprise IT in contrast thrives on complexity, but generally offers an operator to user ratio a couple of orders of magnitude lower. Of course transactional IT requires uptime guarantees that twitter certainly doesn’t, but the mistake we often make in IT is that all workloads require “enterprise-class” reliability.
Ragavan not surprisingly argues that we can have web class cost and capability with enterprise class RAS by learning the lessons of the web, but using VMware virtualisation. The big goal -10,000 virtual machines per administrator.“What has changed? How you deal with failures. in the traditional data center dealing with failure is the most expensive aspect of IT, with complex systems infrastructure needed for high availability, and people available on their pager 24 hours a day. “These modern [web company] data centers though expect hardware to fail, and deal with it in software. Instead of trying to fix the hardware as soon as it happens, if a server fails nothing happens. at the end of the week they send someone to replace it.

“How is this done today? Start with open source code and ad a lot of compsci phDs”

That’s last line is as good an explanation of DevOps as any. But of course most enterprises aren’t going to throw phDs at the IT management problem. No – they want to achieve the same ends by licensing software. Which is where VMware wants to come in, with a new bundle of management products (I won’t call it a suite because the disparate products haven’t been integrated yet.) I find it kind of funny to see a company an an event designed to sell software to ops people talking up node.js, but of course node is the future, and shows VMware as an industry leader.
I like the fact VMware is so explicit about learning from the web. Its a useful approach, that should benefit customers.  Other enterprise vendors take note.

Red Hat has been joined by some major heavyweight - including IBM, Cisco, and Intel - to push a more full-featured version of KVM as an alternative to VMware.

Red Hat is taking on VMware with five enterprise heavyweights through a vendor-neutral virtualisation community project based on its RHEV-M stack.

Red Hat has been joined by Cisco, IBM, Intel, NetApp and SuSE to lead oVirt Project, planning on building a pluggable hypervisor management framework along with an ecosystem of plug-in partners around its virtualisation management tool for KVM.

To seed the project and motivate the community, Red Hat is releasing the RHEV-M code to oVirt under an Apache Software Foundation (ASF) licence.

oVirt’s official launch is in early November, at a workshop day that will be hosted by networking giant Cisco at its San Jose, California, campus. The ASF’d RHEV-M code will be released to Git repositories at this event. Today oVirt consists of 13 projects with the expectation for another two to be added by November…

The Linux distro vendor now hopes the OVA will become heavily involved in marketing oVirt. Joining Red Hat at the OVA launch earlier this year were IBM, Hewlett-Packard, Intel, Novell, BMC, and Eucalyptus Systems; today OVA claims more than 200 members.

The idea behind oVirt is to give people who want an alternative to VMware more than just the choice of going with a Red-Hat subscription for RHEV-M or using Microsoft’s Hyper-V, Red Hat told us.

“This will make things radically different,” Red Hat technical director Carl Trielof told The Reg. “Massive adopt of the technology as an alternative to VMware can only be a good thing.”

“This is going to be the first open and openly governed community that’s an alternative to VMware,” he continued. The watch word here is “OpenStack”, the open-cloud project Red Hat didn’t participate in because it disagreed with a governance model and direction dominated by a single company: Rackspace.

oVirt has been in development for three months with Red Hat working on the governance model in with the other five according to Trielof.

“We are not owning the community as Rackspace did around OpenStack. If we really believed there should be an open virtualisation solution then the first question is not what’s best for Red Hat, but it’s what’s best for the community and the user base. If the project thinks in those terms… then it’ll be good for us and everybody.”

Red Hat won’t lose out by putting RHEV-M into the community, he claimed, because the company will continue to sell enterprise subscriptions to the hypervisor and management software. He drew an analogy between Red Hat’s patronage of Fedora and its use as part of the company’s Linux distro.

Trielof said the oVirt Project is being set up as a hybrid of Apache and Eclipse: the former being a place that hot-houses projects and elevates individuals to leadership roles based on their contributions to projects while the latter is based on the ability to build a technology framework that’s enriched through a lively ecosystem of plug-ins from ISVs and individual developers. ASF founding member Jim Jagielski is helping on oVirt board to get it established.

Red Hat went with an independent organisation rather than dropping the oVirt projects into Apache because some features in the KVM are under the GPL and LGPL.

Also, there’s no opportunity in Apache to release the hoped-for 15 oVirt projects as an integrated release – projects would have to be developed separately.

Trielof added that Red Hat wanted to create a brand and a location for the community rather than simply shunting it into Apache.

The vision is to emulate Eclipse, not just from the perspective of having an ecosystem of plug-in providers, but also in terms of delivering an integrated platform composed of projects and technologies.

Eclipse now provides an annual synchronised update for many of its projects to remove bugs and give those using and developing products on Eclipse a reliable platform. Eclipse in June put out its biggest single release so far: 62 projects, or 46 million lines of code.

Eclipse was kick-started by IBM in November 2001 as a community and technology project. IBM primed the project by donating $40m worth of source code from its Websphere Studio Workbench to create an open-source framework for Java and C++ server-side development.

The idea, on paper, was to build an open IDE framework that didn’t lock in tools partners and that saved ISVs from having to constantly re-invent the basic building blocks of IDE such as syntax editors.

Red Hat was among the founding members of Eclipse with IBM, SuSE, Borland Software – now owned by MicroFocus – TogetherSoft, borged by Borland, and others. The group expanded quickly, hitting 80 members by the end of 2003, ultimately pulling in Oracle and SAP. By the late 2000s when it came to tools, the world was represented by Eclipse on Java and Microsoft on .NET.

Eclipse consolidated IBM’s WebSphere while putting some tools vendors, like Borland, out of business and driving others into the arms of IBM. Eclipse broke out of being just an IDE project for the server in into PC and web development, business intelligence and even runtimes.

If oVirt is successful the Eclipse model might prove a burden for Red Hat as well as hurt VMware. For all the success of Eclipse and its long-list of community members, it’s been IBM that over the years has consistently provided most of the manpower and code commits.

That has meant while Eclipse has helped IBM’s Java business, IBM has been left pick up the bill on development while struggling to sell paid-for tools against Eclipse, whose price point starts at zero dollars.

Chuck Hollis looks at how virtualization is changing the way that organizations design and provision databases. His focus is obviously VMware, but it’s an interesting discussion nonetheless…

Inexorably, virtualization is changing how we think about every aspect of IT.

It’s already vastly changed how we think about physical IT infrastructure: servers, storage and network. 

From static allocations to dynamic pools of resources – without VMware’s popularity, we really wouldn’t be talking much these days about cloud and transforming to IT to a service.

But what about databases and data management?  

Clearly, many of these technologies haven’t made the transition to the new model.  At best, we’ve only been able to encapsulate and containerize legacy databases using virtual infrastructure vs. revisiting how databases might intelligently work in this new model.

How do we get databases to intelligently use dynamic resources?  How do we deliver database as a service?  And how do we make databases as easy to consume as other forms of infrastructure?

…

Today’s Databases Were Designed For A Physical World

The more you look, the more you realize that the vast majority of databases in use today were designed to operate in the physical world, and not the virtual one.  And that’s far from ideal.

One immediate example is the lack of dynamic resource utilization.  All databases use precious resources: memory, CPU and storage.  Even though most database workloads are extremely variable, the vast majority of databases expect a big, fat over-allocation of these resources – they’re not smart enough to request more when they need it, and give it back when they don’t.

Any “virtualized” database should be smart about the environment it’s running in – smart enough to request and release resources as circumstances change.

Another example is the absence of integrated provisioning using standard templates and workflows.  

Today, provisioning infrastructure can be dead simple on something like a Vblock: the administrator defines standard infrastructure services and templates, and creates or changes resource instances using a very high level of automation.

This isn’t true for provisioning new database instances – it’s still mostly a manual process that requires hands-on work by an important and scarce resource – the database administrator.   Any “virtualized” database should be as easy to provision as a virtual machine.

A final example might be the need for self-service portals.  In the infrastructure world, using products like vCloud Director, it’s easy for administrators to expose resources to anyone at all: other IT groups, even end users.  A simple portal explains your choices, collects your details, and gives you what you want: typically with little or no human intervention.  

More importantly, the system administrator is additionally armed with powerful tools that help manage the pool of resources: allocation, service levels, and so on.  Again, if we’re talking fully virtualized databases, the same generic model should apply.  

Consuming a new instance of a database should be as conceptually simple as consuming a virtual machine.  And managing pools of databases ought to be as straightforward as managing pools of virtual machines.

Ideally, virtualized databases would support dynamic resource usage, integrated provisioning and self-service pooled consumption models.  But, outside of a few exceptions, that’s not the case today.

Dynamic Resource Usage

One of the first things that leaps out of the announcement is the virtual enhancements the VMware team has made to the popular PostgreSQL database.  At the outset, a “balloon driver” is able to request and release memory based on changing circumstances.  The same sort of capability seems to be there for GemFire.  The announcement is pretty clear: more options coming over time.

Extending this idea a bit, it would be logical to assume that – eventually – this concept could include to storage performance (perhaps using a variety of mechanisms that are extensions of VAAI: linked clones, Storage vMotion and/or storage service pools) —  creating and releasing additional database storage instances (or perhaps relocating them to different storage tiers) thereby increasing or decreasing performance.  The same expand/contract potential exists for dynamically using virtual or physical cores.

I have observed that massive over-provisioning seems to be the accepted norm in the database world: overprovisioning on memory, overprovisioning on storage performance, and overprovisioning on CPU.  

Wouldn’t it be wonderful if databases were smart enough to take what they needed to meet changing service level requirements, and no more?  If they had the same elastic properties as other portions of the infrastructure?

That’s the goal here.

Integrated Provisioning

Everyone who has had the pleasure of doing physical server provisioning knows all of the sequential, labor-intensive and occasionally error-prone steps involved.  

Indeed, anyone who’s working in a fully integrated virtualized environment (such as a Vblock using UIM) probably doesn’t want to go back to the old way of doing things anytime soon.

Indeed, in these new environments, valuable system administrator time is now spent on more worthwhile, higher-order tasks vs. the drudgery of before.  

The database administrator in many regards is no different – their time is important as well, and they could greatly benefit from the same sorts of capabilities: far less time doing sequential, labor-intensive and occasionally error-prone grunt work; and far more time tackling the more interesting challenges and opportunities.

I haven’t had the opportunity to look at the new vFabric Data Director in gory detail, but from what I can see from the overviews, there appear to be the same sort of templates and automated workflow concepts you see in virtualized server provisioning workflows.

Ease Of Consumption

Today’s pooled and virtualized environments are designed to be easy to consume — that’s what the whole “as a service” thing is about.  

Popular request types can be easily exposed on a portal, and people can get what they need with an absolute minimum of human intervention.  Behind that, resource administrators now have powerful tools that help manage and control the pooled environment in aggregate.No such luck for most of the database world today.  

Getting (or changing) a database instance almost always involves tracking down a database administrator and asking them to do something on your behalf.  And, while database administrators have the tools to manage individual database instances, there’s not much out there that addresses their need to manage and control hundreds or thousands of database instances being delivered as a service.

That changes with the new vFabric Data Director.

Digging Deeper

I think once the novelty wears off, most IT thinkers will realize a few simple truths.

First, there’s a big and obvious problem to be solved here.  

I routinely meet customers who have hundreds and occasionally thousands of database instances swirling around their environment.  Telling people not to create new databases just means they’ll go elsewhere.  Not good.

And no one has the stomach for a massive “gee, let’s go consolidate a bunch of existing databases into a single humungous instance” project.  At least, not twice :)

The only viable approach for many?  Use virtualization techniques to lessen resource usage, control service delivery and manage the pool of database instances more efficiently.  Just like you do server instances.

Second, while the technology is capable of supporting demanding workloads, that’s not where it’s going to be used first.  Just as with server virtualization, the most appealing initial target will be non-critical database workloads vs. the big hairy stuff.   Make no mistake, that too will come — in time.

Third, the underlying hybrid cloud model is extremely relevant here.  If you think for a moment about external database and PaaS offeirngs (e.g. AWS, Azure, et. al.) there’s only one consumption option for each: their particular service.  Easy to get in to, somewhat more difficult to get out of …

Compare and contrast this with the vFabric Data Director approach where you’re free to set it up internally, use any number of compatible external service providers, or any particular combination that suits you.  

Fourth, I’ve met more than a few people that are looking for a different industry model to deliver database services to the business vs. buying more of what they already have.  Here’s a model that’s worthy of serious consideration.

The nitty-gritty details on HP’s new VirtualSystem and CloudSystem converged solutions. Very technical. Be warned.

Just in case you can’t keep the HP product names straight, Blade System is the name of the company’s blade chassis and related x64-based server blades. Blade System Matrix was a special bundle of blade servers and deployment tools created by HP in response to Cisco Systems’ entry into the blade server market more than two years ago, while CloudStart, launched in August 2010, is a private cloud packaging of HP blades and tools with a cloud control freak called Cloud Service Automation, as a possible add-on.

CloudStart is sold by HP Technology Services (which was formerly not part of HP’s Enterprise Server, Storage and Networking group but which now is) and was certified to run VMware’s ESXi and Microsoft’s Hyper-V server virtualization hypervisors and used HP’s own EVA arrays for storage for the blades.

Back in January of this year, HP’s ESSN Group, which seemed to be often working at cross purposes to the services group, rolled out its own cloudy infrastructure based on BladeSystem blades called CloudSystem. This stack includes Cloud Service Automation Suite, which has some control freakery licensed from Adaptive Computing as well as the Matrix goodies HP has created. The CloudSystem also uses HP’s 3Par disk arrays instead of its EVA arrays.

If customers want to build a stack of virtualized servers and don’t yet want to go all the way to CloudStart or CloudSystem cloudy infrastructure, HP lets them back off a notch to something it calls VirtualSystem. HP previewed this VirtualSystem setup back in June and has now certified it to run VMware’s latest ESXi 5.0 hypervisor and its related vSphere 5.0 tools for managing virtual machines.

The VirtualSystem stacks include pretuned hardware – servers, storage, and networking – as well as management tools from both HP and VMware and licenses for the VMware hypervisor. The VirtualSystem is built at HP’s server factory in Houston, Texas, and ships pre-integrated at the rack level. HP says that it can ship it to you and have it up and running, doing useful work, in a matter of days.

The VirtualSystem for vSphere 5.0 stack has all of the storage APIs in the vSphere stack hooked into the Lefthand Networks storage arrays from HP, and in some cases HP is running its Lefthand SAN software on ProLiant blades inside a virtual machine if customers want to do it that way. (This is similar to VMware’s own Virtual Storage Array, launched back in July.) HP has also cooked up its own variant of VMotion for Storage, which allows for the files that comprise a virtual machine to be teleported around the storage to be more closely located to the physical servers where the VMs are running.

This VMotion for Storage feature that is enabled in the VirtualSystem running the vSphere 5.0 stack can operate over metro distances and can work on Lefthand virtual SANs separated over those distances. If you want the full-tilt-boogie CloudSystem with the Cloud Service Automation Suite, HP offers an upgrade path from the VirtualSystem.

The VirtualSystem server virtualization stacks are intended for customers setting up their own pools of virtualized servers, and they come in three sizes. The VS1 entry configuration has eight of HP’s two-socket DL380 G7 servers with a total 96 cores running at 3.46GHz, 1.54TB of of main memory, and 2.34TB of total local disk capacity (that’s 96GB of memory and two 146GB disks per physical machine).

To this HP is adding four Lefthand Networks P4500 arrays with a total of 57.6TB of storage and its homegrown switches with a total of 96 Gigabit Ethernet and 48 10 Gigabit Ethernet ports to link it all together. This configuration is designed to support up to 750 virtual machines and includes licenses to VMware’s vSphere 5.0 Enterprise Plus package. This bundle costs $167,300 including three years of support. If you already have VMware licenses, you can apply them to this VS1 stack to lower the price.

The VirtualSystem VS2 configuration for vSphere 5.0 moves to a bladed server and bladed Lefthand P4800 SAN arrays. The VS2 setup has a dozen BL460c G7 two-socket Xeon blade servers and two BladeSystem c7000 blade server chassis. Each blade has a dozen cores running at 3.06GHz and 192GB of main memory (expandable to 384GB), two 146GB disks, and links out to the network through four VirtualConnect Flex-10 switches (two per chassis). Four SAS switches (two per chassis) link out to the blades that make up the P4800 SAN storage blades, which have a total of 84TB of capacity.

The chassis has room to add another 16 servers, and the VirtualConnect switches implement 48 Gigabit and 48 10 Gigabit Ethernet ports. There is a total of 240Gb/sec of aggregate I/O bandwidth in the VirtualSystem VS2 box. In its base configuration, HP says it can support up to 1,125 VMs, but with the full complement of 28 server blades it can support around 2,500 VMs.

The largest VirtualSystem for vSphere 5.0 setup is the VS3 box, which is designed to support up to 6,000 VMs. This monster has four BladeSystem c7000 chassis, a total of 64 of HP’s ProLiant BL460c G7 servers (that’s 768 cores at 3.06GHz and 6TB of total memory). Instead of Lefthand SAN storage, HP is swapping in 3Par F series SAN arrays with 163.2TB of capacity. Each chassis has tow of the VirtualConnect FlexFabric supporting a total of 56 Gigabit, 64 10GE, and 80 8Gb/sec SAN ports.

Pricing for the VirtualSystem VS2 and VS3 configurations was not available at press time.

The Register explores the the three-headed beast that is VCE. Is it a start-up? A co-operative? Or what? And why is EMC losing so much money on the venture?

Is VCE, the VMware-Cisco-EMC entity set up to sell integrated virtualised server-networking-storage Vblocks, a startup? And how is it doing, given EMC’s revelation that VCE is costing it more than $40m a quarter?

VCE was originally set up as Acadia in November 2009, to integrate and sell Vblocks and provide an initial period of operational support for customers. It began life with a 130-employee headcount and funds from the three founding companies, plus Intel.

It changed its name to VCE – the Virtual Computing Environment company – in January this year, and supplies Vblocks as single products with a single SKU (stock keeping unit) through a network of partners, 130 or so of them, who get pre-built Vblocks. VCE says it was formed by Ciso and EMC with investments from VMware and Intel. We understand EMC owns 58 per cent of VCE and Cisco 35 per cent, leaving 7 per cent for VMware and Intel.

In January VCE had more than 100 customers with average Vblock deal size of $2.5m and a 40 per cent month by month growth rate, plus a $1.2bn pipeline. At that point El Reg estimated VCE accumulated sales revenue was $262.5m.

The EMC SEC filing in August revealed EMC’s accumulated VCE losses were $132m out of total contributed funding of $181.3m; that’s $173.5m cash and $7.8m in stock options. EMC’s VCE investment lost $41.9m lost in the March quarter and $46.6m in the June quarter. We observed that, if viewed as a startup, VCE did not appear that successful. This resulted in quite strong disagreement, as seen here for example.

Is VCE a startup?

Is VCE a startup? Not in the classic Silicon Valley sense, no.

A typical startup involves founders inventing a new product, building a prototype with their own or angel investor funding, getting a demonstrable technology with the prototype, gaining VC funding and building a product…

Their company is started up from scratch and staffed by people new to it. It develops the product, brings it to market, and then markets and sells it, with possible further rounds of funding to build sales and marketing and support operations as well as develop the product further. Eventually and hopefully the new company has a strong enough pipeline of business for profitability to be reliably coming or present and goes public, or is bought, with its VC backers cashing out.

Wikibon’s David Vellante stated: “EMC and Cisco spotted a trend and acted like VCs for a change — by creating a startup that could get to market fast.”

Mark Hopkins of Silicon Angle says VCE is a startup: “When we talked to Michael Capellas at EMC World in 2010, he characterised the work he was doing with Acadia, the predecessor to VCE, as very start-up-ish in nature. Later in 2011, when I was at SxSW in Austin this year, I met Jay Cuthrell [managing principal vArchitect at VCE] in the bloggers’ lounge, who without provocation extolled the virtues of the VCE startup culture.”

Hopkins had some strong words for El Reg. He said: “This is a world that the Register isn’t plugged into, nor are they usually responsible to their audience for understanding this. That combined with their tendency towards sensationalism, and it’s not surprising it wouldn’t even enter into their minds that this company indeed truly functions like a startup, nor would the thought to compare VCE’s progress at 18 months to other startups at 18 months… So, even when you evaluate VCE (purely as a startup) and compare them to their enterprise startup peers [like 3PAR], there’s absolutely no shame in not being profitable at 18 months.”

But VCE is not a startup in the Silicon Valley sense at all, I’d argue, being, in essence, a joint venture to integrate existing and successful products all with a great track record of success.

It was staffed by people who know the products and sold largely through a channel that knew the individual products too. In other words it did not have to develop substantial technology or IP at all, being merely a system integration exercise joint venture by, largely, Cisco and EMC.

Its startup expenses, compared to a Silicon Valley startup developing new hardware, should have been a lot, lot less. It could hit the ground running, selling bundles of known and successful products to customers who knew and trusted the products and their suppliers.

That’s hardly like 3PAR, a then unknown company, selling a brand new product and technology to customers who had to be convinced to take a big risk in buying it.

VCE numbers

VCE did not really have to develop any significant IP at all in fact, so its start costs were basically staffing and facilities.

VCE is now in the 18 months-plus after product launch phase, and we see that it is still costing EMC money, to the tune of $40m plus a quarter. Yet it has made revenues, we estimate, of $262.5m. Extracting EMC’s proportion of that, 58 per cent pro rata, gives us $152.25m.

So EMC has put in $181.3m, made $152.25m, leaving $29.05m, yet has accumulated losses of $132m and is doing so at a rate of $40m-plus per quarter. At this rate it will have to put in more funding, unless VCE sales rise significantly.

What has happened to the $152.25m of VCE revenues that EMC appears to be entitled to? How can it have lost $132m yet seen $152.25m come in? VCE must have been spending a lot of cash…

One source suggests total VCE losses are: “about $228m as of June 2011.”

This VAR guy goes on to say: “Barring a change of direction it sounds like the losses will continue. Back in [a] May 2011 SEC filing, Cisco indicated over the next 12 months, as VCE scales its operations, ‘[Cisco] expects that it will make additional investments in VCE and may incur additional losses, proportionate with the company’s ownership percentage’.”

The same would be true of EMC.

Wikibon’s David Vellante says: “I think VCE has spent heavily. It has had to restructure strategy and messaging to protect its partners. It has been spending money like crazy and has had to get more disciplined. It will continue to alter the operating model. But to conclude that VCE is failing is in my view way off base and at the very least grossly premature.”

As a comparison point, NetApp has its Flexpods, Vblock-like bundles that are, practically speaking, Vblocks with NetApp storage sold by NetApp’s channel. With no VE structure to support, we’d assume Flexpod sales return profits to NetApp whereas, as EMC’s SEC filing reveals, Vblocks don’t yet return a profit to EMC.

EMC’s view

An EMC view is that, SEC filings apart, VCE losses are not losses but expenses. Here’s EMC’s Chuck Hollis commenting on the VAR guy’s blog: “I understand what you’re trying to say here, but I think you’ve missed a fundamental point — the revenues associated with VCE flow back to the parent companies, and aren’t broken out separately. Put differently, the ‘losses’ you’re writing about are actually ‘expenses’. They’re offset by revenue that show up elsewhere. And none of the parent companies break out revenue to that level of detail.” …

We are left with the knowledge that EMC is asserting Vblock sales are rising steadily and there is strong customer interest. VCE is costing EMC money right now but EMC hopes and expects Vblock sales to reach billion dollar annual run rates in the “next several quarters” when it would then, obviously, be profitable.

Four quarters would take us to this time next year and we might expect profitability to be reached by then, and, possibly, unit ship numbers to be revealed or indicated.

If VCE were a Silicon Valley start-up its backers would probably never have countenanced such an ambitious infrastructure build-out and, assuming they had, would be scaling it back to stop it running out of cash.

Ironically there are rumours of VCE layoffs but these are, emphatically, rumours. Commenting on this point and saying he didn’t know about that, Hollis said: “Just about every IT vendor in the industry is continually refining their labour model these days — adding and subtracting to get the right mix. To do otherwise would be foolhardy.”

An interesting example of workload-optimization for virtualized environments using flash storage.

Startup Astute Networks stores VMware virtual machines on its dedicated ViSX G3 flash SAN – leaving bulk data to be stored on standard drive arrays.

Astute claims that “virtualised storage is too slow to support SAP, Oracle, and SQL databases; virtual machine performance is inadequate to meet Microsoft Exchange and SharePoint service levels and user mailbox loads; and backing up or recovering virtualised data stores is unacceptably slow.” Its ViSX G3 is its answer to these problems.

The ViSX G3 is a 3U box connecting to servers with iSCSI on one or 10 gig Ethernet. It comprises flash memory and an Astute network processor, and provide 80,000 IOPS for up to 64 virtual machines (VMs) which could each host a virtual desktop. These could be on one or more physical servers. The company says that VMs which require capacity-optimised data stores remain on disk-based storage.

Astute says the product has a DataPump Engine which includes the networks processor, custom hardware and software. It accelerates TCP/IP, virtualises iSCSI data store traffic, and provides RAID facilities to speed performance and protect flash contents…

The VMware-certified ViSX G3 boxes are managed, Astute says, as virtualised storage via vCenter or VMware-certified 3rd party software. Their use removes VM I/O from drive arrays, freeing up bandwidth for data access.

This is a novel approach to speeding up drive array I/O, and focusses exclusively on VMs that are I/O-bound. It is not a cache, unlike Avere and Alacritech filer accelerators, and neither is it a general, primary data, all-flash array like the Nimbus Data product. It is a highly specific, flash-based SAN for I/O-bound VMware virtual machines only.

The ViSX G3 1200 is priced at $29,000; the ViSX G3 2400 is at $59,000; and the ViSX G3 4800 costs $94,000. Options include extended warranty, expert support and on-site service, and are available now.

Beyond the headline, the interesting news here is that many company are deploying multiple hypervisors in their data centers.

VMware is the primary hypervisor for 58% of organizations that use x86 virtualization software, with Citrix and Microsoft’s Hyper-V splitting the rest of the market, a new survey finds.

Commissioned by virtualization management vendor Veeam and conducted by research group Vanson Bourne, the V-Index survey of more than 500 large businesses found that 92% have adopted virtualization, and deployed an average of 470 virtual machines. Although VMware is still the most widely used, it turns out multiple hypervisors in the same data center is fast becoming the norm. And with VMware’s latest price increases, Microsoft and Citrix have a chance to chip into its lead.

Fully 58% of companies using virtualization call VMware their primary hypervisor, while 20.2% put Citrix at the top of the list and 18.6% count Hyper-V first.

Yet when counting all virtualization use, 84% of virtualizing companies use VMware, 61% use Hyper-V, 55.4% use Citrix’s XenServer and 12% use other hypervisors. Only 3% use an “other” hypervisor as their main one.

Vanson Bourne surveyed IT decision makers at 544 organizations in the United States, United Kingdom, France and Germany, each of which has at least 1,000 employees. Exactly half of the companies use virtualization.

The results suggest that Microsoft, until recently an also-ran in the virtualization market, is making progress. But with VMware being counted as the main hypervisor for nearly 60% of customers, it seems VMware is still the top choice for running mission critical applications.

Veeam said it will update the survey every three months.

While Veeam’s survey provides a snapshot of today’s usage, IDC’s latest quarterly numbers show that customers are continuing to buy VMware products at a rapid pace.

In Q1 2011, worldwide sales of new virtualization licenses deployed into production on new or existing servers was 58% VMware, 26% Microsoft and 8% Citrix, with others accounting for the remainder, according to IDC.

Overall, 20% of new x86 servers have virtualization technology pre-installed or installed at the time of deployment, IDC also says. While Veeam’s numbers suggest that most enterprises are using virtualization, the IDC survey shows that doesn’t mean most of their servers are virtualized.

In fact, Veeam’s survey also showed that just under 40% of servers are virtualized, with an average consolidation ratio of six virtual servers per host.

Microsoft’s share could continue to increase because of anger among VMware customers over new licensing prices. In postings on a VMware community forum, one customer professed to be “totally floored” by the new prices, while another used the phrase “still in shock” and claimed to be emailing a VMware sales rep to complain.

VMware increased the number of licenses required to virtualize systems that meet certain thresholds of sockets and memory, causing one customer to call it a “penalty on density.”

“Our team prefers vSphere over Hyper-V, but with this new pricing we might be forced to switch,” one customer wrote.