Intel 240gb User's Manual
Contents
1. 95 95 95 95 Search Engine Cache 4KB 8KB 16KB 16KB 95 95 95 95 Content Delivery Network 16KB 95 70 95 Based on these usage trends small transfer sizes ranging from 4KB to 16KB and above are common in enterprise and data center applications Also much emphasis is placed on random accesses and although there are varied levels of read and write December 2013 329903 001US White Paper 7 intel Intel SSD DC S3500 Workload Characterization in RAI D Configurations 4 4 4 5 mixes read intensive workloads are more prominent In summary it is important to select the proper SSD for a particular workload The examples presented here use 100 write and 100 read workloads to show the maximum performance in these areas Also 7096 3096 read write and 9096 1096 read write are used in order to simulate typical workloads in the datacenter Drive Endurance Drive endurance or wear is an important consideration when selecting an SSD for a particular application The Intel SSD DC 3500 Series drive is a standard endurance enterprise class drive designed for read heavy workloads It is important to understand how drive wear is affected by the RAID level A RAID level that uses dedicated parity such as RAID 4 will write all parity to a single drive This can potentially cause the parity drive to wear faster than the other drives in the set Distributed parity RAI2. but the progression is not as steep through deeper queues ending between 500 600 uS Average latency for 9096 read and 10096 read continue to improve due to the higher speed of reads over writes Figures 2 3 4 The latency of the two drive set is lower than other drive counts as read percentage increases due to the manner in which the LSI controller deals with the additional drives This is expected behavior For more information please contact LSI for details Figure 4 At 100 read the I OPS performance scales linearly In other words the I OPS for four drives is double that of two drives six drives is triple that of two drives and eight drives is four times that of two drives White Paper December 2013 12 329903 001US Intel SSD DC S3500 Workload Characterization in RAI D Configurations in tel 5 3 RAID 1 Consistency Consistency behind a RAID controller is very important because the performance of any RAID set is limited by the lowest performing drive As a RAID set increases in number of drives the probability of any given drive performing poorly increases Therefore if the model of drive used is inconsistent in its performance the inconsistency increases with the size of the RAID set The Intel SSD DC 3500 Series drive has shown excellent consistency when used as a single drive Figure 5 illustrates the consistent of DC S3500 in RAID sets Notice that the maximum latency is grouped very tightly for both the two d
3. by how the particular application scales and the IO profile of the workload produced by the application The following workload specific characteristics have a direct impact on the ability of the SSD to produce IO e Read Write Mix NAND programming writes and read timing reads differ significantly at the hardware level Because of the higher controller overhead required for processing writes the number of read I OPS are often higher than write IOPS Real world workloads are most often a mix of read and write e Random Sequential Mix IOPS can vary depending on the ratio of sequential versus random accesses With higher random write workloads more data movement and greater data management activity occurs in the drive As random write activity increases the IOPS serviceable to the host typically decreases e Queue Depth Higher queue depths typically allow the SSD to generate higher OPS through concurrent processing of commands However as the queue size increases latency will be negatively impacted e Random Transfer Block Size With a smaller transfer size the SSD controller has to work harder to maintain the logical to physical address mappings In addition the smaller the transfer size the larger the logical space needed for its mapping Once logical space constraint is reached background re mapping will take place These frequent events slow IOPS e Available Spare Area A larger spare area directly impacts random wri
4. drive set possibly consolidating the 4KB blocks into the 256KB stripes Read performance is also very good with eight drives reaching 300K IOPS with 4KB blocks and queue depth 8 per drive Mixed workloads show very good performance with 7096 read hitting 75K IOPS at queue of 8 per drive and 90 read coming in at almost 140K both with 4KB blocks Latency on all workloads is very manageable although as the queue depth increases so does the latency The graphs show that as queues grow latency increases at an increasingly higher rate To summarize In both RAID 1 and RAID 5 the Intel SSD DC S3500 Series drive shows excellent scalability performance and consistency e Very little latency was introduced by the RAID controller in RAID 1 In RAID 5 the overhead and latency are slightly higher e In random mixed read write workloads SSDs perform significantly as much as 100 times better than HDDs in a similar situation e With this RAID controller there is the possibility of greater performance by adding more than eight drives in both RAID 1 and RAID 5 configurations White Paper December 2013 18 329903 001US Intel SSD DC 3500 Workload Characterization in RAI D Configurations intel 8 0 8 1 Appendix RAI D Levels RAID Redundant Array of Independent Disks developed in 1988 to improve performance reliability and scalability of hard disk storage systems has become a standard in datacenters because of these qualit
5. AID 9265 8i controller card e 2x up to 8x Intel SSD DC 3500 Series 480GB drives BIOS configuration changes e Hyper Threading disabled RAID controller configuration e 256KB Striping default e No Read Ahead e Write Through e Direct O Windows Drive Configuration e Basic disk e GUID partition table e Simple volume e Use full available space e NTFS format Test Software configuration e lOMeter 2009 10 22 e 1x worker per drive in RAID set 1 The system was selected to make sure the performance of the RAID card and the SSDs would not be inhibited by the server 2 Hyper Threading is disabled in this test system specifically due to additional latency introduced during benchmark testing In any practical application Hyper Threading would NOT be disabled 3 In the configuration of the RAID set No Read Ahead and Write through are used due to the speed of the SSDs Read and write caching was designed for use with HDDs Caching with SSDs introduces additional overhead thus interfering with the SSDs performance 4 One thread or worker per drive was used in order to simulate the manner in which many applications utilize storage and also to attempt to saturate the communication channels to the SSDs December 2013 White Paper 329903 001US 9 intel Intel SSD DC S3500 Workload Characterization in RAI D Configurations 5 2 Intel SSD DC S3500 Series in RAI D 1 Performance Characterization Data This section provides per
6. D levels RAID 5 and RAID 6 reduce this issue RAID 1 and RAID 5 as tested in this example shows very consistent wear across all drives in the RAID sets This is due to the tests using the full LBA space of the RAID set thereby not creating any hotspot activity Selection of RAI D Controller There are many quality RAID controllers on the market today with varying levels of performance features and price points Below are the important features considered in selecting the RAID controller for this sample test e RAID levels available e Controller chipset e PCle version e SAS SATA speed e Internal External ports e Compatibility with SSDs Within these categories the LSI MegaRAID 9265 8i was chosen for the following reasons e RAID O 1 5 6 10 50 60 are supported e LSISAS2208 Dual Core RAID on Chip ROC e 1 GB 1333MHz DDR3 SDRAM Cache e x8 PCle 2 0 e 6Gb s per port e 8 internal SAS ports e SSD support e Relative availability popularity in the industry White Paper December 2013 8 329903 001US Intel SSD DC 3500 Workload Characterization in RAI D Configurations intel 5 0 RAID 1 5 1 Test System Specifications The system used for RAID 1 testing include the following e Intel R2208GZ4GC IDD 2U rack mount server e Intel S2600GZ server board e 2x Intel Xeon E5 2690 8 core CPUs 2 9 GHz e Intel C602 chipset e 192GB DDR3 1333 memory e Microsoft Windows Server 2008 R2 64 bit e LSI MegaR
7. R aRaRGaYa d 18 8 0 LDuDnIMDpQuM 19 8 1 RAID WU CVElS CM nm 19 December 2013 White Paper 329903 001US 3 Tables Table 1 Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 White Paper 4 Intel SSD DC S3500 Workload Characterization in RAI D Configurations Typical Mixed Workloads in Data Center Applications sss sees 7 RAID 1 Random 100 Write 9 4KB Transfer Size with Average Latency 11 RAID 1 Random 70 Read 4KB Transfer Size with Average Latency 11 RAID 1 Random 90 Read 4KB Transfer Size with Average Latency 12 RAID 1 Random 100 Read 4KB Transfer Size with Average Latency 12 RAID 1 Maximum Latency for 2 drive and 8 drive Configurations 13 RAID 5 Random 100 Write 4KB Transfer Size with Average Latency 15 RAID 5 Random 70 Read 4KB Transfer Size with Average Latency 15 RAID 5 Random 90 Read 4KB Transfer Size with Average Latency 16 RAID 5 Random 100 Read 4KB Transfer Size with Average Latency 16 RAID 5 Maximum Latency for 3 drive and 8 drive Configurations 17 December 2013 329903 001US Intel SSD DC 3500 Workload Characterization in RAI D Configurations in tel 1 0 Revision History Document Revision are Lu Number Number Desc
8. al to the system used for RAID 1 testing except the following changes e 2x Intel Xeon E5 2680 8 core CPUs 2 7 GHz e 3x up to 8x Intel SSD DC S3500 Series 800GB drives For this test 800GB drives were used The rated performance of the Intel SSD DC S3500 Series drive in 800GB 600GB and 480GB capacities are nearly identical per internal Intel testing Intel SSD DC S3500 Series in RAI D 5 Performance Characterization Data This section provides performance characterization data for the Intel SSD DC 3500 Series in RAID 5 configurations To establish baseline expectations for IOPS the Intel SSD DC S3500 Series 800 GB drives were evaluated in RAID 5 sets of 3 4 5 6 7 and 8 drives The data collected was based on a different mix of read and write random and sequential workloads Since higher queue depths can sometimes yield higher IOPS queue depths of 1 2 4 and 8 per drive were used in the test setup Multiple transfer sizes were tested however only selected data is presented here All tests were done using the entire Logical Block Address LBA range of the virtual drive Tests were repeated at least twice to validate results Drives were prepared using OMeter to fill the entire user area of the drive with data Then the first workload of each type Sequential or Random was 100 write performed for 120 minutes Each subsequent workload was run for 12 minutes with average IOPS collected over the final 10 minutes of the run Th
9. e there is a steady increase in performance both as drives are added and as the queue deepens Figures 7 8 9 As read percentage increases the exponential increase in latency is not as prominent with deeper queues This is due to the speed at which reads are performed 6 3 RAID 5 Consistency Consistency behind a RAID controller is very important because the performance of any RAID set is limited by the lowest performing drive As the number of different drives in a RAID set increases so does the likelihood of any given drive performing poorly Therefore if the model of drive used is inconsistent in its performance the inconsistency increases with the size of the RAID set The Intel SSD DC S3500 Series drive has shown excellent consistency when used as a single drive Figure 10 illustrates that the DC S3500 is also very consistent in RAID sets The three drive and eight drive data shows the maximum latency is grouped very tightly indicating that adding more drives would have little impact on consistency White Paper December 2013 16 329903 001US Intel SSD DC S3500 Workload Characterization in RAI D Configurations in tel Figure 10 RAID 5 Maximum Latency for 3 drive and 8 drive Configurations Intel internal testing October 2013 6 4 RAI D 5 Performance Conclusions The RAID 5 write performance data illustrates the additional processing power required of the RAID controller to calculate parity and stripe data across mu
10. e following figures show different levels of performance for a selection of configurations The scale of the IOPS charts is variable in order to clearly show the change as drives are added White Paper December 2013 14 329903 001US Intel SSD DC 3500 Workload Characterization in RAI D Configurations intel Figure 6 RAID 5 Random 100 Write 4KB Transfer Size with Average Latency Intel internal testing October 2013 NOTES There are gains in write performance as drives are added to the RAID 5 set The change at queue depth 1 from three drives to six drives is approximately 58 increase in IOPS For eight drives the change is 9796 increase in I OPS over the three drive set At a queue depth of 1 latency increases as more drives are added to the RAID set most likely caused by the additional overhead of calculating parity and striping across more drives This increases as the queue deepens Figure 7 RAID 5 Random 70 Read 4KB Transfer Size with Average Latency Intel internal testing October 2013 December 2013 White Paper 329903 001US 15 in tel Intel SSD DC 3500 Workload Characterization in RAI D Configurations Figure 8 RAID 5 Random 90 Read 4KB Transfer Size with Average Latency Intel internal testing October 2013 Figure 9 RAID 5 Random 100 Read 4KB Transfer Size with Average Latency Intel internal testing October 2013 Notes Figures 7 8 9 As the workloads become more read intensiv
11. eesesesesesesesee sese nenas a nena sunu k uana ununi ana nnn nan 5 3 0 About This Guide eeeeeeeseeeeeeee enne nennen eens esas nun un entera nn ananas nnne nnn 5 4 0 OVGFVIQW iere covec v execute kae xvetevaud epo a eaaa OS AR VR RE RUNE DVD ERE NEEU CREDE ERR aa UN EON 5 4 1 What Impacts SSD IO Performance sssssssssssess eene 6 4 2 Queue Depth and Latency sssssssssssssssssssssesememne eee ee ee emen eee meses 7 4 3 Why Mixed Workload Is Important sessssssse eene 7 4 4 Drive End ratite i ied retener uel cesa ag gane t An ra n a Fg n c a b nara 8 4 5 Selection Of RAID Controller esses mne eene nnns 8 5 0 D5i og me 9 5 1 Test System Specifications 2 0 0 ccc neers 9 5 2 Intel SSD DC S3500 Series in RAID 1 Performance Characterization Data 10 5 3 RAID 1 Consistency ee EPM UTR II RAIMUNDI E 13 5 4 RAID 1 Performance Conclusions sess 13 6 0 RAID c 14 6 1 Test System SpecificaLblORis ierit arie Et DAEA temp e Re piu reta ER RR iN EN 14 6 2 Intel SSD DC S3500 Series in RAID 5 Performance Characterization Data 14 6 3 RAID 5 Performance ConsiStency ccccceee eee eee ene 16 6 4 RAID 5 Performance Conclusions sss emen nennen 17 7 0 Summary eesesssseseeueusunu auae A RARE AR GR RR GRRRRRGGUSARRRRARSRRTRSRRRRZRRTASRRRRERRTRYRRRRERRRRSRRR R
12. epths can increase IOPS with read intensive workloads Obtaining the best performance for a particular application requires balance The challenge is to achieve high speed or IOPS at an acceptable latency level This white paper presents lower queue depths of 1 2 4 and 8 per drive The results shown demonstrate favorable speed and I OPS generation without pushing latency to extreme levels Why Mixed Workload Is Important Mixed random workloads are predominant in data center and enterprise applications Intel SSDs have been deployed in a variety of these applications ranging from content delivery and video on demand networks to Internet datacenter portals and database management servers Although these applications see unique IO traffic across the storage drive there are commonalities in their usage of random and read write mixed workloads Table 1 shows an overview of transfer sizes read write mixes and randomness in commonly used workloads in data center applications These are based on commonly available industry information and information available through such benchmarks as TPC C TPC E TPC H and TPOX which attempt to mimic these real world applications Typical Mixed Workloads in Data Center Applications Application Transfer Size 9o Random Read Web Servers 4KB 8KB 16KB 75 95 Exchange Email Database OLTP 4KB 4KB 8KB 95 95 70 70 Decision Support Video On Demand 16KB 16KB
13. formance characterization data for the Intel SSD DC 3500 Series in RAID 1 configurations To establish baseline expectations for IOPS the Intel SSD DC S3500 Series 480GB drives were evaluated in RAID 1 sets of 2 4 6 amp 8 drives The data collected was based on a different mix of read and write random and sequential workloads Since higher queue depths can sometimes yield higher IOPS queue depths of 1 2 4 amp 8 per drive were chosen in the test setup Multiple transfer sizes were tested however only selected data is presented here All tests were done using the entire LBA range of the virtual drive Tests were repeated at least twice to validate results Drives were prepared using lOMeter to fill the entire user area of the drive with data Then the first workload of each type Sequential or Random was 100 write performed for 120 minutes Each subsequent workload was run for 12 minutes with average IOPS collected over the last 10 minutes of the run The following figures show the different levels of performance for a selection of configurations Note The scale of the IOPS charts is variable to clarify the changes that occur as drives are added White Paper December 2013 10 329903 001US Intel SSD DC 3500 Workload Characterization in RAI D Configurations in tel Figure 1 RAID 1 Random 100 Write 4KB Transfer Size with Average Latency Intel internal testing October 2013 Notes Figure 1 The write performance
14. hes the single drive specification and increases linearly as drives are added to the set With eight drives at queue depth 1 per drive this configuration processes over 50K write IOPS with 4KB blocks The RAID controller adds very little latency but latency does increase as the queue depth grows and as more drives are added to the array Read performance increases with additional drives and with queue depth reaching over 200K read I OPS with eight drives a queue of 8 per drive and 4KB blocks More importantly the performance on mixed workloads was excellent and increased as more drives were added The 7096 read workload topped out at close to 110K IOPS with 4KB blocks and the 90 went to nearly 150K IOPS with 4KB blocks both at queue of 8 per drive The highest throughput seen in this test was 2300 MB s during 10096 read using eight drives with transfer size of 128KB and a queue of 8 per drive This means the bandwidth limit of the PCle lanes was not reached 4000 MB s for x8 PCle 2 0 It is theoretically possible that more than eight drives could be used and an increase in performance obtained In RAID 5 configurations write operations increase both as drives are added and as the queue depth increases The increase in write performance with queue depth is somewhat surprising and is most likely attributed to the RAID controller and the scaling effect of the cache in each drive This may be due to the way the controller writes the stripes to the
15. ies There are many types or levels of RAID RAID 0 uses block level striping to span one or more drives This does improve performance and increases capacity when more than one drive is used However there is no fault tolerance so failure of any one drive will cause full data loss RAID 1 also called mirroring writes data identically to two drives producing a mirrored set Reads can be serviced by either drive and writes occur in unison on both drives If one drive has a hardware failure the data is protected in the mirrored copy RAID 1 requires two drives Many modern RAID controllers support RAID 1 sets of more than two drives however the original specification was for only two Because of the 5096 overhead RAID 1 is the most expensive RAID type RAID 2 uses bit level striping with dedicated Hamming code parity This is a theoretical model and not used in practice RAI D3 uses byte level striping with dedicated parity This level is not commonly used RAID 4 uses block level striping with dedicated parity All parity data is on a single drive I O requests are handled in parallel increasing performance RAID 5 uses block level striping with distributed parity Data and parity are distributed among all drives and requires all but one drive to be present RAID 5 requires at least three drives and can survive a single drive failure RAID 6 uses block level striping with double distributed parity Identical to RAID 5 in the way
16. ign 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 Contact your local Intel sales office or your distributor to obtain the latest specifications and before placing your product order Copies of documents which have an order number and are referenced in this document or other Intel literature may be obtained by calling 1 800 548 4725 or go to http www intel com design literature htm Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors Any change to any of those factors may cause the results to vary You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases including the performance of that product when combined with other products Intel and the Intel logo are trademarks of Intel Corporation in the U S and other countries Other names and brands may be claimed as the property of others Copyright 2013 Intel Corporation All Rights Reserved White Paper December 2013 2 329903 001US Intel SSD DC 3500 Workload Characterization in RAI D Configurations in tel Contents 1 0 Revision H StOTY PM A 5 2 0 Supporting Documentation
17. intel Intel SSD DC S3500 Series Workload Characterization in RAI D Configurations White Paper December 2013 329903 001US intel Intel SSD DC 3500 Workload Characterization in RAI D Configurations 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 UNLESS OTHERWISE AGREED IN WRITING BY INTEL THE INTEL PRODUCTS ARE NOT DESIGNED NOR INTENDED FOR ANY APPLICATION IN WHICH THE FAILURE OF THE INTEL PRODUCT COULD CREATE A SITUATION WHERE PERSONAL INJURY OR DEATH MAY OCCUR 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 des
18. it writes data however parity is written twice in different locations RAID 6 can survive 2 drive failures therefore it is often used for larger sets of drives RAID levels can also be nested for improved performance or fault tolerance RAID 10 0 1 50 and 60 are common combinations December 2013 White Paper 329903 001US 19
19. ltiple drives There is diminished performance gain be adding drives when compared to RAID 1 Intel s data also shows that in mixed workloads and in pure reads RAID 5 performs well reaching over 300K IOPS in 10096 read at a queue of 8 per drive on eight drives As the workloads become more read heavy latency drops from a high of 2 2 ms 10096 write to a low of 140 us 10096 read The highest throughput achieved was 2400 MB s with eight drives 10096 read 128KB transfer size and queue depth of 8 This leaves room for possible improvement by adding more drives to the RAID set In configurations where RAID 5 would traditionally be used SSDs would provide significant performance gain over HDDs Additionally RAID 5 with SSDs could be used in situations where RAID 5 with HDDs would not perform well The consistency of the drives is well demonstrated in these tests and shows that Intel SSD DC S3500 Series drives consistently offer higher performance with excellent stability even behind a RAID controller December 2013 White Paper 329903 001US 17 intel Intel SSD DC S3500 Workload Characterization in RAI D Configurations Summary The Intel SSD DC S3500 Series drive has proven itself in many applications where speed and reliability are essential The data in this paper shows that this drive is very robust behind a RAID controller In RAID 1 configurations write performance is as expected for this setup that is a two drive set matc
20. of the two drive RAID 1 set matches the write performance of a single Intel DC 3500 drive This indicates very low latency introduced by the RAID controller As more drives are added the write performance scales linearly At four drives the performance is 2x that of a single drive at six drives it is 3x higher and at eight drives it is 4x This is true at all queue depths tested and at transfer sizes from 4KB to 128KB In this case queue depth does not affect performance significantly At a queue depth of 1 the average latency for 100 write at 4KB transfer size is less than 200 us Latency increases as the queue deepens ending at 1 4ms for a queue of 8 It is interesting to note that latency is not affected by the number of drives Figure 2 RAID 1 Random 70 Read 4KB Transfer Size with Average Latency Intel internal testing October 2013 December 2013 White Paper 329903 001US 11 in tel Intel SSD DC 3500 Workload Characterization in RAI D Configurations Figure 3 RAID 1 Random 90 Read 4KB Transfer Size with Average Latency Intel internal testing October 2013 Figure 4 RAID 1 Random 100 Read 4KB transfer size with Average Latency Intel internal testing October 2013 Notes Figures 2 3 In mixed workloads 7096 read and 90 read IOPS increase with additional drives and show slightly exponential growth with deeper queues Figures 2 3 4 Average latency for 7096 read starts out similar to 10096 write
21. ription Revision Date 2 0 Supporting Documentation For more information on Intel SSDs see the corresponding documentation o o poe ee iii Locarion Intel Solid State Drive DC 3500 Series Product Specification 328860 3 0 About This Guide This guide describes Intel SSD DC S3500 Series performance characteristics in RAID configurations across multiple workloads and provides analysis to help optimize performance The audience is technical IT professionals Systems Storage Database and Application Engineers 4 0 Overview The Intel SSD DC S3500 Series provides high random read and write storage Input Output Operations per Second IOPS across mixed read and write workloads This high random performance and the consistency of IOPS under workload deliver robust and scalable operation when used behind a RAID controller Data centers can benefit in both performance and TCO by using the Intel SSD DC S3500 Series in the appropriate applications Compared to the approximately 200 300 random IOPS that a single 15K SAS hard disk drive HDD can provide an Intel SSD DC S3500 Series operates at much higher I OPS up to 75 000 IOPS for random 4KB reads and up to 11 500 IOPS for random 4KB writes over the entire span of the SSD The Intel SSD performance numbers are based on the Intel product specification sheet as derived from internal Intel testing With real world workloads the I OPS that any particular device can produce will va
22. rive and eight drive RAID sets indicating there is very little change in consistency as more SSDs are added Figure 5 RAID 1 Maximum Latency for 2 drive and 8 drive Configurations Intel internal testing October 2013 5 4 RAID 1 Performance Conclusions RAID 1 is a very good choice for data needing robust replication The RAID controller used shows good bandwidth with low latency causing little to no effect on read and write speeds of the SSDs The linear scaling of read and write performance with additional drives shows that adding more drives would provide good ROI in most applications The highest throughput seen in this test was 2300 MB s during 10096 read using eight drives with transfer size of 128KB and a queue of 8 per drive This means the theoretical bandwidth limit of the PCle lanes was not reached 4000 MB s for x8 PCle 2 0 It is theoretically possible that more than eight drives could be used and obtain an increase in performance depending on the latency introduced by the necessary SAS expander The consistency of the drives is well demonstrated in these tests and shows that Intel SSD DC S3500 Series drives provide high performance with excellent stability even behind a RAID controller December 2013 White Paper 329903 001US 13 intel Intel SSD DC S3500 Workload Characterization in RAI D Configurations 6 1 Note 6 2 Note RAID 5 Test System Specifications The system used for RAID 5 testing was identic
23. ry depending on several factors the application s ability to produce IOPS the ratio of random to sequential access the block transfer size the queue depth the read write mix of the workload and overall resource utilization in the server running the workload This guide presents data for RAID 1 and RAID 5 configurations due to their prominence in the datacenter Additional RAID levels are currently being tested and will be presented in future revisions or as separate papers December 2013 White Paper 329903 001US 5 in tel Intel SSD DC S3500 Workload Characterization in RAI D Configurations A selection of workloads that represent both best case performance and real world performance are presented in this white paper These scenarios give the IT professional a better understanding of the capabilities of the Intel SSD DC 3500 Series drive when used in conjunction with a hardware RAID controller More importantly it helps the IT professional understand a variety of workloads and circumstances in which Intel SSD technologies will accelerate those workloads and provide business value for their organization 4 1 What Impacts SSD IO Performance Although Intel SSDs excel in delivering random read and write I OPS it is important to remember that more IO activity at the application level results in higher CPU utilization in the applications host In addition to the abilities of the SSD IO performance in any particular situation is dictated
24. te and mixed read write performance by minimizing the frequency of reclaim activities and freeing up processor cycles to support more host read write requests You can increase the spare area by over provisioning the SSD See the Intel High Performance SATA Solid State Drive Over Provisioning an Intel SSD White Paper for more information In summary the following principles of storage are often true concerning queue depth block size randomness and per 1O transactional latency e As queue depth increases IOPS increase and latency increases e As block size increases throughput increases and latency increases e As randomness increases I OPS decreases and latency increases White Paper December 2013 6 329903 001US Intel SSD DC 3500 Workload Characterization in RAI D Configurations in tel 4 2 4 3 Table 1 Queue Depth and Latency Latency The amount of time needed to service one outstanding 10 to the drive measured in milliseconds ms or with SSDs microseconds us The Intel SSD DC S3500 Series supports a maximum queue depth of 32 per drive In a RAID array the queue depth is multiplied by the number of drives in the RAID set Example In a RAID 5 set of 8 drives the maximum total queue depth would be 256 8 X 32 As more commands are queued in the SSD average latency is impacted Our internal testing indicates that average latency increases sharply with queue depths beyond 8 However these high queue d
Download Pdf Manuals
Related Search