Unix to Linux Migration

Are you frustrated with the price/performance of proprietary UNIX solutions? Are you seeking to take advantage of the multi-vendor, low-cost, high performance environment provided by Intel's x86 hardware and Red Hat Enterprise Linux? By migrating from UNIX to Red Hat Enterprise Linux, you can:

• Improve application performance, for enhanced response times for customer care and business critical operations.
• Lower the total cost of IT operations, by replacing proprietary UNIX systems with low-cost, commodity Intel hardware.
• Leverage existing IT skills you already have, because of the close technology kinship of UNIX and Linux.
• Enhance system security and scalability on an Intel platform.

Red Hat, Intel and our global ecosystem of ISV and hardware partners can provide you with comprehensive solutions for migrations from UNIX to Red Hat Enterprise Linux.

The economic downturn of 2001 and 2002 changed how customers and IT managers
think about selecting and configuring their server solutions. Integral to this shift is
increased interest in low capital-cost alternatives such as Linux servers – which are
gaining considerable market share as shipment rates grew at double-digit rates yearover-
year throughout 2002 and into Q103. The strong growth in this market segment
is in turn accelerating the industry’s acceptance of new suppliers and new
technologies.
Although significant attention has been paid to Linux solutions, IT organizations
continue to invest heavily in Unix/RISC servers. The Unix/RISC installed base
represents many billions of investment that were made over a period of many years.
During the late 1990s, annual spending for Unix/RISC servers averaged more than
$20 billion per year. In 2002, Unix/RISC accounted for 38.6% of the worldwide server
revenue, yet only 11.0% of the shipments. This reflects the higher average sales
prices for scalable Unix servers in the midrange and high-end price bands.
IT organizations today are being pressed to freeze or reduce their annual budgets for
new hardware acquisitions, ongoing maintenance and operations, and personnel
costs. In the case of Unix/RISC servers, customers often are seeking more
economical alternatives for the necessary capabilities and performance. Increasingly,
customers are finding the solution in the form x86 servers and “scale out” clusters
running Linux.
The programming and operational similarities of Linux and Unix are well-known.
Although the operating system kernels were built independently, the surrounding
utilities and shells often are similar or identical. This leads to application-development
and deployment environments that are similar enough to allow existing IT staff to
transfer workloads from one platform to another — providing that the application code
and software skill sets are available. Thus, the differences between these two
computing environments do not preclude Linux servers from being deployed to
support workloads that once were the sole province of Unix servers.
Differences do exist, including fewer off-the-shelf packaged applications for Linux;
less scalability on any individual server node in Linux servers; and higher use of Linux
for "edge", Web, and HPC workloads than for traditional business workloads.
However, IDC believes that these aspects of the Linux server ecosystem will change
quickly in coming years, making Linux servers more entrenched in the overall IT
environment.

WHY MAKE THE MOVE TO L INUX?
O P E R A T I O N A L B E N E F I T S A N D I S S U E S O F
U N I X / L I N U X M I G R A T I O N
Over the past few years, Linux has emerged as a credible alternative solution to
traditional Unix system deployments, particularly for basic infrastructure workloads
traditionally found behind the scenes at businesses, governmental organizations, and
universities. Linux was intended from its introduction in the early 1990s to deliver a
Unix-like experience, and because Linux draws on many of the same utilities and
applications used in Unix, the parallels between these two environments, for
programmers and for system administrators, are remarkably strong. Early
deployments often took the form of a basic network service, such as a print server,
file server, Web server, or firewall system
These early workloads generally were not dependent upon layered infrastructure
software such as transaction processing software, database software or application
server software, and were typically invisible to end users and executive management.
If an organization’s needs were to outgrow the capabilities of the Linux system, or
should the Linux prove unsuitable to the workload assigned, a Unix system could
easily be substituted with little fanfare.
After proving itself in these less glamorous roles, Linux now is increasingly being
seen as a solution suitable for more demanding business workloads. The introduction
of commercial software ranging from databases, Web application servers,
middleware, infrastructure software and management tools, it is apparent that Linux
now offers a software stack that can be a credible alternative to Unix systems.
In particular, the move of Oracle and other vendors to port its database products to
Linux is exactly the kind of industry support that lends credibility to the power of the
open-source software movement. By offering equal or higher capabilities along with
better price/performance than competitive technologies, users can explore the
possibility of migrating elements of their installed base Unix/RISC servers, over time,
to more cost-effective volume server systems (servers priced less than $25,000) and
to midrange enterprise server systems (servers priced from $25,000 to $499,999).
ECONOMICS
Past IDC studies on total cost of ownership (TCO) consistently found that the most
significant cost item associated with a 3- or 5-year TCO calculation is the IT staffing
cost. Considering the similarity between Linux and Unix software environments, it is
not surprising that hourly staffing costs for full-time equivalent (FTE) administrative
personnel are similar. However, these IDC studies have also found that for some
workloads, Linux servers can have a lower TCO, especially for Web-enabled
workloads and infrastructure workloads. These differences often are the result of less
complicated – and therefore less costly – configurations on the part of the Linux
systems.