Posts Tagged ‘consolidation’

HP/HP-UX/Itanium tops IBM/RHEL/x86

alinean2

A recently published whitepaper from Alinean VP and Senior Analyst, Paul Demopoulos took an in-depth look at HP-UX 11i on Itanium-based Integrity Servers vs. Red Hat Linux on IBM x86 Servers.

From the abstract:

“This paper analyzes the three year lifecycle TCO of two alternative platforms, considering the costs to plan, purchase, implement, manage, and use two comparable server configurations for a specified scenario, application, and workload. The comparisons in this study use HP-UX 11i, a UNIX solution hosted on an open system platform using HP Integrity servers (Intel® Itanium®-based servers with 4, 8 and 16 cores depending on function) versus Red Hat Linux running on IBM x86 servers and quantifies how open systems UNIX solutions running on Intel Itanium servers can deliver higher levels of manageability, consolidation, virtualization, adaptability, security, and availability. As a result of the analysis, HP-UX 11i running on HP Integrity servers provides substantially more benefits in business critical computing environments, delivering higher levels of consolidation to deliver total cost of ownership (TCO) savings of more than 20% compared to Red Hat Linux on IBM x86 servers.”

Get the whitepaper here.

Suiting Mission-Critical Apps

Editor’s Note: The Alliance again welcomes Jonathan Eunice, co-founder and principal IT advisor for Illuminata, to our blog. The complete series of four videos featuring Jonathan and Alliance president and executive director Joan Jacobs can be found here.

What do you need to run mission-critical apps? From a feature/function checklist perspective, the answers are pretty clear. They’re things like extremely high performance, rock-solid reliability, high-availability that doesn’t stop, the right application and support software, good value, and strong support.

See the latest and final video installment below, Safe & Sound

See the corresponding slide deck for “Safe & Sound” here.

But let’s look the requirements a little differently, a little more philosophically. What virtues are we looking for in our platforms for our most important applications? I like these:

Headroom – Mission-critical systems run mission-critical apps. They’re important. They take in money, or manage critical business processes. If they don’t perform well enough, everyone’s unhappy. Customers complain, and may balk. Executives become very cross. Bad things happen. In specifying a mission critical system, you want to ensure that there’s enough performance for the apps, services, and requirements that you have today—and those that are coming. If you get more customers, acquire a competitor, or have a massive promotion—whatever the eventuality, you want to be able to handle it with aplomb.

Safe and Sound – If you think customers and executives get cross when systems and applications perform poorly, just wait until those systems and applications go down. Unavailability is a profoundly bad outcome when the business is on the line.

Amidst and Among – You want whatever systems you source to work with the other systems, applications, and services you already have. The real world is heterogeneous, so it has to go along and get along. It has to integrate well—technically, culturally, and in every other way with the choices you’ve already made. Windows, Linux, Unix, whatever. This database vendor or that. This application or that. It must be a well-integrated into the world you have, and the one you’re constructing.

Big Iron is the architectural pattern that most clearly embodies these virtues for mission-critical databases, applications, and services, as well as for highly virtualized, highly consolidated infrastructures.

In a series of four video discussions, Alliance President and Executive Director Joan Jacobs and I have gone into more detail about how Itanium, an architecture born to run Big Iron systems and mission-critical apps, helps deliver these virtues.

SAP on Itanium

Two recent posts on the SAP Community Network blog offer some good insight on designing and running various computing environments on Itanium. Zoran Popovic is a Senior Systems Engineer at Hemofarm, an international pharmaceutical group based in Serbia. The content is geared toward those who own SAP on Itanium and Windows, and for those interested in SAP, Itanium and virtualization.

In “SAP, Linux, Virtualization and - Itanium …” Zoran talks about how his firm’s IT infrastructure, originally based on 10 Integrity servers with HP-UX, then Windows, has grown and evolved over the years. Today, five years later, their system is twice as large with a 1TB database. Zoran states:

“We never had any serious unplanned downtime or system failures, performance, reliability, availability and stability was predictable (apart from some OS problems with Microsoft MSCS and one short storage outage).”

Hemofarm’s SAP landscape is quite diverse and currently consists of: ERP systems, BI systems, CEN system, NW04, Solution Manager, EWA, monitoring, SAP routers, SAP Web Dispatchers, network printing servers, and different sandbox and other systems.
The main argument for Itanium system application is usage within OLTP systems and databases. In order to address low server utilization rates, Zoran suggests that consolidated OS environments be isolated and that is usually done through some form of virtualization.

In a follow up post “SAP, Linux, Virtualization and - Itanium … continued” he illustrates some new findings and results from the system and speaks to those interested in SAP with virtualization about moving from Windows or Unix to Linux.

Some approximate tests were run and improvised benchmarks with virtualized and bare metal SAP systems are posted for ERP, ERM, and ERC. The goal was to make a comparison with similar bare metal systems and different platforms — not making exact results comparable with some official tests.

About Itanium in the data center, Zoran States:

“If it’s about critical business environment and _not_ about best price/performance ratio or HPC, there is no good reason to change CPU architecture to other than Itanium.”

Zoran is interested in making further inquiries and welcomes any suggestions and recommendations both about benchmarking and this environment in general.

Read Zoran’s posts in full here:

SAP, Linux, Virtualization and - Itanium …
SAP, Linux, Virtualization and - Itanium … continued.

Illuminata Video Series Part II: Amidst & Among

The second video in a series of four featuring Jonathan Eunice, co-founder and principal IT advisor for Illuminata, and Joan Jacobs, Alliance president and executive director, can be viewed below. In this episode, Jonathan discusses the reality of diversified datacenters including the management of applications, costs, and risk. See the corresponding slide deck here.

Real Companies…Real Results!

In this new Alliance whitepaper featuring 8 of our 2009 Itanium Innovation Award finalists, read how real companies tackled mission-critical issues and achieved truly outstanding results using Itanium-based solutions. The companies profiled in this whitepaper represent diverse industries from around the globe including healthcare, energy, education and retail. They  have one important thing in common, their reliance on Itanium servers to successfully power their most mission-critical operations. If you are an enterprise customer evaluating your platform options for a mission-critical application, either a move from a legacy platform or a new deployment, this paper is a must read.