HDD vs SATA SSDs in th September 2017
HDD vs. SATA SSDs in th September 2017 NVM SOLUTIONS GROUP 1
You Have a Choice to Make in Data Storage HDD vs SATA-based SSDs Basics SATA-based SSDs 10 K HDD Poor annualized failure rates (AFR) Intel® SSD DC S 4500 and DC S 4600 Series or Higher reliability with up to 3 x lower AFR 2 Higher power requirement up to 3. 2 x lower power requirement 1 Higher cooling requirement up to 3. 2 x lower cooling requirement 1 Larger physical size Store up to 6. 4 x more TB in same space 4 Increased data duplication needs up to 5: 1 data duplication space savings 3 1. Intel TCO tool comparing Intel SSD DC S 4500 960 GB and Seagate Savvio* 10 K. 6 900 GB 10 k SAS HDD & Intel SSD DC S 4500 960 GB and Seagate Savvio 10 K. 6 900 GB 10 k SAS HDD. The workload equates 128 KB (131, 072 bytes) Queue Depth equal to 32 sequential writes. Average power for Seagate drive from http: //www. tomshardware. com/charts/enterprise-hdd-charts/-19 -Power-Requirement-at-Database, 3389. html. http: //estimator. intel. com/ssddc/ 2. Industry AFR Average (2. 11%): Source - Backblaze. com https: //www. backblaze. com/blog/hard-drive-failure-rates-q 1 -2017/ 3. Source – 451 Research, Voice of the Enterprise Storage, Q 1 2016, page 106. 4. Intel® SSD DC S 4500 at 2. 5” 7. 6 TB vs. Seagate Enterprise* Performance 10 K SAS HDD 2. 5” 2. 4 TB *Other names and brands may be claimed as the property of others. NVM Solutions Group 2
Factors Affecting Data Center Storage Decisions Reliability Performance Operating Costs Failure Rates Transaction Throughput Data Protection Read Performance Cooling Health Tracking Write IOPs Rack Space Price Power HDD SSD TCO - Total Cost of Ownership NVM SOLUTIONS GROUP 3
Intel® SSD Reliability: More Uptime, Lower Replacement Costs Annual Failure Rate (AFR) 5% 4% Power Loss Imminent (PLI) 4. 8% HDD AFR Industry Average 1 Power Failure Detection Circuit 3% Power Loss-Tolerant Charge Storage 2% Enhanced Firmware Data Safe Mode 1% Hot-Plug Inrush Current Mitigation 0% Smart Logs . 44% ü Temperature ü Endurance ü NAND ü Power Cycles ü Power-on Hours ü Warnings Intel® SSD GOAL Actual AFR Consistently Better Than Goal 2 Data Protection in Event of Power Loss Event Monitor and Manage the SSD In-band 1) Industry AFR Average: Source - Backblaze. com*, https: //www. backblaze. com/blog/hard-drive-reliability-q 3 -2015/ 2) 10 x Lower AFR: Source - Intel SSD Annualized Fail Rate Report for all of 2015. NVM SOLUTIONS GROUP 4
Driving Storage Transformation with SSDs Improving Competitiveness 1 Reduce Opex Reduce power cost and improve operational efficiency 2 Increase agility Efficiently grow and adapt to changing demand 3 Reduce Capex Reduce infrastructure procurement cost NVM SOLUTIONS GROUP 5
Do more per server with intel® SSDs compared to HDDs NVM Solutions Group 6
Intel® SSD DC S 4500 and DC S 4600 Series Optimized for space efficient capacity 36 Storage 8 Capacity 2. 5” 7 mm 7. 68 TB Intel® SSD DC S 4500 x 24 TB 6. 4 x more t. B per Serve Standard 2 U Server up to To 1 From up to 2. 5” 1. 58 TB 2. 4 TB SATA HDD x 24 Performance measured using FIO* on Linux* with 128 KB (131, 072 bytes) of transfer size with Queue Depth 32 NVM Solutions Group 7
Get More Done with Intel® SSDs Sequential Read MB/s 600 2. 5 x 600 Faster Sequential Write MB/s Faster 500 400 300 200 100 0 10 K SAS HDD (900 GB) 15 K SAS HDD (900 GB) Intel S 3520 (960 GB)Intel S 4500 (960 GB) Intel S 3520 (900 GB) Intel S 4500 (960 GB) 4 K Random Write IOPS 4 K Random Read IOPS 223 x Faster 80000 3 x 35000 25000 60000 20000 50000 Faster 30000 70000 80 x 15000 40000 30000 10000 20000 5000 10000 0 0 10 K SAS HDD (900 GB) 15 K SAS HDD (900 GB) Intel S 3520 (960 GB) Intel S 4500 (960 GB) Performance measured using FIO* on Linux* with 128 KB (131, 072 bytes) of transfer size with Queue Depth 32. Performance with different Queue Depths will be provided upon request. HDD Specs from http: //www. tomshardware. com/charts/enterprise-hdd-charts/- 02 -Read-Throughput-Maximum-h 2 benchw-3. 16, 3372. html. Performance measured using IOMETER with Queue Depth 32. Measurements are performed on a full Logical Block Address (LBA) span of the drive. 4 KB=4, 096 bytes compared with HDD Specs from http: //www. tomshardware. com/charts/enterprise-hdd-charts/- 02 -Read-Throughput-Maximum-h 2 benchw-3. 16, 3372. html *Other names and brands may be claimed as the property of others. NVM Solutions Group 8
Intel® SSD DC S 4500 Series Built to increase storage workload adaptability 80000 More performance/capacity helps… § Increase users with each server § Add more services to a server § Perform more workloads with a server § Repartition to adapt to conditions § Get more value out of every server IOPS/TB 70000 60000 209 X 50000 40000 30000 Faster Storage Workload 1 20000 Do more with a DC S 4500 based server 10000 1 Intel® SSD DC S 4500 960 GB vs. Seagate Savvio* 10 K. 6 900 GB 10 k SAS HDD and Intel® 0 10 K SAS HDD (900 GB) NVM Solutions Group 15 K SAS HDD (900 GB) Intel S 3520 (960 GB) Intel S 4500 (960 GB) SSD DC S 4600 960 GB vs. Seagate Savvio 10 K. 6 900 GB 10 k SAS HD. Comparing Intel spec 4 k Random Read IOPS at QD 32 vs Toms Hardware data of Seagate drive from http: //www. tomshardware. com/charts/enterprise-hdd-charts/- 27 -IOMeter-2006. 07. 27 -4 KRandom-Reads, 3397. html Any differences in your system hardware, software or configuration may affect your actual performance. *Other names and brands may be claimed as the property of others. 9
Intel® SSD DC S 4500 Series Optimizing server agility Read IOPS/TB 80000 300 Read MBPS/W 70000 60000 50000 40000 30000 250 209 X 200 more efficient Performance 150 100 1 20000 8 X more Bandwidth Efficient 2 50 10000 0 0 10 K SAS HDD (900 GB) 15 K SAS HDD (900 GB) Intel S 3520 (960 GB) Intel S 4500 (960 GB) 1 Intel® SSD DC S 4500 960 GB vs. Seagate Savvio* 10 K. 6 900 GB 10 k SAS HDD and Intel® SSD DC S 4600 960 GB vs. Seagate Savvio 10 K. 6 900 GB 10 k SAS HD. Comparing Intel spec 4 k Random Read IOPS at QD 32 vs Toms Hardware data of Seagate drive from http: //www. tomshardware. com/charts/enterprise-hdd-charts/- 27 -IOMeter-2006. 07. 27 -4 K-Random-Reads, 3397. html 2 Source- Intel. Performance measured using FIO* on Linux* with 128 KB (131, 072 bytes) of transfer size with Queue Depth 32. divided by The workload equates 128 KB (131, 072 bytes) Queue Depth equal to 32 sequential reads. RMS (Root Mean Squared) Average Power is measured using Agilent* Power Analyzer over a 100 ms sample period. HDD specs of average power from http: //www. tomshardware. com/charts/enterprise-hdd-charts/- 02 -Read-Throughput-Maximum-h 2 benchw-3. 16, 3372. html *Other names and brands may be claimed as the property of others. NVM Solutions Group 10
Legal Disclaimers Intel may make changes to specifications and product descriptions at any time, without notice. Designers must not rely on the absence or characteristics of any features or instructions marked "reserved" or "undefined". Intel reserves these for future definition and shall have no responsibility whatsoever for conflicts or incompatibilities arising from future changes to them. The information here is subject to change without notice. Do not finalize a design with this information. The products described in this document may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request. Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration. No computer system can be absolutely secure. Check with your system manufacturer or retailer or learn more at intel. com. Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade. Tests document performance of components on a particular test, in specific systems. Differences in hardware, software, or configuration will affect actual performance. Consult other sources of information to evaluate performance as you consider your purchase. Cost reduction scenarios described are intended as examples of how a given Intel- based product, in the specified circumstances and configurations, may affect future costs and provide cost savings. Circumstances will vary. Intel does not guarantee any costs or cost reduction. Intel does not control or audit the design or implementation of third party benchmark data or Web sites referenced in this document. Intel encourages all of its customers to visit the referenced Web sites or others where similar performance benchmark data are reported and confirm whether the referenced benchmark data are accurate and reflect performance of systems available for purchase. Intel and the Intel logo are trademarks of Intel Corporation in the U. S. and/or other countries. © 2017 Intel Corporation. All rights reserved. *Other names and brands may be claimed as the property of others. NVM SOLUTIONS GROUP 11
- Slides: 11