Massive growth in servers has brought about increases in TCO (Total ownership cost) which in turn have caused further cost blow-outs from overly complicated administration. This burden of administration and maintenance fees on TCO has also made companies hesitate to strategically invest for new business.
The result is many enterprises are working on this problem as a business challenge under the banner of "TCO reduction"

Here we look at Server Consolidation using PRIMEPOWER, with case studies, and a great scale-up ^(note) consolidation solution.
Consolidation of your entire system onto PRIMEPOWER not only enables TCO optimization but also ensures high performance and high availability. It also provides you with the scalability and efficiency of use needed to accommodate new business opportunities.

Note) Scale-up: Scale-up operation improves performance by increasing the number of CPUs, memory, and disks, etc., to multiple services on the same server. Multiple OS instances on the one server individually control each operation to provide scale up consolidation. Centralizing multiple operations onto PRIMEPOWER means high service reliability and availability. It also provides the advantages of reduced administration and administration costs as server numbers are significantly reduced. Although initial invest is higher compared with “scale-out” purchase of more servers, long term business and cost benefits accrue from the consolidated management and resource optimization of scale up operation.

Server consolidation whereby scattered systems are integrated into one scalable server enables you to reduce lifecycle and business costs across the entire system as well as and opportunities to grow your business.

Total system lifecycle cost reduction

By integrating your servers into one administrative unit you can also reduce the numbers of CPUs required. Case studies show that customers who have done this can achieve 30% reductions in lifecycle costs through savings in administration, maintenance and software license fees of fewer CPUs.

Alleviation of performance deterioration and business delays

Differences in workloads, between night and day or different seasons mean your system capacity must increase to match the required workload. But as it is hard to predict, systems have been commonly divided along individual business process lines. This means separate sizing for peak-time workloads, leading to excess resourcing across your organization.
PRIMEPOWER provides you with dynamic resource reallocation on demand. This means your spare system resources can be switched to meet the needs of heavily loaded business processes. By sharing resources across several businesses you will reduce the hoarding of resources by any one service.
Fujitsu can show you how customers have succeeded in consolidating their business application and batch application servers to put the differences in peak-time requirements to practical use.

Lessons learnt from Server Consolidation case studies   

PRIMEPOWER’s availability and scalability design enables the following system solutions to be implemented:

• Integration of different sized systems onto one server
  Hardware partitions can be flexibly configured from a minimum size of 1 CPU.
• Flexible system reconfiguration
  Server resources including CPUs and memory can be dynamically allocated to any required system.
• Peak-time preparation not required
  Server resources can be shared across multiple business processes. The wasteful requirement of needing extra resources per system is not longer necessary.
• Individual system failure is completely isolated from all other systems
  Each partition acts as a totally intact server. The multiple business processes therefore operate on isolated hardware, where applications, OS, and firmware do not affect other business services.
  

PRIMEPOWER optimized for server consolidation  

PRIMEPOWER Line-up

PRIMEPOWER is the result of Fujitsu’s long experience in high speed and high reliability technology garnered from long-term mainframe, super computer and UNIX development know-how.
The world’s first 90-nanometer processor, SPARC64™ V, made PRIMEPOWER a world record holder in major benchmarks such as Specjbb® 2005 (*). Monitoring by check circuits throughout the system and with separate management processors; business continuity is protected even in case of hardware problems.
PRIMEPOWER900, 1500, and 2500 deliver practical scalability by providing the partitioning and DR (Dynamic Reconfiguration) functions essential to server consolidation.
This state of the art technology is complemented by industry standard compatibility, a myriad of ISV/IHV solutions, and substantial regional configuration and testing centers.

(*) As of 18th Oct., 2006, PRIMEPOWER 25000 holds No.1 record of SPECjbb® 2005 in JVM/bops measurement.

For more information on server consolidation please contact your local Fujitsu Office. No other UNIX server solution vendor in today’s market can provide better guaranteed service levels, lower TCO and support for your enterprises’ evolving IT requirements.
Fujitsu is ready to help you streamline your most demanding computing environment and free-up valuable resources that will allow you to focus on more strategic business issues.