by Hang-sheng Wang, Xinping Zhu, Li-shiuan Peh, Sharad Malik
in Proceedings of 35th Annual International Symposium on Microarchitecture
http://www.ee.princeton.edu/~peh/publications/orion.pdf
Add To MetaCart
Abstract:
With the prevalence of server blades and systems-ona-chip (SoCs), interconnection networks are becoming an important part of the microprocessor landscape. However, there is limited tool support available for their design. While performance simulators have been built that enable performance estimation while varying network parameters, these cover only one metric of interest in modern designs. System power consumption is increasingly becoming equally, if not more important than performance. It is now critical to get detailed power-performance tradeoff information early in the microarchitectural design cycle. This is especially so as interconnection networks consume a significant fraction of total system power. It is exactly this gap that the work presented in this paper aims to fill. We present Orion1, a power-performance interconnection network simulator that is capable of providing detailed power characteristics, in addition to performance characteristics, to enable rapid power-performance tradeoffs at the architectural-level. This capability is provided within a general framework that builds a simulator starting from a microarchitectural specification of the interconnection network. A key component of this construction is the architectural-level parameterized power models that we have derived as part of this effort. Using component power models and a synthesized efficient power (and performance) simulator, a microarchitect can rapidly explore the design space. As case studies, we demonstrate the use of Orion in determining optimal system parameters, in examining the effect of diverse traffic conditions, as well as evaluating new network microarchitectures. In each of the above, the ability to simultaneously monitor power and performance is key in determining suitable microarchitectures. 1 Orion is one of the most powerful (brightest) network of stars (constellation) in the sky. It also stands for Open Research Infrastructure for Optimizing Networks. 1
Citations
|
577
|
Wattch: A Framework for Architectural-Level Power Analysis and Optimizations
– Brooks, Tiwari, et al.
- 2000
|
|
361
|
Virtual-channel flow control
– Dally
- 1992
|
|
245
|
Route packets, not wires: on-chip interconnection networks,” DAC - Design Automation Conference
– Dally
- 2001
|
|
220
|
The torus routing chip
– Dally, Seitz
- 1986
|
|
218
|
An enhanced access and cycle time model for on-chip caches
– Wilton, Jouppi
- 1994
|
|
104
|
The design and use of simplepower: a cycle-accurate energy estimation tool
– Ye, Vijaykrishnan, et al.
- 2000
|
|
89
|
A static power model for architects
– Butts, Sohi
- 2000
|
|
63
|
Microarchitectural exploration with Liberty
– Vachharajani, Vachharajani, et al.
- 2002
|
|
57
|
The SP2 High-performance Switch
– Stunkel, Shea, et al.
- 1995
|
|
49
|
Addressing the System-on-a-Chip Interconnect Woes through Communication-Based Design,'' Design Automation Conference
– Sgroi, Sheets, et al.
- 2001
|
|
39
|
The energy complexity of register files
– Zyuban, Kogge
- 1998
|
|
38
|
A delay model and speculative architecture for pipelined routers
– Peh, Dally
- 2001
|
|
35
|
The Alpha 21364 network architecture
– Mukherjee, Bannon, et al.
|
|
33
|
Interconnect Architecture Exploration for Low-Energy Reconfigurable SingleChip DSPs
– Zhang, Wan, et al.
|
|
27
|
Power constrained design of multiprocessor interconnection networks
– Patel, Chai, et al.
- 1997
|
|
26
|
Analysis of power consumption on switch fabrics in network routers
– Ye, Benini, et al.
- 2002
|
|
24
|
Pipelined memory shared buffer for VLSI switches
– Katevenis, Vatsolaki, et al.
- 1995
|
|
22
|
Value-Based Clock Gating and Operation Packing: Dynamic Strategies for Improving Processor Power and Performance
– Brooks, Martonosi
- 2000
|
|
20
|
Analytical Energy Dissipation Models for Low-Power Caches
– Kamble, Ghose
- 1997
|
|
18
|
A hierarchical modeling framework for on-chip communication architectures
– Zhu, Malik
- 2002
|
|
12
|
Low-power area-efficient high-speed I/O circuit techniques
– Lee, Dally, et al.
- 2000
|
|
12
|
A Hardware Architecture for Dynamic Performance and Energy Adaptation
– STANLEY-MARBELL, HSIAO, et al.
- 2002
|
|
12
|
A Power Model for Routers: Modeling Alpha 21364
– Wang, Peh, et al.
- 2002
|
|
7
|
Power-efficient interconnection networks: Dynamic voltage scaling with links
– Shang, Peh, et al.
- 2002
|
|
3
|
InfiniBand 8-port 12x switch
– IBM
|
|
2
|
Alpha 21364 (EV7). http://www.eecs.umich. edu/vlsi seminar/f01/slides/bannon.pdf
– Bannon
|
|
1
|
PowerCompiler datasheet. http://www. synopsys.com/products/power/power bkg.html
– Corp
|
