Presentation is loading. Please wait.

Presentation is loading. Please wait.

Reducing Power and Area by Interconnecting Memory Controllers to Memory Ranks with RF Coplanar Waveguides on the Same Package WEED 2011, ISCA Mario D.

Similar presentations


Presentation on theme: "Reducing Power and Area by Interconnecting Memory Controllers to Memory Ranks with RF Coplanar Waveguides on the Same Package WEED 2011, ISCA Mario D."— Presentation transcript:

1 Reducing Power and Area by Interconnecting Memory Controllers to Memory Ranks with RF Coplanar Waveguides on the Same Package WEED 2011, ISCA Mario D. Marino, Kevin Skadron Dept. of Computer Science – UVA

2 2 What is the problem? Excessive power usage by the physical memory channel – 2mW/Gbits/s by Palmer et al. ISSCC07 – 160W for 10TB/s (Vantrease et al., ISCA08) – Poor scaling in physical channel: RC load in package

3 3 Outline Hypothesis: Wired-RF (ie, coplanar waveguides--CPWs) solves all these problems in technology that is easier to adopt than optical. Architecture for CPW memory interface Evaluation: area, power, and performance Conclusion PS: note that this is over wires (CPWs), not wireless!

4 4 Hypothesis: why wired-RF (RF) as a bandwidth solution? RF Low latency media and modulation (Chang et al., Near Speed-of-Light Signaling Over On-Chip Electrical, 2003) All electrical (impedances matching), development costs closer to CMOS distances from 1mm to 30cm (delays, energy, data rate; RF for Future Chips, Tam et al. 2011) Beckmann et al., Transmission Line Caches, MICRO03 Frank Chang et al. (caches, modulation, high bandwidth, latency ad power reduction; MICRO08, HPCA08) Quilt-packaging (RF coplanar waveguide connecting two dies, > 200GHz, low insertion loss, built), Liu, Buckhanan et al., Notre Dame Intel-Tera (Polka, ITJ07): on-package Modulation and high speed from optical

5 5 Why can't we use RF in a traditional fashion? Different impedances: I/O pad, inner and outer wire bonds, PCB pads, PCB [Liu, 2006]

6 6 Contributions Evaluate power and area gains by replacing power-hungry MC circuitry with on-die RF transceivers + CPW + Quilt packaging Evaluate architectural performance gains due to power and area gains

7 7 Diagram of the proposed organization Example with 1 core and 1 RFMC RF path from a specific core to its rank > 1mm

8 8 Detailed Organization RFMC: MCs coupled to on-die RF transceivers and on- and inter-die coplanar waveguides (CPW)

9 9 Quilt The use of Quilt (inter-dies distance ~40um) allows: – Extending on-die CPWs – Built for RF/low insertion loss: 0.1 dB – Use of processor-die and DRAM dies, RF transceivers, and UCLA RF models – Versus traditional power hungry transceivers (Palmer et al., ISSCC 2007) – Co-planar, not flip-chip – See Lius PhD dissertation and Buckhanan et al., UGIM10

10 10 Interfacing on-dies CPW and Quilt

11 11 Quilt Packaging is a CPW Extension of the interconnection of two dies facing each other Designed for frequencies larger than 200GHz Prototype from Notre Dame tested up to 60 GHz Insertion loss (*): 0.1 dB So far, no transceivers needed for Quilt; due to its low insertion loss

12 12 Transceivers: Power and Area Extracted from Chang, Tam with 10% power reduction on the amplifier to account for savings for Quilt-type packaging Technology (nm) Data rate Per band (Gbits/s) #carriers to match DRAM Power (TX+RX) (mW) Energy per bit (pJ) Area (TX + RX) mm

13 13 Area Comparison MC Area decreases for all components, but RF essentially eliminates PHY 2.4X area savings MCRFMC

14 14 Energy Comparison-PHY Even with technology improvements, RF is more efficient for distances >= 1mm and < 10mm Net power savings (incl. FE & TE) of 4.6X at 5mm

15 Performance Evaluation M5 and DRAMsim 32K L1s, 1MB/core L2 8 cores 1 DRAM rank per MC, DDR2, at 2 GHz Same FE, TE for both MC, RFMC No RF latency benefits in the performance evaluation

16 16 Performance: Stream Baselinecurrent CPUs: 3 or 4 MC RFMC is up to 2.4x faster than MC

17 17 Conclusions RF architecture for on-package CPU-DRAM interconnection Evolutionary changes to CPU and DRAM designstraightforward manufacturability Area and power benefits (preliminary; improve with Quilt dedicated circuits) Benefits on performance for more cores (limited to the number of ranks if the same proportion core-to-rank is desired)

18 18 Thanks!

19 19 Power Comparison FE and TE present power reduction PHY/RF part is evaluated in the next slide (McPAT does not model RF)


Download ppt "Reducing Power and Area by Interconnecting Memory Controllers to Memory Ranks with RF Coplanar Waveguides on the Same Package WEED 2011, ISCA Mario D."

Similar presentations


Ads by Google