Sie sind auf Seite 1von 2

Intel’s introduction of Hyper-Threading technology represents a significant improv

ement in processor utilization and performance. This technology boosts system pe


rformance without going to a higher clock rate or adding more processors. This i
mprovement is achieved by making multiple instruction streams, called threads, i
nternally available to a single processor at the same time. These threads allow
the processor the opportunity to better schedule the use of internal resources a
nd improve utilization.
HP servers offer this new technology by making use Intel® Xeon processors, which i
ncorporate Hyper-Threading. This technology brief describes the Hyper-Threading
concept, as well as its benefits and limitations for the user. Some HP performan
ce test results are also included to show the improvement seen by the use of Hyp
er-Threading.
Introduction
Hyper-Threading technology is a groundbreaking innovation from Intel that enable
s multi-threaded server software applications to execute threads in parallel wit
hin each processor in a server platform. The Intel® Xeon™ processor family uses Hype
r-Threading technology, along with the Intel® NetBurst™ micro architecture, to incre
ase computer power and throughput for today’s Internet, e-Business, and enterprise
server applications. This level of threading technology has never been seen bef
ore in a general-purpose microprocessor. Hyper-Threading technology helps increa
se transaction rates, reduces end-user response times, and enhances business pro
ductivity providing a competitive edge to e-Businesses and the enterprise. The I
ntel® Xeon™ processor family for servers represents the next leap forward in process
or design and performance by being the first Intel® processor to support thread-le
vel parallelism on a single processor.
With processor and application parallelism becoming more prevalent, today’s server
platforms are increasingly turning to threading as a way of increasing overall
system performance. Server applications have been threaded (split into multiple
streams of instructions) to take advantage of multiple processors. Multi-process
ing-aware operating systems can schedule these threads for processing in paralle
l, across multiple processors within the server system. These same applications
can run unmodified on the Intel® Xeon™ processor family for servers and take advanta
ge of thread-level-parallelism on each processor in the system. Hyper-Threading
technology complements traditional multi-processing by offering greater parallel
ism and performance headroom for threaded software.
Hyper-Threading
This new technology from Intel enables multi-threaded applications to execute th
reads in parallel on each individual processor. Available on Intel Xeon processo
rs, Hyper-Threading provides the user with increased computing power to meet the
needs of today’s server applications.
Need for the technology
Improving processor utilization has been an industry goal for years. Processor s
peeds have advanced until a typical processor today can run at frequencies over
2 gigahertz, but much of the rest of the system is not capable of running at tha
t speed. To enable performance improvements, memory caches have been integrated
into the processor to minimize the long delays that can result from accessing ma
in memory. Xeon processors, for example, now include three cache levels on the d
ie.
Large server-based applications tend to be memory intensive due to the difficult
y of predicting access patterns. The working data sets are also quite large. The
se two things can create bottlenecks, regardless of memory prefetching technique
s. Latency due to these bottlenecks only gets worse when pointer-intensive appli
cations are executed. Any mistake in prediction can force a pipeline to be clear
ed, incurring a delay to refill this data.
It is this latency that drives processor utilization down. Despite improvements
in application development and parallel processing implementations, reaching hig
her utilization rates remained an unmet goal.
What is Hyper-Threading??
Hyper-Threading Technology enables one physical processor to execute two separat
e threads at the same time. To achieve this, Intel designed the Xeon processor w
ith the usual processor core, but with two Architectural State devices. Each Arc
hitectural State (AS) tracks the flow of a thread being executed by core resourc
es.
After power-up and initialization, these two internal Architectural States defin
e two logical processors. Individually they can be halted, interrupted, or can e
xecute a specific thread independently of the other logical processor. Each AS h
as an instruction pointer, advanced programmable interrupt controller (APIC) reg
isters, general-purpose registers, and machine state registers.
The two logical processors then share the remaining physical execution resources
. An application or operating system (OS) can submit threads to two different lo
gical processors just as it would in traditional multiprocessor system.
Operating Systems and Applications
A system with processors that use Hype Threading
Technology appears to the operating system and application software as having tw
ice the number of
Processors than it physically has. Operating systems manage logical processors a
s they do physical processors, scheduling run able tasks or threads to logical p
rocessors. However, for best performance, the operating system should implement
two optimizations
The first is to use the HALT instruction if one logical processor is active and
the other is not. HALT will allow the processor to transition to either the ST0-
or ST1-mode. An operating system that does not use this optimization would exec
ute on the idle logical processor a sequence of instructions that repeatedly che
cks for work to do. This so-called “idle loop” can consume significant execution res
ources that could otherwise be used to make faster progress on the other active
logical processor.
The second optimization is in scheduling software threads to logical processors.
In general, for best threads to logical processors on different physical
processors before scheduling multiple threads to the
Same physical processor. This optimization allows
Software threads to use different physical execution
Resources when possible.
Benefits of Hyper-Threading Technology
Hyper-Threading technology can result in many benefits to e-Business and the ent
erprise:
• Improved reaction and response times for end-users and customers
• Increased number of users that a server system can support
• Handle increased server workloads
• Higher transaction rates for e-Businesses
• Greater end-user and business productivity
• Compatibility with existing server applications and operating systems
• Headroom to take advantage of enhancements offered by future software releases
• Headroom for future business growth and new solution capabilities

Reference: http://www.seminarprojects.com/Thread-hyperthreading-download-the-sem
inar-report#ixzz1L4BZR4q9

Das könnte Ihnen auch gefallen