Get documents about "paraid
Document Sample
PARAID: A Gear-Shifting
Power-Aware RAID
Charles Weddle, Mathew Oldham, Jin Qian, An-I Andy Wang – Florida St. University
Peter Reiher – University of California, Los Angeles
Geoff Kuenning – Harvey Mudd College
1
Motivation
Energy costs are rising
An increasing concern for servers
No longer limited to laptops
Energy consumption of disk drives
24% of the power usage in web servers
27% of electricity cost for data centers
More energy more heat more cooling lower
computational density more space more costs
Is it possible to reduce energy consumption
without degrading performance while
maintaining reliability?
2
PARAID: A Gear-Shifting Power-Aware RAID
Challenges
Energy
Not enough opportunities to spin down RAIDs
Performance
Essential for peak loads
Reliability
Server-class drives are not designed for
frequent power switching
3
PARAID: A Gear-Shifting Power-Aware RAID
Existing Work
Most trade performance for energy
savings directly
e.g. vary speed of disks
Most are simulated results
4
PARAID: A Gear-Shifting Power-Aware RAID
Observations
RAID is configured for peak performance
RAID keeps all drives spinning for light loads
Unused storage capacity
Over-provision of storage capacity
Unused storage can be traded for energy savings
Fluctuating load
Cyclic fluctuation of loads
Infrequent on-off power transitions can be effective
5
PARAID: A Gear-Shifting Power-Aware RAID
Performance vs. Energy
Optimizations
Performance benefits
Realized under heavy loads
Energy benefits
Realized instantaneously
6
Power-Aware RAID
Skewed striping for energy savings
Preserving peak performance
Maintaining reliability
Evaluation
Conclusion
Questions
7
PARAID: A Gear-Shifting Power-Aware RAID
Skewed Striping for Energy Saving
Use over-provisioned spare storage
Organized into hierarchical overlapping subsets
1 2 3 4 5
RAID
8
PARAID: A Gear-Shifting Power-Aware RAID
Skewed Striping for Energy Saving
Each set analogous to gears in automobiles
1 2 3 4 5
RAID
Gears
1
2
3
9
PARAID: A Gear-Shifting Power-Aware RAID
Skewed Striping for Energy Saving
Soft states can be reclaimed for space
Persist across reboots
1 2 3 4 5
Soft
States
RAID
Gears
1
2
3
10
PARAID: A Gear-Shifting Power-Aware RAID
Skewed Striping for Energy Saving
Operate in gear 1
Disks 4 and 5 are powered off
1 2 3 4 5
Soft
States
RAID
Gears
1
2
3
11
PARAID: A Gear-Shifting Power-Aware RAID
Skewed Striping for Energy Saving
Approximate the workload
Gear shift into most appropriate gear
Minimize the opportunity lost to save power
Conventional RAID PARAID
Energy
( Powered
On Disks ) workload
Workload
( Disk Parallelism )
12
PARAID: A Gear-Shifting Power-Aware RAID
Skewed Striping for Energy Saving
Adapt to cyclic fluctuating workload
Gear shift when gear utilization threshold is met
load
utilization threshold
gear shift time
13
PARAID: A Gear-Shifting Power-Aware RAID
Preserving Peak Performance
Operate in the highest gear
When the system demands peak performance
Uses the same disk layout
Maximize parallelism within each gear
Load is balanced
Uniform striping pattern
Delay block replication until gear shifts
Capture block writes
14
PARAID: A Gear-Shifting Power-Aware RAID
Maintaining Reliability
Reuse existing RAID levels (RAID-5)
Also used in various gears
Drives have a limited number of power
cycles
Ration number of power cycles
15
PARAID: A Gear-Shifting Power-Aware RAID
Maintaining Reliability
Busy disk stay powered on, idle disks stay powered off
Outside disks are role exchanged with middle disks
role exchange
Disk 1 Disk 2 Disk 3 Disk 4 Disk 5 Disk 6
busy power idle
disks cycled disks
Gear 1 disks
Gear 2
Gear 3
16
PARAID: A Gear-Shifting Power-Aware RAID
Logical Component Design
User space Admin tool
Linux kernel File system
RAID
Reliability manager PARAID block mapping
Load monitor Soft RAID
Gear manager
Disk device driver
17
PARAID: A Gear-Shifting Power-Aware RAID
Data Layout
Resembles the data flow of RAID 1+0
Parity for 5 disks does not work for 4 disks
For example, replicated block 12 on disk 3
Disk 1 Disk 2 Disk 3 Disk 4 Disk 5
Gear 1 (1-4) 8 12 ((1-4),8,12)
RAID-5 16 20 (16,20,_) _
1 2 3 4 (1-4)
5 6 7 (5-8) 8
Gear 2 9 10 (9-12) 11 12
RAID-5
13 (13-16) 14 15 16
(17-20) 17 18 19 20
18
PARAID: A Gear-Shifting Power-Aware RAID
Data Layout
Cascading parity updates
For example, updating block 8 on disk 5
Disk 1 Disk 2 Disk 3 Disk 4 Disk 5
Gear 1 (1-4) 8 12 ((1-4),8,12)
RAID-5 16 20 (16,20,_) _
1 2 3 4 (1-4)
5 6 7 (5-8) 8
Gear 2 9 10 (9-12) 11 12
RAID-5
13 (13-16) 14 15 16
(17-20) 17 18 19 20
19
PARAID: A Gear-Shifting Power-Aware RAID
Update Propagation
Up-shift propagation (e.g. shifting from 3 to
5 disks)
Full synchronization
On-demand synchronization
Need to respect block dependency
Downshift propagation
Full synchronization
20
PARAID: A Gear-Shifting Power-Aware RAID
Asymmetric Gear-Shifting Policies
Up-shift (aggressive)
Moving utilization average + moving standard
deviation > utilization threshold
Downshift (conservative)
Modified utilization moving average + moving
standard deviation < utilization threshold
Moving average modified to account for fewer drives
and extra parity updates
21
PARAID: A Gear-Shifting Power-Aware RAID
Implementation
Prototyped in Linux 2.6.5
Open source, software RAID
Implemented block I/O handler, monitor,
disk manager
Implemented user admin tool to configure
device
Updated Raid Tools to recognize PARAID
level
22
PARAID: A Gear-Shifting Power-Aware RAID
Evaluation
Challenges
Prototyping PARAID
Commercial machines
Conceptual barriers
Benchmarks designed to measure peak
performance
Trace replay
Time consuming
23
PARAID: A Gear-Shifting Power-Aware RAID
Evaluation
Measurement framework
client RAID
USB cable
P4 2.8 Ghz, 1 GB RAM
160 GB 7200 RPM SATA RAID
crossover RAID
cable multimeter
server RAID
power
Xeon 2.8 Ghz, 512 MB RAM measurement
RAID
36.7 GB 15k RPM SCSI probes
BOOT 12v & 5v
SCSI power lines
power
cable supply
24
PARAID: A Gear-Shifting Power-Aware RAID
Evaluation
Three different workloads using two different
RAID settings
Web trace - RAID level 0 (2-disk gear 1, 5-disk gear 2)
Mostly read activity
Cello99 - RAID level 5 (3-disk gear 1, 5-disk gear 2)
I/O-intensive workload with writes
PostMark - RAID level 5
Measure peak performance and gear shifting overhead
Speed up trace playback
To match hardware
Explore range of speed up factors and power savings
25
PARAID: A Gear-Shifting Power-Aware RAID
Web Trace
UCLA CS Dept Web Servers (8/11/2006 – 8/14/2006)
File system: ~32 GB (~500k files)
Trace replay: ~95k requests with ~4 GB data (~260 MB unique)
0.6
0.5
0.4
GB/hour 0.3
0.2
0.1
0
1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96
hours
26
PARAID: A Gear-Shifting Power-Aware RAID
Web Trace Power Savings
64x – 60 requests/sec
60
50 Energy Savings
40
RAID-0 64x - 34%
watts 30
PARAID-0
20
128x - 28%
10
0 256x - 10%
0 5 10 15 20 25 30
hours
128x – 120 requests/sec 256x – 240 requests/sec
60 60
50 50
40 40
RAID-0 RAID-0
watts 30 watts 30
PARAID-0 PARAID-0
20 20
10 10
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
hours hours
27
PARAID: A Gear-Shifting Power-Aware RAID
Web Trace Latency
256x
1
0.8 Overhead
0.6 RAID-0 256x - within 2.7%
0.4 PARAID-0
64x - 240%
0.2
80ms vs. 33ms
0
1 10 100 1000 10000 100000
msec
128x 64x
1 1
0.8 0.8
0.6 RAID-0 0.6 RAID-0
0.4 PARAID-0 0.4 PARAID-0
0.2 0.2
0 0
1 10 100 1000 10000 100000 1 10 100 1000 10000 100000
msec msec
28
PARAID: A Gear-Shifting Power-Aware RAID
Web Trace Bandwidth
256x
180
160
140 Overhead
120
100 RAID-0 256x - within
MB/sec
80 PARAID-0 1.3% in high
60 gear
40
20
0
0 5 10 15 20 25 30
hours
128x 64x
180 180
130 130
RAID-0 RAID-0
MB/sec 80 MB/sec 80
PARAID-0 PARAID-0
30 30
-20 0 5 10 15 20 25 30 -20 0 5 10 15 20 25 30
hours hours
29
PARAID: A Gear-Shifting Power-Aware RAID
Cello99 Trace
Cello99 Workload
HP Storage Research Labs
50 hours beginning on 9/12/1999
1.5 million requests (12 GB) to 440MB of unique blocks
I/O-intensive with 42% writes
30
PARAID: A Gear-Shifting Power-Aware RAID
Cello99 Power Savings
32x – 270 requests/sec
50
40 Energy Savings
30 RAID-5
watts 32x - 13%
20 PARAID-5
64x - 8.2%
10
0 128x - 3.5%
0 10 20 30 40 50
hours
64x – 550 requests/sec 128x – 1000 requests/sec
50 50
45
40 40
35
30 30 RAID-5
RAID-5 watts
watts 25
PARAID-5 20 PARAID-5
20
15
10
10
5 0
0
0 10 20 30 40 50
0 10 20 30 40 50
hours hours
31
PARAID: A Gear-Shifting Power-Aware RAID
Cello99 Completion Time
128x
1
0.98 Overhead
0.96 RAID-5 32x - 1.8ms,
0.94 PARAID-5
26% slower
0.92 due to time
0.9 spent in low
1 10 100 1000 10000 100000 gear
msec
64x 32x
1 1
0.98 0.98
0.96 RAID-5 0.96 RAID-5
0.94 PARAID-5 0.94 PARAID-5
0.92 0.92
0.9 0.9
1 10 100 1000 10000 100000 1 10 100 1000 10000 100000
msec msec
32
PARAID: A Gear-Shifting Power-Aware RAID
Cello99 Bandwidth
128x
1000
Overhead
100
RAID-5
MB/sec
PARAID-5
< 1% degra-
10 dation during
peak hours
1
0 500000 1000000 1500000
request number
64x 32x
1000 1000
100 100
RAID-5 RAID-5
MB/sec MB/sec
PARAID-5 PARAID
10 10
1 1
0 500000 1000000 1500000 0 500000 1000000 1500000
requests request number
33
PARAID: A Gear-Shifting Power-Aware RAID
PostMark Benchmark
Popular synthetic benchmark
Generates ISP-style workloads
Stresses peak read/write performance of storage
device
34
PARAID: A Gear-Shifting Power-Aware RAID
Postmark Performance
200
150
seconds 100
50
0
1K files, 50K trans 20K files, 50K trans 20K files, 100K
trans
RAID-5 PARAID-5 high gear PARAID-5 low-gear
35
PARAID: A Gear-Shifting Power-Aware RAID
Postmark Power Savings
80
70
60
50
watts
R A ID 5
40
PA R A ID
30
20
10
0
1 11 21 31 41 51 61 71 81 91 10 111 12 13 14 151 16 171
seconds
36
PARAID: A Gear-Shifting Power-Aware RAID
Related Work
Pergamum
EERAID
RIMAC
Hibernator
MAID
PDC
BlueFS
37
PARAID: A Gear-Shifting Power-Aware RAID
Future Work
Try more workloads
Optimize PARAID gear configuration
Explore asynchronous update propagation
Speed up recovery
Live testing
38
PARAID: A Gear-Shifting Power-Aware RAID
Lessons Learned
Third version of design, early design too
complicated
Data alignment problems
Difficult to measure system under normal load
Hard to predict workload transformations due to
complex system optimizations
Challenging to match trace environments
39
PARAID: A Gear-Shifting Power-Aware RAID
Conclusion
PARAID reuses standard RAID-levels without
special hardware while decreasing their energy
use by 34%.
Optimized version can save even more energy
Empirical evaluation important
40
PARAID: A Gear-Shifting Power-Aware RAID
Research Theme
Data flow management
Storage
MANETs
Current state
Reminiscent of plumbing industry 200 years
ago
Limited interchangeable parts
Poorly understood interactions
41
Research Areas
Power-Aware RAID
Electric-field-based routing for MANETs
Conquest disk-persistent-RAM hybrid file
system
Optimistic replication
Real-time systems
42
Questions
PARAID: A Gear-Shifting
Power-Aware RAID
Contact
Andy Wang – awang@cs.fsu.edu
http://www.cs.fsu.edu/~awang/conquest-2
43
PARAID Recovery
2.7 times slower than conventional raid
For example, 2 gear PARAID device
First, the soft state must recover
Second, data must be propagated
Third, conventional raid must recover
Recovery not as bad for read intensive
workloads
44
PARAID Gear-Shifting
Web Trace Gear-Shifting Stats
256x 128x 64x
Number of gear switches 15.2 8.0 2.0
% time spent in low gear 52% 88% 98%
% extra I/Os for update 0.63% 0.37% 0.21%
propagations
Cello99 Gear-Shifting Stats
128x 64x 32x
Number of gear switches 6.0 5.6 5.4
% time spent in low gear 47% 74% 88%
% extra I/Os for update 8.0% 15% 21%
propagations
45
