Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi...
-
Upload
charleen-ward -
Category
Documents
-
view
216 -
download
0
Transcript of Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi...
![Page 1: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/1.jpg)
Designing Tomorrow’s Computing Platforms
Challenges, Solutions, and Tools
Sudhanva Gurumurthie-mail: [email protected]
![Page 2: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/2.jpg)
Talk Outline
• Modern Computer Architecture– The Good– The Bad– The Ugly
• My Previous Work
• Current and Future Research
![Page 3: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/3.jpg)
The Good
![Page 4: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/4.jpg)
Source: http://www.intel.com/technology/silicon/mooreslaw/
![Page 5: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/5.jpg)
Microprocessor Technology Advancement
• Plentiful Transistors– Superscalar, SMT, CMP– Larger caches, deeper memory-hierarchy– High-bandwidth access to memory
• Simultaneously, clock frequencies have grown tremendously
![Page 6: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/6.jpg)
Storage Has Become Ubiquitous
Density
Speed
![Page 7: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/7.jpg)
Source: Hitachi GST Technology Overview Charts, http://www.hitachigst.com/hdd/technolo/overview/storagetechchart.html
Growth in Drive Performance
![Page 8: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/8.jpg)
The Bad
![Page 9: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/9.jpg)
Power Dissipation
0
10
20
30
40
50
60
70
80
90
Po
we
r (W
)
8086 286 386 486 Pentium PentiumIII
Pentium4
![Page 10: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/10.jpg)
Particle Induced Soft-Errors
01
Source: FACT Group, Intel
![Page 11: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/11.jpg)
Are you kidding me?
• No!!– In 2000, Sun Microsystems reported random crashes
in one of its server products due to no parity-protection in the caches.
– Eugene Normand’s study of the error-logs of large systems indicated several such errors
– There are conference sessions and even full conferences/workshops devoted to this problem
– Have personal experience collecting and analyzing soft-error data
![Page 12: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/12.jpg)
Where Do These Particles Come From?
• Neutrons– Terrestrial cosmic rays
• Alpha particles– Packaging
![Page 13: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/13.jpg)
Should we worry?
• Yes!!– Thanks to Moore’s Law
• Lower operating voltages• Exponential increase transistor integration density• Power management (voltage-scaling)
– Larger systems
• Impractical to shield against cosmic rays– Need several feet of concrete– Radiation-hardening hurts performance, area, and
cost
![Page 14: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/14.jpg)
Redundant Multi-Threading
InputReplicator
OutputComparator
Rest of the System
Source: Mukherjee et al, “Detailed Design and Evaluation of Redundant Multithreading Alternatives”, ISCA’02
![Page 15: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/15.jpg)
Performance of Redundant Multi-Threading
0
5
10
15
20
25
30
35
40
45
Pe
rce
nta
ge
of
IPC
Lo
st
gzip swim vpr gcc mesa art mcf equake parser vortex bzip2
![Page 16: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/16.jpg)
Temperature Affects Disk Drive Reliability
• Heat-Related Problems– Data corruption– Higher off-track errors– Head-crashes
Disk drive design constrained by the thermal-envelope• Puts a limit on drive performance
Source: D. Anderson et al, “More than an Interface – SCSI vs. ATA”, FAST 2003.
![Page 17: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/17.jpg)
Power =~ (# Platters)*(RPM)2.8(Diameter)4.6
Increase RPM
Thermal-Constrained Design
Increase RPM
Lower Capacity
Shrink Platter
1 platterData Rate =~ (Linear-Density)*(RPM)*(Diameter)
(RPM)2.8 (Dia)4.6 (# Platters)
Lower Data Rate
Data-Rate Capacity
Temperature
40% AnnualIDR Growth
![Page 18: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/18.jpg)
The BadDrive Temperature
10
100
1000
Year
Tem
pera
ture
(C
)
2.6" 2.1" 1.6"
Thermal-Envelope
![Page 19: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/19.jpg)
The BadData Rate
![Page 20: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/20.jpg)
30-60% Performance Boostfor 10,000 RPM Increase
![Page 21: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/21.jpg)
Search-Engine Thermal Behavior
Thermal Envelope = 45.22 C
![Page 22: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/22.jpg)
The Ugly
![Page 23: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/23.jpg)
Design Tools
• Designing complex systems requires extensive simulation
• Need to model all aspects of the system– Software layers– Power– Temperature– Effect of faults
![Page 24: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/24.jpg)
Simulation Problems
• Painfully slow– Speed vs. Accuracy
• No good support available for modeling effects like temperature and reliability
• Can themselves be hard to write
• Buggy
![Page 25: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/25.jpg)
My Previous Work
![Page 26: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/26.jpg)
Thesis Work:Power Management of
Enterprise Storage Systems
![Page 27: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/27.jpg)
Enterprise Storage Market Growth
• Storage demand growing at annual rate of 60%– By 2008, a company would manage 10 times the
storage it has today.Sources:
1. “Enterprise Storage: A Look into the Future”, TNM Seminar Series, Oct. 31, 2000
2. “More Power Needed”, Energy User News, Nov. 2002
![Page 28: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/28.jpg)
Power Demands of Data Centers“What matters most to the computer designers at Google is not speed but
power – low-power – because data centers can consume as much electricity as a city”, Eric Schmidt, CEO, Google
• Data centers consume several Megawatts of power
• Electricity bill– $4 billion/year– Disks account
for 27% of computing-load costs
• Difficult to cool at high power-densities
Sources:
1. “Intel’s Huge Bet Turns Iffy”, New York Times article, September 29, 2002
2. “Power, Heat, and Sledgehammer, Apr. 2002.
3. “Heat Density Trends in Data Processing, Computer Systems, and Telecommunications Equipment”, 2000.
![Page 29: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/29.jpg)
Data Center Cooling Costs
• Data center of a large financial institution in New York City– Power consumption ~ 4.8 MW
Source: “Energy Benchmarking and Case Study – NY Data Center No. 2”, Lawrence Berkeley National Lab, July 2003.
51%42%
7%
Servers Air-Conditioning Other
![Page 30: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/30.jpg)
Where Does Power Go?
Spindle Motor(SPM)
Voice-Coil Motor(VCM)
Idle = 9 W
Seek = 13 W
Standby = 1 W
4 W
Active = 11 W
![Page 31: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/31.jpg)
Traditional Power Management (TPM)
Disk Active
Spindown
Standby Mode
Spinup
Disk Request
Time
Disk Active
Idleness
Detected
Idle
![Page 32: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/32.jpg)
I/O Characteristics of Server Systems
• Large number of disks– RAID arrays
• Heavier I/O loads sustained over long periods.• Stringent performance requirements.• Server disks physically different
– Not made to use spindowns.– Longer spindown/spinup latencies
• Server Disk - Hitachi Ultrastar – 15 seconds/26 seconds• Laptop Disk - Hitachi Travelstar – 4.5 seconds
![Page 33: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/33.jpg)
Feasibility of Applying TPM
• No prior study on how to tackle this problem systematically.
• Questions1. Is there idleness?2. Can we do TPM?
• Answers1. Yes2. No! Why??
• Large number of very short duration (few ms) idle-periods
![Page 34: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/34.jpg)
The Solution
• Traditional Power Management– Not effective for server workloads
• Power =~ (# Platters)*(RPM)2.8(Diameter)4.6
– All three can be varied at design-time to meet the power budget
• Laptop vs. Server disk
– RPM could be varied dynamically
• Dynamic RPM (DRPM)
![Page 35: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/35.jpg)
Potential Benefits of DRPM
0
10
20
30
40
50
60
70
80
% S
av
ing
s in
E idle
10 100 500 1000 10000 100000
Mean Inter-Arrival Time (ms)
TPMperf DRPMperf Combined
![Page 36: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/36.jpg)
Control-Policy Performance
![Page 37: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/37.jpg)
Research Impact
• The feasibility study [ISPASS’03] started off new research in server disk power management– Active groups: UIUC, Rutgers, UMass, UArizona,
Rochester
• DRPM paper [ISCA’03] widely cited in architecture and systems conferences like ISCA, HPCA, ASPLOS, SOSP, OSDI
• Multi-speed drives starting to appear in the market– Hitachi Deskstar 7K400
![Page 38: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/38.jpg)
My Other Work
• Microarchitectural Techniques to Enhance Redundant Multi-Threading Performance– Instruction Reuse [ISCA’04]
• Soft-Error Data Collection and Analysis from Actual Systems (Intel)
• Soft-Error Tolerant Cache Coherence-Protocols (Intel)
• Simulator Design– SoftWatt [HPCA’02]
– MEMSIM (IBM Research)
![Page 39: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/39.jpg)
More Details About My Work
• Papers:– S. Gurumurthi et al., Disk Drive Roadmap from the Thermal
Perspective: A Case for Dynamic Thermal Management, ISCA 2005.
– A. Parashar et al., A Complexity-Effective Approach to ALU Bandwidth Enhancement for Instruction-Level Temporal Redundancy, ISCA 2004.
– S. Gurumurthi et al., DRPM: Dynamic Speed Control for Power Management in Server Class Disks, ISCA 2003.
– S. Gurumurthi et al., Using Complete Machine Simulation for Software Power Estimation: The SoftWatt Approach, HPCA 2002.
• Available via my CS Department homepage.
![Page 40: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/40.jpg)
Some Research Directions
• Temperature-Aware Storage Systems– Devices– Systems issues
• Multi-Dimensional Approach to Fault Tolerance– Tradeoffs between performance, power, reliability– Dynamic adaptation
• Microarchitectural Support for Security• Design of accurate and fast simulation tools
![Page 41: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/41.jpg)
Research Directions in Storage
• Storage architecture is still quite a nascent field
• Plenty of research opportunities:– Emerging technologies
• MEMS, holographic, molecular storage
– New Research Avenues• Security• Application/Content-Awareness• Active disks• Long-term and survivable storage
![Page 42: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/42.jpg)
Looking for Students!
• Shall be offering a research course in Spring 2006.– Many project opportunities
• Contact Information:– E-mail: gurumurthi@cs– Office: 236B, Olsson Hall
![Page 43: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/43.jpg)
Divider Slide
![Page 44: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/44.jpg)
Approach 1:Seek Throttling
T
E
M
P
E
R
A
T
U
R
E
TIME
Thermal-Envelope
VCM On
VCM Off
![Page 45: Designing Tomorrow’s Computing Platforms Challenges, Solutions, and Tools Sudhanva Gurumurthi e-mail: gurumurthi@cs.virginia.edu.](https://reader031.fdocuments.us/reader031/viewer/2022032206/56649efd5503460f94c11cd1/html5/thumbnails/45.jpg)
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
Benchmark
Perc
en
tag
e o
f IP
C G
ap
(S
IE-D
IE)
reco
vere
d
DIE-IRB-1K-sat
DIE-2xALU
DIE-IRB-ideal
Results2-42% reduction in IPC
gap (avg. 23%)