HP NetServer LP 2000r 8-way Systems Enable Enterprise Applications - Page 3

Issues to Consider

Page 3 highlights

FXVWRPHUýXVDJHýJXLGH +3ý 1HW6HUYHUý /;Uýåèíí Issues to Consider 9JCV"MKPF"QH"FCVC"GPXKTQPOGPV"GZKUVU CV"VJKU"HCEKNKV[! Is the facility centralized or decentralized? Is it consolidating the servers? What application(s) does it support and are they processor intensive, I/O intensive or network intensive? What are the performance requirements? 9JCV"MKPF"QH"ITQYVJ"KU"GZRGEVGF! Has the organization done capacity planning? Are more servers anticipated? Already ordered? 9JCV"CTG"VJG"CXCKNCDKNKV["TGSWKTGOGPVU! What does it cost the company if the servers are down? What is the level of availability currently on this system and in this facility and how much should it improve? As part of the accumulation and understanding of costs of downtime, it is important to perform a complete risk assessment. Has the organization looked at disaster recovery? Has it addressed the potential for growth and the means for accommodating increased system usage? Scalability means the ability of a server system to expand to meet the growth in the enterprise. This concept is also called headroom and can be a difficult demand to meet. Every significant corporate computing trend of the past decade-LANs, client/server computing and especially the Internet-has required more processing power and more bandwidth than initially forecasted. The Intel Profusion technology provides CPU scalability to eight processors, memory scalability to 32 GB and I/O scalability with ten PCI slots on four 64-bit PCI buses. A server's ability to support an upgrade to future processor generations is one method that will increase the load carrying capacity of a system. Increased processing power will permit the server to support more users and/or more demanding applications. The ability to upgrade other components, such as controllers and storage, is another way to accommodate growth. The LXr 8500 follows the HP NetServer history of providing investment protection in the form of upgrades, for example, to future Intel Pentium Xeon processors. Client/Server Application Architectures The client/server application architecture is a versatile, message-based, modular infrastructure intended to improve interoperability, flexibility and scalability as compared to more centralized, mainframe-style, timesharing computing. The client/server method of structuring and distributing hardware and software has shown great benefits in the operational effectiveness of business information systems. This architecture is widely used with many off-the-shelf application software packages. There are three major functions that the software performs in the client/server architectural model. These functions, summarized here, reside in different pieces of equipment, depending upon the structure. 1. Presentation - the user interface 2. Business Logic - the application data processing 3. Data Management - A database stores and retrieve information Two common hardware deployment configurations are as a centralized environment (see Figure 1) two-tier client/server and a more distributed environment, three-tier client/server (see Figure 2). 3

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15

FXVWRPHU XVDJH JXLGH
3
+3
1HW6HUYHU
/;U °L±±
Scalability
means the ability of a server system to
expand to meet the growth in the enterprise. This
concept is also called headroom and can be a difficult
demand to meet. Every significant corporate computing
trend of the past decade—LANs, client/server computing
and especially the Internet—has required more
processing power and more bandwidth than initially
forecasted. The Intel Profusion technology provides CPU
scalability to eight processors, memory scalability to
32 GB and I/O scalability with ten PCI slots on four 64-bit
PCI buses.
A server’s ability to support an upgrade to future
processor generations is one method that will increase
the load carrying capacity of a system. Increased
processing power will permit the server to support more
users and/or more demanding applications. The ability to
upgrade other components, such as controllers and
storage, is another way to accommodate growth. The
LXr 8500 follows the HP NetServer history of providing
investment protection in the form of upgrades, for
example, to future Intel Pentium Xeon processors.
Client/Server Application Architectures
The client/server application architecture is a versatile,
message-based, modular infrastructure intended to
improve interoperability, flexibility and scalability as
compared to more centralized, mainframe-style, time-
sharing computing. The client/server method of
structuring and distributing hardware and software has
shown great benefits in the operational effectiveness of
business information systems. This architecture is widely
used with many off-the-shelf application software
packages.
There are three major functions that the software performs in the client/server architectural
model. These functions, summarized here, reside in different pieces of equipment,
depending upon the structure.
1.
Presentation - the user interface
2.
Business Logic - the application data processing
3.
Data Management - A database stores and retrieve information
Two common hardware deployment configurations are as a centralized environment (see
Figure 1) two-tier client/server and a more distributed environment, three-tier client/server
(see Figure 2).
Issues to Consider
9JCV MKPF QH FCVC GPXKTQPOGPV GZKUVU
CV VJKU HCEKNKV[!
Is the facility centralized or decen-
tralized? Is it consolidating the servers?
What application(s) does it support and
are they processor intensive, I/O
intensive or network intensive? What are
the performance requirements?
9JCV MKPF QH ITQYVJ KU GZRGEVGF!
Has the organization done capacity
planning? Are more servers anticipated?
Already ordered?
9JCV CTG VJG CXCKNCDKNKV[ TGSWKTGOGPVU!
What does it cost the company if the
servers are down? What is the level of
availability currently on this system and
in this facility and how much should it
improve?
As part of the accumulation and
understanding of costs of downtime, it is
important to perform a complete risk
assessment. Has the organization
looked at disaster recovery? Has it
addressed the potential for growth and
the means for accommodating increased
system usage?