Using GPUs to Enable Highly Reliable Embedded Storage Matthew Curry - - PowerPoint PPT Presentation

Using GPUs to Enable Highly Reliable Embedded Storage

University of Alabama at Birmingham 115A Campbell Hall 1300 University Blvd. Birmingham, AL 35294-1170 Computer Science Research Institute Sandia National Laboratory PO Box 5800 Albuquerque, NM 87123-1319

Matthew Curry (curryml@cis.uab.edu) Lee Ward (lee@sandia.gov) Anthony Skjellum (tony@cis.uab.edu) Ron Brightwell (rbbrigh@sandia.gov) High Performance Embedded Computing (HPEC) Workshop 23-25 September 2008 Approved for public release; distribution is unlimited.

The Storage Reliability Problem

• Embedded environments are subject to

harsh conditions where normal failure estimates may not apply

• Since many embedded systems are

purposed for data collection, data integrity is of high priority

• Embedded systems often must contain as

little hardware as possible (e.g. space applications)

Current Methods of Increasing Reliability

• RAID

– RAID 1: Mirroring (Two-disk configuration) – RAID 5: Single Parity – RAID 6: Dual Parity

• Nested RAID

– RAID 1+0: Stripe over multiple RAID 1 sets – RAID 5+0: Stripe over multiple RAID 5 sets – RAID 6+0: Stripe over multiple RAID 6 sets

Current Methods of Increasing Reliability

• RAID MTTDL (Mean Time to Data Loss)

– RAID 1: MTTF2/2 – RAID 5: MTTF2/(D*(D-1)) – RAID 6: MTTF3/(D*(D-1)*(D-2))

• Nested RAID MTTDL

– RAID 1+0: MTTDL(RAID1)/N – RAID 5+0: MTTDL(RAID5)/N – RAID 6+0: MTTDL(RAID6)/N

RAID Reliabliity (1e7 hours MTTF, 24 hours MTTR)

1.00E+00 1.00E+01 1.00E+02 1.00E+03 1.00E+04 1.00E+05 1.00E+06 1.00E+07 1.00E+08 1.00E+09 1.00E+10 1.00E+11 1.00E+12 1.00E+13 1.00E+14 1.00E+15 1.00E+16 1.00E+17 1.00E+18 1.00E+19 4 5 6 8 10 12

Number of Disks MTTDL

RAID N+3 RAID 6+0 RAID 6 RAID 1+0 RAID 5+0 RAID 5 RAID 0

Why N+3 (Or Higher) Isn’t Done

• Hardware RAID solutions largely don’t

support it

– Known Exception: RAID-TP from Accusys uses three parity disks

• Software RAID doesn’t support it

– Reed-Solomon coding is CPU intensive and inefficient with CPU memory organization

An Overview of Reed-Solomon Coding

• General method of generating arbitrary

amounts of parity data for n+m systems

• A vector of n data elements is multiplied

by an n x m dispersal matrix, yielding m parity elements

• Finite field arithmetic

Multiplication Example

• {37} = 32 + 4 + 1 = 100101 = x5 + x2 + x0
• Use Linear Shift Feedback Register to

multiply an element by {02}

x0 x1 x2 x3 x4 x5 x6 x7

Multiplication Example

• Direct arbitrary multiplication requires

distributing so that only addition (XOR) and multiplication by two occur.

– {57} x {37} – {57} x ({02}5 + {02}2 + {02}) – {57} x {02}5 + {57} x {02}2 + {57} x {02}

• Potentially dozens of elementary
• perations!

Optimization: Lookup Tables

• Similar to the relationship that holds for real

numbers: elog(x)+log(y) = x * y

• This relationship translates (almost) directly to

finite field arithmetic, with lookup tables for the logarithm and exponentiation operators

• Unfortunately, parallel table lookup capabilities

aren’t common in commodity processors

– Waiting patiently for SSE5

NVIDIA GPU Architecture

• GDDR3 Global Memory
• 16-30 Multiprocessing Units
• One shared 8 KB memory region per

multiprocessing unit (16 banks)

• Eight cores per multiprocessor

Integrating the GPU

3+3 Performance

200 400 600 800 1000 1200 3 1 2 2 1 3 3 9 1 2 3 4 8 6 6 8 4 1 2 1 2 1 3 8 1 5 6 1 7 4 1 9 2 2 1 2 2 8 2 4 6 2 6 4 2 8 2 3 3 1 8 3 3 6 3 5 4 3 7 2 3 9 Data Size (KB) Throughput (MB/s) 3+3

29+3 Performance

1300 1320 1340 1360 1380 1400 1420 1440 1460 1480 1500 58 116 174 232 290 348 Data Size (KB) Throughput (MB/s) 29+3

Neglecting PCI Traffic: 3+3

500 1000 1500 2000 2500 3 12 21 30 39 12 30 48 66 84 102 120 138 156 174 192 210 228 246 264 282 300 318 336 354 372 390 Data Size (KB) Throughput (MB/s) 3+3 (No PCI Traffic)

Conclusion

• GPUs are an inexpensive way to increase

the speed and reliability of software RAID

• By pipelining requests through the GPU,

N+3 (and greater) are within reach

– Requires minimal hardware investment – Provides greater reliability than available with current hardware solutions – Sustains high throughput compared to modern hard disks