26-11-2010, 10:32 AM
[attachment=7295]
Presented By:Anil Kumar
What is Stream Computing?
Stream Computing based on parallel processing, generally used for 2D or 3D graphics applications, to solve real-world problems.
Stream computing uses Software Algorithms that analyzes the data in real time.
It increase speed and accuracy when dealing with data handling and analysis.
A model that uses sequences of data and computation kernels to increase efficiency of concurrency.
Evolution of Stream Computing
In June 2007, IBM announced its stream computing system, called System S.
System S
Runs on 800 Microprocessor.
Enables Application to split up tasks and then reassemble the data.
ATI Technologies
Describes technology that enables the graphics processors (GPUs).
Work with high-performance.
Solve complex computational problems.
Applications that run on the GPU instead of a CPU.
Stream Processing
It is a computer programming related to SIMD (Single Instruction Multiple Data).
In Stream processing, for a given a set of data (a stream), a series of operations (kernel functions) are applied to each element in the stream.
Where we require Stream Processing?
Stream processing is especially suitable for applications that exhibit three application characteristics.
Compute Intensity:
The Large number of arithmetic operations per I/O.
More algorithmic complexity in process.
Data Parallelism :
Exists in a kernel and the same function is applied to all records of an input stream.
Number of records can be processed simultaneously without waiting for results from previous records.
Data Locality :
locality is common.
Intermediate data within kernel functions.
Why Stream Processing?
Stream Processing enables…
High message/data rates,
Low (msec-secs) latency,
Advanced analysis
Its Applications in :-
Embedded systems
Cell phones, handheld computers.
Desktop workstations
Streaming media, real-time encryption.
High-performance servers
Radar tracking, HDTV editing consoles.
Limitations
Enable GPU processor cores to communicate, synchronize, and share data.
Other Hardware device required for stream computing.
Presented By:Anil Kumar
Stream Computing
What is Stream Computing?
Stream Computing based on parallel processing, generally used for 2D or 3D graphics applications, to solve real-world problems.
Stream computing uses Software Algorithms that analyzes the data in real time.
It increase speed and accuracy when dealing with data handling and analysis.
A model that uses sequences of data and computation kernels to increase efficiency of concurrency.
Evolution of Stream Computing
In June 2007, IBM announced its stream computing system, called System S.
System S
Runs on 800 Microprocessor.
Enables Application to split up tasks and then reassemble the data.
ATI Technologies
Describes technology that enables the graphics processors (GPUs).
Work with high-performance.
Solve complex computational problems.
Applications that run on the GPU instead of a CPU.
Stream Processing
It is a computer programming related to SIMD (Single Instruction Multiple Data).
In Stream processing, for a given a set of data (a stream), a series of operations (kernel functions) are applied to each element in the stream.
Where we require Stream Processing?
Stream processing is especially suitable for applications that exhibit three application characteristics.
Compute Intensity:
The Large number of arithmetic operations per I/O.
More algorithmic complexity in process.
Data Parallelism :
Exists in a kernel and the same function is applied to all records of an input stream.
Number of records can be processed simultaneously without waiting for results from previous records.
Data Locality :
locality is common.
Intermediate data within kernel functions.
Why Stream Processing?
Stream Processing enables…
High message/data rates,
Low (msec-secs) latency,
Advanced analysis
Its Applications in :-
Embedded systems
Cell phones, handheld computers.
Desktop workstations
Streaming media, real-time encryption.
High-performance servers
Radar tracking, HDTV editing consoles.
Limitations
Enable GPU processor cores to communicate, synchronize, and share data.
Other Hardware device required for stream computing.