2013-02 Heterogeneous Computing Paper 2013-02 CUDA Hardware Architecture 2013-02 CUDA Software Programming 2013-02 Advances in Kepler K20 2013-02 High Level Compilation

This article provides a big picture description of what high performance computing (HPC) is, who are doing it, what they do with it, and where are we positioned in this big picture.  We use a superfast train as the icon to express the speed of a HPC system.

What is HPC?

HPC is Parallel Computing which aggregates the computing power of multiple units of machines.  There are two approaches for parallel computing.  The first is homogeneous based on multiple units of the same type of processors. The second is heterogeneous based on a combination of CPU and GPU. 

Who does HPC?

There is an organisation http://www.top500.org/ that compiles data on the top 500 performance systems in the world.  This organisation publishes the Top 500 list every 6 months and measures system performance with a hardware benchmarking application called LINPACK.  LINPACK is an open source application and a standard, and anyone can download it to measure the performance of his own system. 

The single most important performance indicator produced by LINPACK is FLOPS standing for floating point operations per second.  This is done on Double Precision (16 digits) basis as against Single (7 digits).  On the Top 500 list, we can see the ranking of systems based solely on FLOPS.  The list shows also the number of cores in the system and the total electricity power consumed by the system.  We are able to rank systems based on FLOPS per core and FLOPS per Watt.

What is HPC for?

HPC is for problems that take a time period to compute longer than we desire.  The faster the HPC system the more desirable would it be.  Space, price and electricity consumption are the restraints.  Such problems exist both in the academic/research world as well as in the commercial world.  Compucon uses the terms Digital Concept Creation and Digital Content Creation to describe the 2 main camps of applications.   

SuperMUC, Top 4 in the world as of June 2012, is being used to develop Cave Automated Virtual Environment for example.  “Pirates of the Caribbean” was produced with HPC system amidst a US$180 million budget as another example.  AMBER taking 1 day to simulate 64ns of bio-molecular behaviour is the 3rd. 

Where is Compucon Positioned?

Compucon could be the first local hardware technology company in New Zealand that has systematically attempted to master the eco-systems for heterogeneous parallel computing along the Nvidia CUDA direction.

For more public information on GPU-based HPC solutions, please click.

For general server and other HPC solutions please explore here.

2013-02 Heterogeneous Computing Paper 2013-02 CUDA Hardware Architecture 2013-02 CUDA Software Programming 2013-02 Advances in Kepler K20 2013-02 High Level Compilation

This article provides a big picture description of what high performance computing (HPC) is, who are doing it, what they do with it, and where are we positioned in this big picture.  We use a superfast train as the icon to express the speed of a HPC system.

What is HPC?

HPC is Parallel Computing which aggregates the computing power of multiple units of machines.  There are two approaches for parallel computing.  The first is homogeneous based on multiple units of the same type of processors. The second is heterogeneous based on a combination of CPU and GPU. 

Who does HPC?

There is an organisation http://www.top500.org/ that compiles data on the top 500 performance systems in the world.  This organisation publishes the Top 500 list every 6 months and measures system performance with a hardware benchmarking application called LINPACK.  LINPACK is an open source application and a standard, and anyone can download it to measure the performance of his own system. 

The single most important performance indicator produced by LINPACK is FLOPS standing for floating point operations per second.  This is done on Double Precision (16 digits) basis as against Single (7 digits).  On the Top 500 list, we can see the ranking of systems based solely on FLOPS.  The list shows also the number of cores in the system and the total electricity power consumed by the system.  We are able to rank systems based on FLOPS per core and FLOPS per Watt.

What is HPC for?

HPC is for problems that take a time period to compute longer than we desire.  The faster the HPC system the more desirable would it be.  Space, price and electricity consumption are the restraints.  Such problems exist both in the academic/research world as well as in the commercial world.  Compucon uses the terms Digital Concept Creation and Digital Content Creation to describe the 2 main camps of applications.   

SuperMUC, Top 4 in the world as of June 2012, is being used to develop Cave Automated Virtual Environment for example.  “Pirates of the Caribbean” was produced with HPC system amidst a US$180 million budget as another example.  AMBER taking 1 day to simulate 64ns of bio-molecular behaviour is the 3rd. 

Where is Compucon Positioned?

Compucon could be the first local hardware technology company in New Zealand that has systematically attempted to master the eco-systems for heterogeneous parallel computing along the Nvidia CUDA direction.

For more public information on GPU-based HPC solutions, please click.

For general server and other HPC solutions please explore here.

Time is a reference framework, a resource, and a question we have no answer of.  Every person has a limited span of time in life and so a limited amount of what each can do or achieve.  When two persons or two machines over two different places want to meet, time will be needed as a criterion to synchronize.  We have heard that people adjusted their clocks when a train in Japan or Germany arrived in the train station.  This implies that trains in Japan and Germany were very punctual and their arrivals were used for synchronization.

The Definition of Time Act of London in 1888 was among the first legislations in the world on time.  The Greenwich Royal Observatory was a first official party to implement official time.  Unfortunately scientific communities did not think Greenwich Mean Time was accurate enough for the world and various other observatories have published their own time.

The US Naval Observatory has been very serious on time and done a lot of work to improve timeliness.  The latest we read about USNO was that its time was produced by over 30 caesium beams and 13 hydrogen masers.  As many as 70 other national observatories around the world have been engaging in similar activities.  Fortunately they all submit their work to the Bureau of Weights and Measures based in France for collating into an International Atomic Time (TAI), and further publishing as the Coordinated Universal Time (UTC) for real life use to include consideration of the Earth’s Rotation. We have added 25 leap seconds since 1 January 1972 when UTC was first established and up to June 2012.  When a leap second is added, the time would read HR:MIN:60 which does not normally exist.

A second is a defined unit of measurement.  The original definition is the time taken by the Earth to rotate 1/86400 of one cycle which consists of 24 hours of 60 minutes of 60 seconds. We add leap seconds to adjust laboratory atomic measurements to produce UTC, and leap years to produce the official solar calendar. 

NTP (Network Time Protocol) allows computers around the world to synchronize with the USNO time over the Internet.  USNO has a discrepancy of up to 26ns with UTC, but the Internet transmission delay has a magnitude of milliseconds.

Omega was the official time keeper for the Olympics Games in 1932 and the last one in London in 2012 among other years.  Time keepers used palm and mechanical based stop watches up to split second accuracy in 1932.  Now time keeping is not palm based and is electronic and atomic based and has a resolution of 1us.  This is 0.000001 second!

Sunlight takes about 8 minutes to reach Earth.  We can use a Time Space 2-dimensional diagram to denote this scenario.  This is a scientific interpretation of time and an illustration of the imaginative capabilities of human brains.

Does time have a beginning and an end?  If a big bang is the outcome of a big crunch and vice versa, then the answer is no.  What would happen if there is a beginning or an end of time?