03-03-2011, 02:49 PM
presented by:
Meenakshi Satnami
Crusoe Processor,cse.ppt (Size: 1,022 KB / Downloads: 75)
Crusoe Processor : Outline
Mobile Computing Demands :
Mobile Application Compatibility
Performance Sufficiency
High Energy Efficiency
Peak Performance
Long Battery Life
Low Power Dissipation
Low Heat Generation
What Is Crusoe Processor :
x86 based CPU chip developed by Transmeta Corporation.
Designed specifically for meeting the mobile platform user’s requirements.
Strong performance with low power consumption.
Hybrid hardware-software design.
Simplicity of hardware – less number of transistors.
Places more of the processing burden on the software.
Can emulate other Instruction Set Architectures (ISA’s).
Comparision Of Crusoe And x86 Architecture :
Blue represents hardware part.
Yellow represents software layer.
Consists of an hardware engine logically surrounded by a software layer.
The engine is a very long instruction word (VLIW) CPU capable of executing up to 4 operations in each clock cycle.
VLIW’s native instruction set bears no resemblance to the x86 instruction set.
The software layer (Code Morphing Software) dynamically “morphs” x86 instructions into VLIW instructions.
The surrounding software layer gives x86 instructions the impression that they are running on x86 hardware.
Crusoe Processor Architecture :
VLIW Hardware :
2 integer units,1 floating point unit,a memory(load/store) unit,a branch unit.
Molecule : A long(64 or 128 bits) instuction word which may contain up to 4 RISC –like instructions, called atom.
All atoms within a molecule are executed in parallel, and the molecule format directly determines how atoms get routed to functional units.
• This greatly simplifies the decode and the dispatch hardware
The integer register file contains :
• 64 general purpose registers and 32 floating point working registers shadowed by 48 general purpose and 16 floating point registers.
• CMS allocates some registers to hold x86 state while others can contain state internal to the system, or can be used as temporary registers.
Structure Of A Molecule :
To keep the processor running at full speed, molecules are packed as fully as possible with atoms.
Superscalar Architecture :
This type of processor hardware is much more complex than Crusoe Processor’s simple VLIW engine.
Code Morphing Software (CMS) :
Dynamic binary translation of x86 ISA into Crusoe internal native VLIW format using software at run-time.
It resides in the ROM and is the first program to start execution when the processor boots.
The only program written directly for the VLIW engine is the CMS itself.
Completely invisible to operating system –looks like x86 hardware processor
CMS provides isolation.
Translations cached to avoid translator overhead on repeated execution.
Optimizes across x86 instruction boundaries to improve performance.
CMS : Caching
Translation cache resides in a separate memory slot that is inaccessible to x86 code.
Takes advantage of Locality Of Reference
Translation cache purposes (trace cache)
• A place to keep x86 code’s translations.
• A way to build longer code, that are better suited for optimizations.
After a block has been translated once, repeated execution “hits” in the translation cache and the hardware can then execute the optimized translation at full speed.
Successive executions of the translation invokes only the optimizer, not the translator.
CMS : Decoding & Scheduling
A traditional x86 translates each x86 instruction every time it is executed.
Code Morphing can translate an entire group of x86 instructions at once, creating a translation, whereas a superscalar x86 translates single instructions in isolation.
Transmeta’s software translates instructions once, saving the resulting translation in a translation cache. The next time the x86 code is executed, the system skips the translation step and directly executes the existing optimized translation.
Code Morphing approach can amortize the cost of translation over many executions, allowing it to use much more sophisticated translation and scheduling algorithms.
The translation software can optimize the generated code and potentially reduce the number of instructions executed in a translation.
Code Morphing Benefits :
Molecules explicitly encode the instruction-level parallelism, hence they can be executed by a simple VLIW engine.
• Hardware does not need to perform complex instruction reordering.
• Simplicity means fast and low-power design.
Processor upgrades are simplified.
• Software layer means that software developers don’t have to recompile programs.
• New hardware architecture only needs a new code morphing software
• Code morphing software can be upgraded independently into flash ROM.
Software Layer increases the perfomance.
LongRun Technology :
A clever and innovative approach to power and thermal management that does have the effect of scaling power consumption to software demands.
LongRun Technology is tightly coupled with Code Morphing Software.
LongRun Technology aims at :
• Adaptive Power Management
Dynamically reduce core processor power consumption to near-optimal levels in response to application workload requirements.
• Thermal Management
Intelligently adapts processor operation to system thermal environments.
Power Management :
LongRun provides CMS with the ability to adjust :-
• Processor core voltage (V).
• Clock frequency (f).
Dynamically adjust V and f depending on current application load on processor.
Produces cubic reductions in power consumption (Power = (C*f*(V^2))/2)
Conventional processors can only scale down power linearly by reducing f .
LongRun power management policies implemented within CMS.
Thermal Management :
Thermal management integrated into dynamic power management operating point policies.
Crusoe provides an integrated on-die thermal diode.
• Can be connected to an external temperature sensor and processor temperature monitored by system BIOS and application software.
Manages processor thermal environment by using frequency operating point shifts as a substitute for thermal throttling.
• Doesn’t force the processor to shift from active to sleep state to dissipate heat.
• Delivers higher performance at same die temperature, or same performance at lower die temperature.
Crusoe And P3 Comparision :
Pentium 3 and Crusoe running DVD movie.
Crusoe requires no active cooling.
Conclusion :
Revolutionary hybrid software-hardware design.
Reduced power consumption.
Low heat generation.
Efficient for mobile computing.
"Considering the complexity of the project, it is amazing how well it works, how fast it works, and how low-power it is. For the end-user, this is just a normal PC, but under the hood, it is a technological marvel."
--- Marc Fleischmann, Transmeta
Meenakshi Satnami
Crusoe Processor,cse.ppt (Size: 1,022 KB / Downloads: 75)
Crusoe Processor : Outline
Mobile Computing Demands :
Mobile Application Compatibility
Performance Sufficiency
High Energy Efficiency
Peak Performance
Long Battery Life
Low Power Dissipation
Low Heat Generation
What Is Crusoe Processor :
x86 based CPU chip developed by Transmeta Corporation.
Designed specifically for meeting the mobile platform user’s requirements.
Strong performance with low power consumption.
Hybrid hardware-software design.
Simplicity of hardware – less number of transistors.
Places more of the processing burden on the software.
Can emulate other Instruction Set Architectures (ISA’s).
Comparision Of Crusoe And x86 Architecture :
Blue represents hardware part.
Yellow represents software layer.
Consists of an hardware engine logically surrounded by a software layer.
The engine is a very long instruction word (VLIW) CPU capable of executing up to 4 operations in each clock cycle.
VLIW’s native instruction set bears no resemblance to the x86 instruction set.
The software layer (Code Morphing Software) dynamically “morphs” x86 instructions into VLIW instructions.
The surrounding software layer gives x86 instructions the impression that they are running on x86 hardware.
Crusoe Processor Architecture :
VLIW Hardware :
2 integer units,1 floating point unit,a memory(load/store) unit,a branch unit.
Molecule : A long(64 or 128 bits) instuction word which may contain up to 4 RISC –like instructions, called atom.
All atoms within a molecule are executed in parallel, and the molecule format directly determines how atoms get routed to functional units.
• This greatly simplifies the decode and the dispatch hardware
The integer register file contains :
• 64 general purpose registers and 32 floating point working registers shadowed by 48 general purpose and 16 floating point registers.
• CMS allocates some registers to hold x86 state while others can contain state internal to the system, or can be used as temporary registers.
Structure Of A Molecule :
To keep the processor running at full speed, molecules are packed as fully as possible with atoms.
Superscalar Architecture :
This type of processor hardware is much more complex than Crusoe Processor’s simple VLIW engine.
Code Morphing Software (CMS) :
Dynamic binary translation of x86 ISA into Crusoe internal native VLIW format using software at run-time.
It resides in the ROM and is the first program to start execution when the processor boots.
The only program written directly for the VLIW engine is the CMS itself.
Completely invisible to operating system –looks like x86 hardware processor
CMS provides isolation.
Translations cached to avoid translator overhead on repeated execution.
Optimizes across x86 instruction boundaries to improve performance.
CMS : Caching
Translation cache resides in a separate memory slot that is inaccessible to x86 code.
Takes advantage of Locality Of Reference
Translation cache purposes (trace cache)
• A place to keep x86 code’s translations.
• A way to build longer code, that are better suited for optimizations.
After a block has been translated once, repeated execution “hits” in the translation cache and the hardware can then execute the optimized translation at full speed.
Successive executions of the translation invokes only the optimizer, not the translator.
CMS : Decoding & Scheduling
A traditional x86 translates each x86 instruction every time it is executed.
Code Morphing can translate an entire group of x86 instructions at once, creating a translation, whereas a superscalar x86 translates single instructions in isolation.
Transmeta’s software translates instructions once, saving the resulting translation in a translation cache. The next time the x86 code is executed, the system skips the translation step and directly executes the existing optimized translation.
Code Morphing approach can amortize the cost of translation over many executions, allowing it to use much more sophisticated translation and scheduling algorithms.
The translation software can optimize the generated code and potentially reduce the number of instructions executed in a translation.
Code Morphing Benefits :
Molecules explicitly encode the instruction-level parallelism, hence they can be executed by a simple VLIW engine.
• Hardware does not need to perform complex instruction reordering.
• Simplicity means fast and low-power design.
Processor upgrades are simplified.
• Software layer means that software developers don’t have to recompile programs.
• New hardware architecture only needs a new code morphing software
• Code morphing software can be upgraded independently into flash ROM.
Software Layer increases the perfomance.
LongRun Technology :
A clever and innovative approach to power and thermal management that does have the effect of scaling power consumption to software demands.
LongRun Technology is tightly coupled with Code Morphing Software.
LongRun Technology aims at :
• Adaptive Power Management
Dynamically reduce core processor power consumption to near-optimal levels in response to application workload requirements.
• Thermal Management
Intelligently adapts processor operation to system thermal environments.
Power Management :
LongRun provides CMS with the ability to adjust :-
• Processor core voltage (V).
• Clock frequency (f).
Dynamically adjust V and f depending on current application load on processor.
Produces cubic reductions in power consumption (Power = (C*f*(V^2))/2)
Conventional processors can only scale down power linearly by reducing f .
LongRun power management policies implemented within CMS.
Thermal Management :
Thermal management integrated into dynamic power management operating point policies.
Crusoe provides an integrated on-die thermal diode.
• Can be connected to an external temperature sensor and processor temperature monitored by system BIOS and application software.
Manages processor thermal environment by using frequency operating point shifts as a substitute for thermal throttling.
• Doesn’t force the processor to shift from active to sleep state to dissipate heat.
• Delivers higher performance at same die temperature, or same performance at lower die temperature.
Crusoe And P3 Comparision :
Pentium 3 and Crusoe running DVD movie.
Crusoe requires no active cooling.
Conclusion :
Revolutionary hybrid software-hardware design.
Reduced power consumption.
Low heat generation.
Efficient for mobile computing.
"Considering the complexity of the project, it is amazing how well it works, how fast it works, and how low-power it is. For the end-user, this is just a normal PC, but under the hood, it is a technological marvel."
--- Marc Fleischmann, Transmeta