|
|
| Date Posted:
7/21/2008 11:48:48 PM
Status:
Live |
|
|
Asker's Rating:
None Provided
Moderator's Rating: Somewhat Helpful
|
|
Response:
hii frnd,
ya offcourse i also here for ur help..
1)an computer engineering, computer architecture is the conceptual design and fundamental operational structure of a computer system. It is a blueprint
and functional description of requirements (especially speeds and
interconnections) and design implementations for the various parts of a
computer — focusing largely on the way by which the central processing unit (CPU) performs internally and accesses addresses in memory.........
2)computer system architecture is
presented. A single numeric processor and multiple interactive
processors are used to provide sustained compute performance in numeric
applications. The numeric processors' unique directed dataflow
architecture supports the parallelization of a much broader range of
algorithms than vector processor architectures. Applications can be
ported to the Cydra 5 and achieve very high performance with
significantly less reprogramming than on alternative architectures. The
numeric processor requires significantly less application reprogramming
to make efficient use of its architecture to accelerate typical numeric
applications. The numeric processor parallelizes programs with
recurrences, conditionals within loops, unstructured memory reference,
and other difficult-to-vectorize program constructs...........
3)
Abstract
A computer system includes a bus system; a pluggable central processing
unit circuit board, coupled to the bus system; a pluggable logic board
coupled to the pluggable central processing unit circuit board through the
bus system; a pluggable input/output board coupled to the pluggable logic
board through the bus system; a first connector unit for directly
connecting the pluggable central processing unit circuit board to a first
predetermined location on the pluggable logic board; and a second
connector unit for directly connecting the pluggable logic board to a
predetermined location on the pluggable input/output board............
4)
Abstract:
Terascale computing is commonplace in today's HPC world. For example, at
LLNL, users have a choice of five separate terascale systems. With ASCI
Purple, BlueGene/L, and a 11 TF Linux cluster, LLNL will continue this
trend into the foreseeable future. We have found that this unprecedented
degree of parallelism exposes performance limitations in new and existing
applications. However, analyzing and predicting performance at this scale
is challenging for several reasons. First, performance analysis techniques
must strike the appropriate balance between instrumentation resolution and
overhead. Second, users and architects must gain insight from these
potentially massive datasets. Third, to predict application performance,
designers need efficient simulation strategies for these large, complex
datasets.
To this end, we propose three new solutions for these challenges,
respectively. First, we introduce a novel message sampling technique for
MPI applications that reduces instrumentation overhead dramatically and
allows runtime analysis of performance data. Second, we help users distill
massive performance datasets by using decision tree classification, a
supervised machine learning technique, to classify the performance of an
application's individual communication operations. Third, to accelerate
the practice of performance prediction, we use tracing to capture
high-level communication and computation behaviors, and then, we use a
trace-driven simulator to experiment with the architectural design space.
Experimental results from numerous applications on systems demonstrate
that our new solutions can improve the process of performance analysis.
i hope it will help u....
|
|
|
|