Multi-core processing
Stanford researcher outlines three techniques in the labs
Rick Merritt
EE Times
(06/18/2009 12:38 PM EDT)
SAN JOSE, Calif. — Help is on the way for programmers trying to wrestle greater performance from
multicore processors, but the big leaps forward in parallel programming are still in the labs, according to
a keynoter at the Multicore Virtual Conference.
Processor vendors are planning support for thread-level speculation and transactional memory as two
ways assist developers in getting more parallelism. But it will require a new generation of domain-specific
languages still in an early research phase before the average programmer can harness the parallelism
in tomorrow's many-core chips, said Kunle Olukotun who directs the Pervasive Parallelism Lab at
Stanford.
Both Intel and Sun aim to support transactional memory in hardware in future processors, said Olukotun
in his keynote. "The question is how the software will develop to take advantage of transactional memory
hardware," he said.
Pervasive Parallelism Lab's Director
Kunle Olukotun
The technique, originally developed by database specialists, essentially groups multiple memory
accesses into a single mega-transaction that is executed—or aborted—as a single task. With hardware
support, the technique can deliver as much as a seven-fold performance improvement for some
applications compared to software-only implementations, he said.
But "transactional memory doesn't solve all the problems," he said. "We still haven't addressed how we
find independent tasks, map them to threads and optimize the locality and communications [of data] to
provide predictable and scalable performance," added Olukotun, a professor of electrical engineering
and computer science at Stanford.
To meet those goals, researchers are working on domain-specific languages that provide a level of
abstraction. The languages would let programmers focus on defining applications-level jobs that the
languages translate into parallel tasks.
SQL, MatLab, Ruby on Rails and OpenGL serve similar functions today for their targeted user groups.
Olukotun said he hopes his lab can deliver a platform as early as 2012 to help create such tools.
"We want to make [parallel programming] practical for the masses of programmers," he said.
He also discussed thread-level speculation, a near term technique for reducing the need to synchronize
tasks running in parallel. With hardware support, the technique can deliver significant performance
gains, he said, showing results of an automated translator of x86 binary code into parallel software.
Some chip makers will support the technique within two or three years, he predicted. However, the
technique is limited by its inability to scale.
Mapou
Contract software development
commented on Jun 18, 2009 4:06:41 PM
The recent reports that Sun Microsystems cancelled Rock, its costly multicore processor project, does
not bode well for the processor industry. The industry just suffered a painful blow. Prof. Olukutun may
delude himself into believing that multithreading and hardware transactional memory are part of the
future of parallel processing but all he is doing is prolonging the suffering. When the pain becomes
unbearable (it's all about money), it will suddenly dawn on everybody in the business that it is time to
force the baby boomers (the Turing Machine worshippers) into retirement in order to break away from
the hopelessly flawed computing models of the 20th century. Prof Olukutun is not doing computer
science a favor by clinging to the old bankrupt paradigms. Rest assured that the Rock is not the last of
the big chip failures. Don't be surprised to see Intel's Larrabee and AMD's Fusion projects come
crashing down like so many Hindenburgs. How much more pain can we take? It's time for the industry to
wake up. Unfortunately, Prof. Olukutun is not helping. He's a product of the old school and rocking the
boat is not on his agenda. Too bad because, sooner or later, that is exactly what is going to happen.
Mapou
Contract software development
commented on Jun 18, 2009 4:11:36 PM
One more thing. Those who are truly interested in finding a solution to the parallel programming crisis
should go to the following link. Prof. Olukutum and the experts at Stanford's Pervasive Parallelism lab
need not bother. http://rebelscience.blogspot.com/2008/07/how-to-solve-parallel-programming.html
http://bits.blogs.nytimes.com/2009/06/15/sun-is-said-to-cancel-big-chip-project/
June 15, 2009, 3:46 pm
Sun Is Said to Cancel Big Chip Project
By Ashlee Vance
Sun Microsystems may have dropped a bit of weight by the time Oracle officially acquires the company.
According to two people briefed on Sun’s plans, the company has canceled its Rock chip project, putting
an end to one of its biggest revitalization bets.
Sun has been working on the Rock project for more than five years, hoping to create a chip with many
cores that would trounce competing server chips from I.B.M. and Intel. The company has talked about
Rock in the loftiest of terms and built it up as a game-changing product. In April 2007, Jonathan
Schwartz, the chief executive of Sun, bragged about receiving the first test versions of Rock.
But the two people familiar with Sun’s plans say Rock has met with an unceremonious end. The people
requested anonymity, as they are not authorized to speak with the press about Sun’s plans.
Michelle Parkinson, a Sun spokeswoman, said the company had no comment.
Rock was at one point scheduled to ship last year, but Sun has delayed it a number of times, and over
the past few months company insiders have complained about various glitches hurting the product.
This marks the second high-end chip in a row that Sun has canceled before its release. These types of
products cost billions of dollars to produce, and Sun now has about a 10-year track record of investing
in game-changing chips that failed to materialize.
Sun has been relying on chips from Fujitsu for its larger servers while it waited for the Rock development
to be finished. Now it is likely to just continue using Fujitsu chips, which should lower research and
development costs. That’s probably good news for Oracle, which is in the process of acquiring Sun.
Sun does still design a line of chips for its smaller servers, and those products have proved popular in
recent years.
Next
Index
Stanford researcher outlines three techniques in the labs
Rick Merritt
EE Times
(06/18/2009 12:38 PM EDT)
SAN JOSE, Calif. — Help is on the way for programmers trying to wrestle greater performance from
multicore processors, but the big leaps forward in parallel programming are still in the labs, according to
a keynoter at the Multicore Virtual Conference.
Processor vendors are planning support for thread-level speculation and transactional memory as two
ways assist developers in getting more parallelism. But it will require a new generation of domain-specific
languages still in an early research phase before the average programmer can harness the parallelism
in tomorrow's many-core chips, said Kunle Olukotun who directs the Pervasive Parallelism Lab at
Stanford.
Both Intel and Sun aim to support transactional memory in hardware in future processors, said Olukotun
in his keynote. "The question is how the software will develop to take advantage of transactional memory
hardware," he said.
Pervasive Parallelism Lab's Director
Kunle Olukotun
The technique, originally developed by database specialists, essentially groups multiple memory
accesses into a single mega-transaction that is executed—or aborted—as a single task. With hardware
support, the technique can deliver as much as a seven-fold performance improvement for some
applications compared to software-only implementations, he said.
But "transactional memory doesn't solve all the problems," he said. "We still haven't addressed how we
find independent tasks, map them to threads and optimize the locality and communications [of data] to
provide predictable and scalable performance," added Olukotun, a professor of electrical engineering
and computer science at Stanford.
To meet those goals, researchers are working on domain-specific languages that provide a level of
abstraction. The languages would let programmers focus on defining applications-level jobs that the
languages translate into parallel tasks.
SQL, MatLab, Ruby on Rails and OpenGL serve similar functions today for their targeted user groups.
Olukotun said he hopes his lab can deliver a platform as early as 2012 to help create such tools.
"We want to make [parallel programming] practical for the masses of programmers," he said.
He also discussed thread-level speculation, a near term technique for reducing the need to synchronize
tasks running in parallel. With hardware support, the technique can deliver significant performance
gains, he said, showing results of an automated translator of x86 binary code into parallel software.
Some chip makers will support the technique within two or three years, he predicted. However, the
technique is limited by its inability to scale.
Mapou
Contract software development
commented on Jun 18, 2009 4:06:41 PM
The recent reports that Sun Microsystems cancelled Rock, its costly multicore processor project, does
not bode well for the processor industry. The industry just suffered a painful blow. Prof. Olukutun may
delude himself into believing that multithreading and hardware transactional memory are part of the
future of parallel processing but all he is doing is prolonging the suffering. When the pain becomes
unbearable (it's all about money), it will suddenly dawn on everybody in the business that it is time to
force the baby boomers (the Turing Machine worshippers) into retirement in order to break away from
the hopelessly flawed computing models of the 20th century. Prof Olukutun is not doing computer
science a favor by clinging to the old bankrupt paradigms. Rest assured that the Rock is not the last of
the big chip failures. Don't be surprised to see Intel's Larrabee and AMD's Fusion projects come
crashing down like so many Hindenburgs. How much more pain can we take? It's time for the industry to
wake up. Unfortunately, Prof. Olukutun is not helping. He's a product of the old school and rocking the
boat is not on his agenda. Too bad because, sooner or later, that is exactly what is going to happen.
Mapou
Contract software development
commented on Jun 18, 2009 4:11:36 PM
One more thing. Those who are truly interested in finding a solution to the parallel programming crisis
should go to the following link. Prof. Olukutum and the experts at Stanford's Pervasive Parallelism lab
need not bother. http://rebelscience.blogspot.com/2008/07/how-to-solve-parallel-programming.html
http://bits.blogs.nytimes.com/2009/06/15/sun-is-said-to-cancel-big-chip-project/
June 15, 2009, 3:46 pm
Sun Is Said to Cancel Big Chip Project
By Ashlee Vance
Sun Microsystems may have dropped a bit of weight by the time Oracle officially acquires the company.
According to two people briefed on Sun’s plans, the company has canceled its Rock chip project, putting
an end to one of its biggest revitalization bets.
Sun has been working on the Rock project for more than five years, hoping to create a chip with many
cores that would trounce competing server chips from I.B.M. and Intel. The company has talked about
Rock in the loftiest of terms and built it up as a game-changing product. In April 2007, Jonathan
Schwartz, the chief executive of Sun, bragged about receiving the first test versions of Rock.
But the two people familiar with Sun’s plans say Rock has met with an unceremonious end. The people
requested anonymity, as they are not authorized to speak with the press about Sun’s plans.
Michelle Parkinson, a Sun spokeswoman, said the company had no comment.
Rock was at one point scheduled to ship last year, but Sun has delayed it a number of times, and over
the past few months company insiders have complained about various glitches hurting the product.
This marks the second high-end chip in a row that Sun has canceled before its release. These types of
products cost billions of dollars to produce, and Sun now has about a 10-year track record of investing
in game-changing chips that failed to materialize.
Sun has been relying on chips from Fujitsu for its larger servers while it waited for the Rock development
to be finished. Now it is likely to just continue using Fujitsu chips, which should lower research and
development costs. That’s probably good news for Oracle, which is in the process of acquiring Sun.
Sun does still design a line of chips for its smaller servers, and those products have proved popular in
recent years.
Next
Index
 | |  | |  | |  |