560 likes | 573 Views
Learn about Conquest-2's innovative system enhancing energy efficiency and performance by utilizing a disk/RAM hybrid approach. Discover the benefits and evolution in storage technology.
E N D
Conquest-2: Improving Energy Efficiency and Performance Through a Disk/RAM Hybrid File System An-I Andy Wang Florida State University (NSF CCR-0098363, CNS-0410896)
Conquest-2 Team Members • FSU • An-I Andy Wang (PI), Charles Weddle, Cory Fox, Jin Qian, Dragan Lojpur, Mark Carpenter, Ryan Fishel • UCLA • Peter Reiher (Co-PI), Erik Kline • Harvey Mudd College • Geoff Kuenning • Former members: • Mathew Oldham, Noriel Lu, RuGang Xu
Motivation • Computers are becoming cheaper • Energy is not • Energy consumption by storage devices • 8% for laptops • 24% for Web servers • 77% for proxy servers • 27% of the operating costs for data centers Motivation–Conquest–Conquest-2– Power-Aware RAID – Conclusion
Disk Energy Consumption • Laptops: 8% 20 min of battery life • Proxy server: • Higher energy cost higher cooling cost lower density of servers more space cost Motivation–Conquest–Conquest-2– Power-Aware RAID – Conclusion
Performance vs. Energy Benefits • Performance • More relevant during peak loads • Energy savings • Realized instantaneously Motivation–Conquest–Conquest-2– Power-Aware RAID – Conclusion
Roadmap • Conquest • Existing energy-saving approaches • Emergence of memory-rich storage era • Conquest-2 Motivation–Conquest–Conquest-2– Power-Aware RAID – Conclusion
A disk/persistent-RAM hybrid file system Deliver all file system services from memory, with the exception of high-capacity storage Two separate and specialized data paths Benefits: Simplicity Performance Conquest Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
106 105 Hardware Evolution CPU (50% /yr) 1 GHz Memory (50% /yr) Accesses Per Second (Log Scale) 1 MHz 1 KHz Disk (15% /yr) 1990 1995 2000 (1 sec : 6 days) (1 sec : 3 months) Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Magnetic RAM? battery-backed DRAM (write once) flash memory disk tape persistent RAM Storage Media Alternatives $/MB (log) 10-3 106 100 103 accesses/sec (log) 10-3 Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion [Caceres et al., 1993; Hillyer et al., 1996; Qualstar 1998; Tanisys 1999; Micron Semiconductor Products 2000; Quantum 2000]
Booming of digital photography 4 to 10 GB of persistent RAM paper/film Persistent RAM 1” HDD 2.5” HDD 3.5” HDD Price Trend of Persistent RAM 102 101 $/MB (log) 100 10-1 10-2 1995 2000 2005 Year Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion [Grochowski 2000]
User Access Patterns • Small files • Take little space (10%) • Represent most accesses (90%) • Large files • Take most space • Mostly sequential accesses • Except database applications Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion [Iram 1993; Douceur et al., 1999; Roselli et al., 2000]
Files Stored in Persistent RAM • Small files (< 1MB) • No seek time or rotational delays • Fast byte-level accesses • Metadata • Fast synchronous update • No dual representations • Executables and shared libraries • In-place execution Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Large-File-Only Disk Storage • Allocate in big chunks • No fragmentation management • No tricks for small files • Storing data in metadata • Wrapping a balanced tree onto disk cylinders Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion [Devlinux.com 2000]
PostMark Benchmark • Conquest is comparable to ramfs • At least 24% faster than the LRU disk cache • ISP workload (emails, web-based transactions) 40 to 250 MB working set with 2 GB physical RAM Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion [Katcher 1997; Sweeney et al., 1996; Card et al., 1999; Namesys 2002]
PostMark Benchmark • When working set > RAM, Conquest is 1.4 to 2 times faster than ext2fs, reiserfs, and SGI XFS 10,000 files, 3.5 GB working set with 2 GB physical RAM Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Conquest-2 • Conquest has made advancements in terms of better performance • Can we extend Conquest to improve performance and reduce energy consumption at the same time? Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Conquest-Based Numbers • A UCLA Webserver • Single disk • File size threshold of 32KB • Spin down whenever the disk idle time > 10s • Conquest: 84% energy savings • LRU: 64% energy savings • Greater benefits for multiple disks Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Existing Approaches • Provide degraded service • Reduced disk rotation speed • Speculative methods • Predicting idle periods for shutting down a disk • Not suitable for servers • High loads • Uniform data striping among disks Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Just Use Laptop Drives? • Cannot simply replace server drives with laptop ones Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Persistent RAM Storage? • RAM performance/energy savings and disk capacity? Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Why not Conventional Caching? • High overhead to access data stored in RAM storage • 90% cache hit rate ≠ 90% disk idle time • 10% of cache misses can keep a drive spinning all the time • e.g., multimedia workloads Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
What if you have multiple disks? Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Access patterns And access patterns are skewed Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Better Off Caching Cold Disks Spin down cold disks Access patterns Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Conquest-2 Approach • Strategic use of memory storage • Improve performance • Energy-aware memory manager • Power down unused banks • Power-aware RAIDs (PARAIDs) • “Gear-shift” individual drives according to performance demands Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
New Roles of Memory • Shaping the frequency, timing, and predictability of disk accesses • Low frequency of disk access • Better performance • Energy savings • Predictability • Hide the latency to spin a disk up Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
File Access Characterizations Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
frequently used (index, data) infrequently used (index, data) Energy-Aware Memory Management indices data Conceptually simple, but difficult in practice Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Linux Memory Manager (1) • Page allocator maintains individual pages Page allocator Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Zone allocator Linux Memory Manager (2) • Zone allocator allocates memory in power-of-two sizes Page allocator Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Slab allocator Linux Memory Manager (3) • Slab allocator groups allocations by sizes to reduce internal memory fragmentation Zone allocator Page allocator Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
Linux Memory Manager (4) • Difficult to collocate information according to energy constraints Slab allocator Zone allocator Page allocator Motivation –Conquest–Conquest-2– Power-Aware RAID – Conclusion
load drives time Conventional RAID
load load drives drives time time 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 Motivation –Conquest–Conquest-2–Power-Aware RAID – Conclusion
Power-Aware RAID • Observations • RAIDs are configured for peak performance • Uniform striping keeps all drives spinning for light loads • Over-provision of storage capacity • Unused storage can be traded for energy savings • Cyclic fluctuation of loads • Infrequent on-off power transitions can be effective Motivation –Conquest–Conquest-2–Power-Aware RAID – Conclusion
Cyclic Fluctuation of Loads load utilization threshold time gear 2 gear 1 Motivation –Conquest–Conquest-2–Power-Aware RAID – Conclusion
Skewed Striping for Energy Saving • Use over-provisioned spare storage • Can use fewer drives for light loads disk 1 disk 2 disk 3 disk 4 soft-state block replication RAID-5 layout gear 1 gear 2 gear 3 Motivation –Conquest–Conquest-2–Power-Aware RAID – Conclusion
Preserving Peak Performance • Based on RAID-5 • All drives on for peak loads • Full parallelism • Fewer drives on for light loads • Lower latency for small files • Degraded throughput for large files Motivation –Conquest–Conquest-2–Power-Aware RAID – Conclusion
role exchange Disk 1 Disk 2 Disk 3 Disk 4 Disk 5 Disk 6 busy disks power cycled disks idle disks Gear 1 Gear 2 Gear 3 Reliability • Drives have a limited number of power cycles • Form bi-modal distribution of busy/idle drives Motivation –Conquest–Conquest-2–Power-Aware RAID – Conclusion
100% gear 1 gear 2 % of power cycles gear 3 0% power cycles 0 20,000 Reliability • Drives have a limited number of power cycles • Form bi-modal distribution of busy/idle drives • Rotate drives with more power cycles Motivation –Conquest–Conquest-2–Power-Aware RAID – Conclusion
Reliability • Drives have a limited number of power cycles • Form bi-modal distribution of busy/idle drives • Rotate drives with more power cycles • Ration number of power cycles • Distributed parity (RAID-5) • Tolerate single-disk failures Motivation –Conquest–Conquest-2–Power-Aware RAID – Conclusion
Other Issues • Update propagations • Gear-shifting policies disk utilization disk utilization gear 2 utilization threshold gear shift utilization threshold downshift gear 1 utilization threshold gear shift time 300s, 60s, 10s moving averages time Motivation –Conquest–Conquest-2–Power-Aware RAID – Conclusion
load load time time Gear-Shifting Policies • Ideal • In practice
Empirical Measurements • Servers are not measurement friendly • Time consuming • Cannot easily apply the trick of skipping idle times Motivation –Conquest–Conquest-2–Power-Aware RAID – Conclusion
everything on or off geared switching linear scaling no choices time Workload Selection • Need to match with the hardware setup energy consumption load
Experiment Set 1 • Workload • FSU CS Department Web Server trace • A single day trace • Hardware • Dell 2600 with 5 drives • PARAID • 2 gears (3-disk RAID-0 and 5-disk RAID-0) • No energy-aware memory management Motivation –Conquest–Conquest-2–Power-Aware RAID – Conclusion
Web Trace Replay 512x 1024x 1920x
Experiment Set 2 • Workload • Cello99 server I/O trace from HP • A 50-hr trace • Hardware • Dell 2600 with 5 drives • PARAID • 2 gears (3-disk RAID-5 and 5-disk RAID-5) • No energy-aware memory management Motivation –Conquest–Conquest-2–Power-Aware RAID – Conclusion
Cello99 • 50hr Trace 32x 64x 128x