Understanding Performance of Ssds in the Enterprise Will Akin Principal Engineer, Intel NAND Solutions Group Dr

SF 2009

Understanding Performance of SSDs in the Enterprise Will Akin Principal Engineer, Intel NAND Solutions Group Dr. John Busch Co-founder, CEO, Schooner Information Technology MEMS002 Agenda Enterprise Performance Tuning

• Why Intel SSDs in enterprise

• Performance nuances of an SSD

• Schooner – Demonstrating the value of Performance

2 Agenda Enterprise Performance Tuning

• Why Intel SSDs in enterprise

• Performance nuances of an SSD

• Schooner – Demonstrating the value of Performance

3 Traditional Storage Optimizations

using traditional high performance hard disk drives DRAM HDD . Add lots of . Lots of DRAM Spindles . Store the . Reduce Seeks, working set in Increase IOPS DRAM to avoid disk latency The Bottom Line

Power & IOPS Cost & TCO Energy Costs

Add DRAM & Spindles to address bottlenecks in the server storage hierarchy

4 What if we fill one rack with SSDs? 120 HDDs 120 SSDs

. 36,000 IOPS . 4,200,000 IOPS . 12 GB/sec . 36 GB/sec sustained BW sustained BW . 1452 Watts . 288 Watts

Per HDD*: Per SSD**: – R/W 100 MB/sec – Read 250 MB/sec – 300 IOPS – Write 170 MB/sec – 12.1 W (active) – 35,000 IOPS (Read) * Typical 143GB 15K RPM HDD – 2.4 W (active) **Intel® X25-E SSD 160 GB

Energy Costs Read IOPS Sustained BW 5X 115X 3X Reduction Increase Increase

SSD changing the economics of data centre higher performance with lower energy cost

5 Agenda Enterprise Performance Tuning

• Why Intel SSD in enterprise

• Nuances of SSD Performance

• Schooner – Demonstrating the value of Performance

6 Nuances of SSD Performance

Four Elements to SSD Performance

1. Type of NAND • Single Level Cell (SLC) • Multi Level Cell (MLC) 2. Indirection system • Erasing and Writing Blocks 3. Host traffic pattern • Workload and Fullness of SSD 4. Spare area • SSD Workspace

7 Nuance of the “Indirection System”

• Logical to physical LBA mapping removes need for atomic operations like read modify write (RMW) – The placement of new LBA information can be packed into pages that are at new physical locations • Data placement in previously erased blocks makes foreground work (Host IO operations) faster • Indirection “clean up” needs to reclaim invalid physical locations in background

SSD L0 to P0 Logical L0 to P348 Host LBA L0 to P5120

SSD converts a Physical Page to Logical LBA. Logical LBA will not reside in the same physical location each time it is written

8 Host Traffic Pattern: Empty vs Full • An empty SSDs achieves its maximum write performance under all workloads Sequential • Once initially filled data performance will decrease • Steady State write performance is achieved when the SSD has settled into a consistent write latencies pattern • A Steady State can be observed when SpareSpare AreaArea – User capacity is full – Consistent work load is provided SSDs steady state performance will have dependencies on the amount of spare area

9 Host Traffic Pattern: Sequential vs Random

• Steady state performance of an SSD full of sequential data is better than the steady state of an SSD full of data random data

Sequential – Sequential sectors will be invalidated in larger linear clusters than random. – Invalidation of sectors within a block is spotty in random writes.

• Changing the workload of an SSD from sequential to random will cause SpareSpare Area Area the performance to fall whereas changing from random to sequential data

Random will increase performance over time.

Valid Data

Invalid Data 10 11 12 Drive Performance vs Spare Area

160GB MLC Performance scaling w/ over provisioning 10000 600% IOP 8000 500% Improvement 400% • As spare area increases so 6000 300% IOPS 4000 does performance 200% IOPS 2000 100% %IOPS 0 0% • MLC has a greater % 7% 17% 27% 47% 57% performance increase due % Spare Area to the relative smaller 64GB SLC Performance scaling w/ over spare area to start with provisioning 10000 70% IOP 60% 8000 Improvement 50% 6000 40%

4000 30% IOPS(avg) IOPS 20% 2000 % IOPS 10% 0 0% 27% 39% 51% Intel Confidential % Spare Area

IOPS scales with increase in spare area 13 Summary of Performance Impacts

• NAND type SLC VS MLC NAND Speed SLC vs MLC • Indirection system optimization + Workload • Host traffic pattern – empty vs full, sequential vs random random vs sequential Read/Writes + • Spare Area contributes to the write Spare Area amplification factor Spare Capacity work space + • Efficient Firmware optimizes the use Firmware Efficiencies of the NAND, Spare Area, and Efficient NAND writes and Workload wearleveling

Workload and Spare Area are outside influences on performance

14 Agenda Enterprise Performance Tuning

• Performance Nuances of an SSD

• Impacts on SSD Performance

• Schooner – Demonstrating the Value of Performance

15 Agenda Schooner Overview Caching Tier Database Tier

16 Agenda Schooner Overview

17 Too Much Rack, Power, and Pipe

For every 100 U.S. data- The number of units of energy Datacenter centers use installed servers From 2003 to 2008 piped into a data equipment is more energy in the U.S. will the data size of the center, only only utilized than the increase from 2.2 average web page has three are used 6% to 10%. entire nation million in 2007 to more than tripled. for actual - William Forrest 6.8 million in 2010. computing. of Sweden. –websiteoptimization.com Forbes - U.S. Department - EE Times - Frost & Sullivan of Energy

18 Typical Web 2.0 and Cloud Deployment

Web Caching Database Storage Tier Tier Tier Tier

Slaves Web App Servers Cache Servers Master Pool 1 Pool

End User WWW Pool 2 Pool

! Memory Bound ! Disk Bound

19 Schooner Balanced Platform Hardware

Parallel Flash • Hardware Features Controllers SSDs – Dual quad-core Intel® Xeon® 64GB DRAM processors 5500 series Dual Xeon PCle – ½ terabyte of Intel® X25-E 5500 Processors HUB Extreme SATA Solid-State Drive flash memory 8x Multi – 64 GB DRAM GbE External – 1/10 Gb Ethernet Networking – Built from the ground up for 10GbE reliability and serviceability; (Memcached appliance only) all critical components are field-replaceable – IBM X-Series M2 3650 server platform

20 The Schooner Operating Environment Optimizing Parallel Flash Memory and Multi-Core Processors

Networked Service Application: 1/10 G E-Net Management  Application Protocol Handling

Data Fabric: Object Attributes  Thread and Core Management Synchronization/Concurrency Management  DRAM Cache Management Admin  Container Management Object Metadata Management Configure Replication Management Monitor Control Flash Management Subsystem: Optimize Space Allocation and Shard Management Object Replacement (cache mode)  Persistency Management Tiered Storage Management

Flash Access: Asynchronous I/O Handling  Data Striping  Interrupt Batching

21 Schooner Appliance Overview

Schooner Data Access Appliances . Purpose-built for Web 2.0 and cloud computing datacenters . 8x performance improvements . 1/8th the power and space requirements . 60% lower TCO . 100% compatible with existing applications and management tools . Manufactured, sold and supported worldwide by IBM

Initial Products in the Schooner Appliance Family . Schooner Appliance for Memcached . Schooner Appliance for MySQL Enterprise™

22 23 24 Caching Tier: Schooner Appliance for Memcached

25 Schooner Appliance for Memcached

• Features Feature Ordinary Schooner Servers Appliance for – Instantaneous and transparent data Memcached persistence, replication, and recoverability Containers  – Multiple containers, allowing separate Persistence and  Memcached domains on a single appliance recoverability – Cache mode or persistent key/value Replication and  store mode for each container auto failover – Plug-and-play auto configuring Plug-and-play  – 100% compatible with existing auto-configuration Memcached applications and GUI and CLI  monitoring tools • Benefits – 8x performance improvement over traditional Memcached servers – Replaces ordinary servers at a ratio 360,000 TPS of up to 8:1, yielding both capex and opex savings

45,000 TPS

26 Memcached Appliance TCO Analysis

With Schooner, consolidation and higher Typical savings for a mid-size 1TB operational efficiency cuts TCO for a rapidly Web 2.0 datacenter growing 1TB Memcached workload. 3yr TCO: Without Schooner $619,000 Caching Tier Caching

OpEx 3yr TCO: $302,000

OpEx

CapEx: 33 2U servers, 18.7 kW (and growing) CapEx: 4 33 Servers Appliances

Legacy Schooner With Schooner

The Bottom Line:

. Immediate capex savings . 51% TCO savings ($317,000) over 4 2U Schooner appliances, 2.5 kW three years. (Schooner has 8x the capacity of legacy servers, and can sustain 8x the throughput.) . Power and space reductions enable © 2009 Schooner Information Technology. All Rights Reserved green datacenter initiatives.

27 Database Tier: Schooner Appliance for MySQL Enterprise™

28 Schooner Appliance for MySQL Enterprise™

• Features Feature Ordinary Schooner Servers Appliance Optimized for OLTP (read-write for MySQL – intensive workloads), as well as OLAP (read mostly workloads) 512 GB flash memory  – Multi-core, high-capacity storage Reduces or eliminates  enables scalability and reduces or sharding eliminates sharding Fast recovery  Fast recovery and warm-up after – restart or failover and warm-up Built-in tools enable failover and one- Auto failover and one-  – click replication/recovery click replication and recovery Fully compatible with existing client – applications and monitoring tools Optimized InnoDB 1.0.3  Incorporates highly optimized InnoDB – 1.0.3 • Benefits 63,000 TPM 8x performance improvement over – legacy disk storage 7,000 TPM

Replaces ordinary severs at a ratio of Minute – up to 8:1, yielding both capex and

opex savings per Transactions

29 MySQL Appliance: TCO Analysis With Schooner, consolidation and Typical savings for a mid-size higher operational efficiency cuts TCO 1TB Web 2.0 datacenter for a rapidly growing 1TB MySQL workload. 3yr TCO: $563,000

Database Tier Database Without Schooner OpEx

3yr TCO: 20 2U servers, 14.7 kW (and growing) $197,000 CapEx: 20 Servers OpEx

CapEx: 2 Schooner With Schooner Appliances

2 2U Schooner appliances, 1.4 kW The Bottom Line: (Schooner has 8x the capacity of legacy servers, . Immediate capex savings and can sustain 8x the throughput.) . 62% TCO savings ($366,000) over © 2009 Schooner Information Technology. All Rights Reserved three years . Power and space reductions enable green datacenter initiatives

30 Intel SLC Flash Is a Key Enterprise Technology

COST PER READ B/W WRITE B/W ERASE LAT. READ LAT. GB HDD 100 mb/s 150.00 mb/s 5,000.00 us $0.10 NAND MLC 250 mb/s 70.00 mb/s 3.5 ms 85.00 us $3.50 NAND SLC 250 mb/s 170.00 mb/s 1.5 ms 75.00 us $11.00 NOR SLC 58 mb/s 0.13 mb/s 5,000.00 ms 0.27 us $70.00 DRAM 2,000 mb/s 2,000.00 mb/s 0.08 us $75.00

31 Limited Benefits of PCI-E or Host-Based Solutions Fusion-io* Intel® X25-E Metric io-drive SSD 8x Array Unit Note

Read latency uS 4KB 75 75 uSEC

Write Latency uS 4KB 250 85 uSEC Write latency of hardware Read CPU uS 50 15 uSEC Increases with garbage Write CPU uS 50-100 15 uSEC collection

Read MBPS 16KB 650 1400 MBPS Sustained with garbage Write MBPS 16KB 150 260 MBPS collection Read IOPS 4K 116,000 230,000 IOPS Sustained with garbage Write IOPS 4K 20,000 50,000 IOPS collection Capacity 160 512 GB Cost $7,200 $5,800 USD Cost/Gbyte $56.25 $11.33 USD * Write throughput assuming continuous random writes spread across the drive, with 20% reserve capacity configured

In typical and worst-case wear scenarios, life expectancy for SSDs in Schooner appliances ranges from 10 to 30 years:

Innodb - DBT2 Memcache 5% SET Slab (Typical) Memcache 50% SET Slab Memcache 50% SET FIFO 8 Drives 8 Drives 8 Drives 8 Drives 64 GB/drive 64 GB/drive 64 GB/drive 64 GB/drive 45,000 TPM 240,000 req/Sec 100,000 req/Sec 150,000 Req/sec 750 Neworders/second 5% SET fraction 50% SET fraction 50% SET fraction 15 Buffer writes/new order 1024 Bytes/object 1024 Bytes/object 1024 Bytes/object 4096 Bytes/page 12000 SET/Sec per node 50000 SET/Sec per node 75000 SET/sec per node 11250 Writes/sec per node 1500 Writes/Sec per drive 6250 Writes/Sec per drive 9375 Writes/sec per drive 1406 Writes/sec per drive 1.5 MB/Sec/drive 6.1 MB/Sec/drive 9.2 MB/sec/drive 5.5 MB/Ssec/drive 44739 seconds to rewrite drive 10737 seconds to rewrite drive 7158 Seconds to rewrite drive 11930 Seconds to rewrite drive 12.4 hours to rewrite drive 3.0 hours to rewrite drive 2.0 Hours to rewrite drive 3.3 Hours to rewrite drive 3.8 Write amplification 3.8 Write amplification 1 Write amplification 3.8 Write amplification 3.3 page reprogram hours 0.8 page reprogram hours 2.0 Page reprogram hours 0.9 Page reprogram hours 327041 Hours lifetime 78490 Hours lifetime 22.7 Year lifetime 87211 Hours lifetime 37.3 Year lifetime 9.0 Year lifetime 10.0 Year lifetime

Notes: . RAID5 spreads access uniformly across SSDs. . Schooner appliance for Memcached also uses hash map and multiple internal volumes to distribute load and wear. . FIFO is a write-optimizing policy that performs group writes to SSD. © 2009 Schooner Information Technology. All Rights Reserved

33 Summary: Schooner+Intel+IBM Key Advantages

. Significantly Lower TCO: Schooner appliances replace traditional servers by a ratio of up to 8:1, providing immediate capex savings and 60% opex savings

. Higher Performance: Schooner appliances provide up to 8x higher performance than traditional servers

. Seamless Compatibility: Schooner appliances are 100% compatible with existing client applications and management tools

. Quick Deployment: Schooner appliances provide plug-and-play installation for easy set-up and deployment

. Easy Management: The Schooner Administrator employs extensive monitoring and optimization features which are easily integrated with existing management tools

. Higher Reliability: Schooner appliances employ persistence, replication, and recovery software to deliver enterprise-class reliability and dramatically increase mean time between failures (MTBF)

. World-Class Support: IBM provides global, 24/7/365, single-point-of-contact service and support for every Schooner appliance

. More Revenue: In addition to improving existing datacenter operations, Schooner appliances support new revenue-producing applications enabled by fast access to terabyte-scale data © 2009 Schooner Information Technology. All Rights Reserved

34 Thank you.

35 Want More Info on SSDs?

• Attend or download these SSD-related sessions Tuesday, Sept 22nd – EBLS001 - Extending Battery Life of Mobile PCs: An Overview Wednesday, Sept 23rd – MEMS001 - Designing Solid-State Drives into Data Center Solutions – MEMS002 - Understanding the Performance of Solid-State Drives in the Enterprise – MEMS003 - Enterprise Data Integrity and Increasing the Endurance of Your Solid- State Drive – MEMS004 - Future Solid-State Drive Innovations – MEMQ002 - Open Q&A for SSD sessions Thursday, Sept 24th – MPTS006 - Extreme Notebook Design: Architecting the Most Powerful Mobile Platforms for Gaming & Workstation Applications – RESS006 - Differentiated Storage Services: Making the Most of Solid-State Drives – STOS004 - Intel® Modular Server with Intel® Solid-State Drives

• Visit our Booth #532 on Level 1 of the Tech Showcase – SSD vs HDD comparisons, gaming demo and more! • Visit us online at www.intel.com/go/ssd – Product briefs, datasheets, whitepapers, videos, technical support

36 Legal Disclaimer • INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL® PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN INTEL’S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, INTEL ASSUMES NO LIABILITY WHATSOEVER, AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF INTEL® PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT. INTEL PRODUCTS ARE NOT INTENDED FOR USE IN MEDICAL, LIFE SAVING, OR LIFE SUSTAINING APPLICATIONS. • Intel may make changes to specifications and product descriptions at any time, without notice. • All products, dates, and figures specified are preliminary based on current expectations, and are subject to change without notice. • Intel, processors, chipsets, and desktop boards 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. • Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. • Intel, Xeon and the Intel logo are trademarks of Intel Corporation in the United States and other countries. • *Other names and brands may be claimed as the property of others. • Copyright © 2009 Intel Corporation.

37 Risk Factors

The above statements and any others in this document that refer to plans and expectations for the third quarter, the year and the future are forward-looking statements that involve a number of risks and uncertainties. Many factors could affect Intel’s actual results, and variances from Intel’s current expectations regarding such factors could cause actual results to differ materially from those expressed in these forward-looking statements. Intel presently considers the following to be the important factors that could cause actual results to differ materially from the corporation’s expectations. Ongoing uncertainty in global economic conditions pose a risk to the overall economy as consumers and businesses may defer purchases in response to tighter credit and negative financial news, which could negatively affect product demand and other related matters. Consequently, demand could be different from Intel's expectations due to factors including changes in business and economic conditions, including conditions in the credit market that could affect consumer confidence; customer acceptance of Intel’s and competitors’ products; changes in customer order patterns including order cancellations; and changes in the level of inventory at customers. Intel operates in intensely competitive industries that are characterized by a high percentage of costs that are fixed or difficult to reduce in the short term and product demand that is highly variable and difficult to forecast. Additionally, Intel is in the process of transitioning to its next generation of products on 32nm process technology, and there could be execution issues associated with these changes, including product defects and errata along with lower than anticipated manufacturing yields. Revenue and the gross margin percentage are affected by the timing of new Intel product introductions and the demand for and market acceptance of Intel's products; actions taken by Intel's competitors, including product offerings and introductions, marketing programs and pricing pressures and Intel’s response to such actions; and Intel’s ability to respond quickly to technological developments and to incorporate new features into its products. The gross margin percentage could vary significantly from expectations based on changes in revenue levels; capacity utilization; start-up costs, including costs associated with the new 32nm process technology; variations in inventory valuation, including variations related to the timing of qualifying products for sale; excess or obsolete inventory; product mix and pricing; manufacturing yields; changes in unit costs; impairments of long-lived assets, including manufacturing, assembly/test and intangible assets; and the timing and execution of the manufacturing ramp and associated costs. Expenses, particularly certain marketing and compensation expenses, as well as restructuring and asset impairment charges, vary depending on the level of demand for Intel's products and the level of revenue and profits. The current financial stress affecting the banking system and financial markets and the going concern threats to investment banks and other financial institutions have resulted in a tightening in the credit markets, a reduced level of liquidity in many financial markets, and heightened volatility in fixed income, credit and equity markets. There could be a number of follow-on effects from the credit crisis on Intel’s business, including insolvency of key suppliers resulting in product delays; inability of customers to obtain credit to finance purchases of our products and/or customer insolvencies; counterparty failures negatively impacting our treasury operations; increased expense or inability to obtain short-term financing of Intel’s operations from the issuance of commercial paper; and increased impairments from the inability of investee companies to obtain financing. The majority of our non-marketable equity investment portfolio balance is concentrated in companies in the flash memory market segment, and declines in this market segment or changes in management’s plans with respect to our investments in this market segment could result in significant impairment charges, impacting restructuring charges as well as gains/losses on equity investments and interest and other. Intel's results could be impacted by adverse economic, social, political and physical/infrastructure conditions in countries where Intel, its customers or its suppliers operate, including military conflict and other security risks, natural disasters, infrastructure disruptions, health concerns and fluctuations in currency exchange rates. Intel's results could be affected by adverse effects associated with product defects and errata (deviations from published specifications), and by litigation or regulatory matters involving intellectual property, stockholder, consumer, antitrust and other issues, such as the litigation and regulatory matters described in Intel's SEC reports. A detailed discussion of these and other risk factors that could affect Intel’s results is included in Intel’s SEC filings, including the report on Form 10-Q for the quarter ended June 27, 2009.

Rev. 8/3/09