Posts Tagged Windows

Open Source Virtualization Saves Money

Virtualization is a powerful technology that enables numerous benefits detailed here, particularly saving money. The savings come mainly through the IT resource consolidation. When you add open source to the virtualization equation, it creates another avenue to savings.

 Virtualization technology, in the form of the hypervisor, is not exactly cheap. VMware, the industry’s 900lbs. hypervisor gorilla commands significant license fees. With its latest pricing plan introduced last year, the standard license starts at about $1000. Enterprise costs are based on processor sockets and memory and, given how it is calculated, VMware can require four times as many licenses as previously needed, which dramatically increases the cost.  Here’s VMware’s FAQ on pricing. Depending on the amount of memory licensing costs could run into the tens of thousands of dollars.

Open source virtualization, noted Jean Staten Healy, IBM’s worldwide Cross-IBM Linux and Open Virtualization Director, presents opportunities to reduce virtualization costs in numerous ways.  For example, the inclusion of open source KVM in enterprise Linux distributions reduces need for additional hypervisors, enabling the organization to avoid buying more VMware licenses.  KVM also enables higher virtual machine density for more savings. IDC’s Al Gillen and Gary Chen put out a white paper detailing the recent KVM advances.

 The ability to manage mixed KVM- VMware virtualization through a single tool further increases the cost efficiency of open source virtualization. IBM’s System Director VMControl is one of the few tools providing such mixed hypervisor, cross-platform management.  For general hypervisor management, Linux and KVM standardized on libvirt and libguestfs as the base APIs for managing virtualization and images. These APIs work with other Linux hypervisors beyond KVM (higher-level tools, such as virsh and virt-manager, are built on top of libvirt).

The combination of KVM technical advances, the slow but steadily increasing adoption of KVM, and the inclusion of KVM as a core feature of the Linux operation system is driving more enterprises deploy KVM along with VMware. Of course, the fact that KVM now comes for free as part of the Linux core means you can try it at no cost and minimal risk.

Enterprise Linux users are now using KVM where they previously would have not bothered to virtualize a particular workload due to cost. This makes sense for several reasons; free being just one of them. Other reasons include the integration of the KVM toolset with the Linux toolset and the fact that Linux admins already know how to use it.

One large bank used Linux and KVM as a development and test resource in a private cloud. Normally, they would have needed to request more budget for VMware but since they had Linux with KVM they could just add Windows virtual machines. And by developing in Java, they could roll out prototypes fast.  In the process, the bank achieved high virtual machine density at minimal cost.

Another financial services firm set up virtual machines with KVM to monitor Linux usage for under-utilized hosts and then deployed virtual machine images to the host as warranted. The result: an ad-hoc grid of KVM virtual machines with high utilization, again at minimal cost.

KVM is a natural for private clouds. IBM reports private clouds being built using Moab with xCAT and KVM. The resulting private cloud handles both VMware and KVM equally well, making them plug-compatible.  With this approach, organizations can gradually expand their use of KVM and reduce, or at least delay, the need to buy more VMware licenses, again saving money.

KVM also is being exercised in a big way as the hypervisor behind IBM’s public Smart Cloud Enterprise, demonstrating how enterprise-capable this free, open source hypervisor is.

BottomlineIT expects VMware will remain the dominant x86 virtualization platform going forward. However, it makes sense to grab every opportunity to use KVM for enterprise-class, multi-platform virtualization and save money wherever you can.


, , , , ,

Leave a comment

New IBM z114—a Midrange Mainframe

IBM introduced its newest mainframe in the zEnterprise family, the z114, a business class rather than enterprise class machine. With the z114IBM can now deliver a more compelling total cost of acquisition (TCA) case, giving midrange enterprises another option as they consolidate, virtualize, and migrate their sprawling server farms. This will be particularly interesting to shops running HP Itanium or Oracle/Sun servers.

The z114 comes with a $75,000 entry price. At this price, it can begin to compete with commodity high end servers on a TCA basis, especially if it is bundled with discount programs likeIBM’s System z Solution Editions and unpublicized offers from IBM Global Finance (IGF). There should be no doubt, IBM is willing to deal to win midrange workloads from other platforms.

First, the specs, speeds, and feeds:  the z114 is available in two models; a single-drawer model, the M05, and a two-drawer model, the M10, which offers additional capacity for I/O and coupling expansion and/or more specialty engines. It comes with up to 10 configurable cores, which can be designated as general purpose or specialty engine (zIIP, zAAP, IFL, ICF) or used as spares. The M10 also allows two dedicated spares as well, a first for a midrange mainframe.

The z114 uses a superscalar design that runs at 3.8 GHz, an improved cache structure, a new out-of-order execution sequence, and over 100 new hardware instructions that deliver better per-thread performance, especially for database, WebSphere, and Linux workloads. The base z114 starts at 26 MIPS but can scale to over 3100 MIPS across five central processors and the additional horsepower provided by its specialty engines.

The z114 mainly will be a consolidation play. IBM calculates that workloads from as many as 300 competitive servers can be consolidated onto a single z114. IBM figures the machine can handle workloads from 40 Oracle server cores using just three processors running Linux. And compared to the Oracle servers IBM estimates the new z114 will cost 80% less. Similarly, IBM figures that a fully configured z114 running Linux on z can create and maintain a Linux virtual server for approximately $500 per year.

As a consolidation play, the zEnterprise System will get even more interesting later this year when x86 blades supporting Windows become available. Depending on the pricing, the z114 could become a Windows consolidation play too.

Today even midrange enterprises are multi-platform shops. For this, the z114 connects to the zBX, a blade expansion cabinet, where it can integrate and manage workloads running on POWER7-based blades as well as the IBM Smart Analytics Optimizer and WebSphere DataPower blades for integrating web-based workloads. In addition, IBM promises support for Microsoft Windows on select System x server blades soon.

To achieve a low TCA, IBM clearly is ready to make deals. For example, IBM also has lowered software costs to deliver the same capacity for 5-18% less through a revised Advanced Workload License Charges (AWLC) pricing schedule.  A new processor value unit (PVU) rating on IFLs can lower Linux costs as much as 48%.

The best deal, however, usually comes through the System z Solution Edition Program, which BottomlineIT’s sister blog, DancingDinosaur, has covered here and here.  It bundles System z hardware, software, middleware, and three years of maintenance into a deeply discounted package price. Initial System Editions for the z114 will be WebSphere, Linux, and probably SAP.

IFG also can lower costs, starting with a six month payment deferral. You can acquire a z114 now but not begin paying for it until the next year. The group also is offering all IBM middleware products, mainly WebSphere Application Server and Tivoli, interest free (0%) for twelve months. Finally, IFG can lower TCA through leasing. Leasing could further reduce the cost of the z114 by up to 3.5% over three years.

By the time you’ve configured the z114 the way you want it and netted out the various discounts, even with a Solutions Edition package, it will probably cost more than $75,000. Even the most expensive HP Itanium server beats that. As soon as there are multiple servers in a consolidation play, that’s where the z114 payback lies.

, , , , , , , , , , , , , , , , , , , , ,

Leave a comment

Open Source KVM Takes on the Hypervisor Leaders

The hypervisor—software that allocates and manages virtualized system resources—usually is the first thing that comes to mind when virtualization comes up. And when IT considers server virtualization the first option typically is VMware ESX, followed by Microsoft’s Hyper-V.

But that shouldn’t be the whole story. Even in the Windows/Intel world there are other hypervisors, such as Citrix Xen.  And IBM has had hypervisor technology for its mainframes for decades and for its Power systems since the late 1990s. A mainframe (System z) running IBM’s System z hypervisor, z/VM, can handle over 1000 virtual machines while delivering top performance and reliability.

So, it was significant when IBM announced in early May that it and Red Hat, an open source technology leader, are working together to make products built around the Kernel-based Virtual Machine (KVM) open source hypervisor for the enterprise. Jean Staten Healy, IBM’s Director of Worldwide Cross-IBM Linux, told IT industry analysts that the two companies together are committed to driving adoption of the open source virtualization technology through joint development projects and enablement of the KVM ecosystem.

Differentiating this approach from those taken by the current x86 virtualization leaders VMware and Microsoft is open source technology. An open source approach to virtualization, Healy noted, lowers costs, enables greater interoperability, and increases options through multiple sources.

The KVM open source hypervisor allows a business to create multiple virtual versions of Linux and Windows environments on the same server. Larger enterprises can take KVM-based products and combine them with comprehensive management capabilities to create highly scalable and reliable, fully cloud-capable systems that enable the consolidation and sharing of massive numbers of virtualized applications and servers.

Red Hat Enterprise Virtualization, for example, was designed for large scale datacenter virtualization by pairing its centralized virtualization management system and advanced features with the KVM hypervisor. BottomlineIT looked at the Red Hat open source approach a few weeks ago, here.

The open source approach to virtualization also is starting to gain traction. To that end Red Hat, IBM, BMC, HP, Intel, and others joined to form the Open Virtualization Alliance. Its goal is to facilitate  the adoption of open virtualization technologies, especially KVM. It intends do this by promoting examples of customer successes, encourage interoperability, and accelerate the expansion of the ecosystem of third party solutions around KVM. A growing and robust ecosystem around KVM is essential if the open source hypervisor is to effectively rival VMware and Microsoft.

Healy introduced what might be considered the Alliance’s first KVM enterprise-scale success story, IBM’s own Research Compute Cloud (RC2), which is the first large-scale cloud deployed within IBM. In addition to being a proving ground for KVM, RC2 also handles actual IBM internal chargeback based on charges-per-VM hour across IBM. That’s real business work.

RC2 runs over 200 iDataplex nodes, an IBM x86 product, using KVM (90% memory utilization/node). It runs 2000 concurrent instances, is used by thousands of IBM employees worldwide, and provides 100TB of block storage attached to KVM instances via a storage cloud.

KVM was chosen not only to demonstrate the open source hypervisor but because it was particularly well suited to the enterprise challenge. It provides a predictable and familiar environment that required no additional skills, auditable security compliance, and the open source licensing model that kept costs down and would prove cost-effective for large-scale cloud use, which won’t be long in coming. The RC2 team, it seems, already is preparing live migration plans for support of federated clouds. Stay tuned.

, , , , , , , , , , , , , , , ,

1 Comment

Run Windows on the Mainframe

Last week IBM published a general statement of direction that essentially said the zEnterprise will have Windows in the 4th quarter via an x blade for the zBX. The announcement reads, in part, “IBM intends to offer select IBM System x blades running Microsoft Windows in the IBM zEnterprise BladeCenter Extension Model 002.”

The zBX is the extension cabinet for the new zEnterprise/z196 mainframe and for the older z10. When the product is released later this year organizations will be able to run native Windows applications through their mainframe.

This isn’t exactly new. IBM has been hinting about this since the zEnterprise announcement last July although it refrained from making any statement that might legally commit them to anything. I have been writing about it for about as long. Here is a Windows on z piece from earlier this year.

Windows on the System z (or the zBX as is the case here) raises some interesting issues. Of immediate interest is the question of which Windows workloads might be best moved to the zEnterprise. Microsoft Exchange and Microsoft SharePoint, probably the company’s two most popular enterprise server applications, are good bets to stay where they are now.

Similarly, the Microsoft Office Suite will stay where it is unless someone wants to attempt to virtualize a slew of Office desktops on a handful of zBX IBM x blades as a VDI play. IBM, however, is pushing VDI separately as a cloud workload.

In response to which workloads, IBM’s reply is as follows: enterprises with one or more applications running in a complex, heterogeneous, multi-tiered environment have the opportunity to upgrade that infrastructure with zEnterprise and enjoy the management benefits that the Unified Resource Manager brings. The most likely candidates are those Windows apps that make use of data or processing residing the z, which is IBM’s basic fit-for-purpose strategy.

IBM also is not expecting a wholesale migration of Windows servers to the z. Again the IBM spokesman explains: Many of the largest data centers already have far more blades and/or rack-mounted [Windows] servers in their inventory than could realistically fit into the zBX.

What IBM does hope for, according to the spokesman, are workloads that rely on System z for data serving and contain other application components running on System z, Power, or Intel that are required to complete the end-to-end business process. Companies with such applications running in these two- or three-tier environments may be ideal candidates for zEnterprise running the new System x blades.

Beyond these basic fit-for-purpose statements IBM isn’t saying much about blade pricing, performance, or software licensing. This serves two purposes for IBM: 1) it avoids binding promises to which it might be held, and probably as importantly, 2) it freezes the competitive market. Multi-platform, multi-tier enterprises considering an upgrade of their Windows servers to the latest offerings from HP or Oracle (Sun) or others now know this IBM option is coming in the fourth quarter and may wait until they see the pricing, licensing, and performance details.

With the System x blade and Windows IBM indeed can make a strong case for re-centralizing on the zEnterprise, especially for multi-tier, multi-platform enterprises. This case will be based on the zEnterprise’s centralized management efficiency, the potential for greater optimization, and the resulting performance improvements. It’s a good story that could catch on if IBM keeps the pricing competitive.

But a more intriguing question: is it enough to attract NON-mainframe shops to the zEnterprise/zBX? Maybe, especially if IBM offers compelling discounts as it has been known to do.

, , , , , , , , , , ,

Leave a comment

High End Servers—the Software Differentiator

Some may argue that given the recent tumult in the high end server market there is no high end server available today that can match the IBM zEnterprise (z196/zBX combo) in processing power, reliability, and scalability. It holds its own in terms of speeds and feeds, number of cores, memory. Software, however, may turn out to be the biggest differentiator for high end servers, and IBM has optimized a ton of software for its platforms, something the others mainly just talk about.

The high end server market has suddenly entered a period of change. In March Oracle announced that will no longer support Itanium processors. HP immediately countered with a statement of support for Itanium. SGI announced a 256-core Xeon Windows system. Also in March, Quanta Computer, a Chinese operation, reported squeezing 512 cores into a pizza box server running the Tilera multi-core processor.  Tilera’s roadmap goes out to 2013 when it expects to pack 200 cores onto a processor.  Of course, IBM launched the first hybrid server, the zEnterprise consisting of the multi-core z196 coupled with the zBX last summer.

This recent flurry of server activity at the large-scale, multi-core end of the market leaves server buyers somewhat confused. One wrote to BottomlineIT: What will be the ultimate retail price per core?  What’s the current price per core of, perhaps, a chassis full of 8-core IBM System p blades, an HP Superdome, or an SGI UV 1000 running Windows or Linux?

Fair questions, for sure. The published OEM price last year was $900 per chip for a 64-core Tilera processor, which rounds to $14 per core. SGI reports that the Altix UV starts at $50,000 with Microsoft software an additional $2,999 per four sockets (32 cores). A buyer could end up facing different vendors and technologies competing at the $50, $100, $500, $1000, $5,000 and $10,000 per core price points. Each vendor will be promoting a different architecture, configuration, memory optimization, performance, and even form factor (multi u, pizza box, blades) attributes.

This is not just about price but integration, internal communication speeds, optimization, and more. At this point all the vendors need to be more forthcoming and transparent.

But this may not turn out to be a hardware, processor, memory, speeds and feeds battle. It may not even turn into a price-per-core battle or a total cost of ownership (TCO) vs. total cost of acquisition (TCA) battle. Ultimately, it has to come down to workloads supported and delivered, and that means software. And when it comes to workload optimization and software IBM already has an advantage, especially when compared to Oracle and HP. Remember when Oracle bought Sun CEO Larry Ellison said it would optimize Oracle software for SPARC? Have you seen much yet?

A quick peek at IBM’s software lineup suggests the company has a lot of topnotch software to run on its hardware, most of it optimized for its hardware platforms.  Factor in the ISV ecosystem and the IBM picture gets even better.

Let’s start with Gartner naming IBM the worldwide market share leader overall in the application infrastructure and middleware software segment.  If you drill down into the various submarkets, IBM often comes up as leader there too. For example, IBM leads the business process management (BPM) market, with better than double the share of its closest competitor. IBM also leads in the message oriented middleware market, the transaction processing monitor market, and the combined markets for Enterprise Service Bus (ESB) and integration appliances.

Critical segments for sure, but businesses need more. For that IBM offers DB2, a powerful enterprise database management system that can rival Oracle. WebSphere goes far beyond being just an application server; it encompasses a wide range of functionality including portals and commerce. With Rational, IBM can cover the entire application development lifecycle, and with Lotus IBM nails down communication and collaboration. And don’t forget Cognos, a powerful BI tool, plus all the IBM Smart Analytics tools. Finally, IBM provides the Tivoli product set to manage systems and storage.

The point: when it comes to high end servers it is not just about processor cores. It’s about systems optimized for the software you need to run your workloads.


, , , , , , , , , , , , , , , , ,

Leave a comment