Data Protector Tape Backup Performance DATA PROTECTOR BACKUP
Data Protector Tape Backup Performance DATA PROTECTOR BACKUP PERFORMANCE WITH TAPE DRIVES 10 March 2009 page 1
Data Protector Tape Backup Performance local backup Cell manager SAN backup application server disk agent media agent tape media agent (robotic control) tape storage area network-based backup Cell manager IDB IDB 30 + 30 = 20? 60 MB/sec in. 20 MB/sec out. Gbit LAN disk agent media agent 30 MB/sec disk agent 10 March 2009 tape 30 via NDMP MB/sec 20 ? ? ? ? ? MB/sec 2: 1 Compression tape library LTO 4 Drives page 22 Page
Data Protector Tape Backup Performance In This Presentation § § § Performance - What do we mean? Is there a performance issue for HP DP Support? Proviso Why HP DP Support does not “do” performance. Why HP DP Support does help with performance. Streaming – The Secret to High Performance Stream-Fail – The Secret to Dismal Performance Data collection Analysis Options Reference 10 March 2009 page 3
Data Protector Tape Backup Performance What does performance mean? Ø In this presentation we are talking about how fast a Backup backs up data to tape. Ø This is not about Restores, although similar. Ø This is not about Virtual Tape. Ø This is about Physical Tape. o All drive references here are to LTO/Ultrium, but, except that Ultrium drives have a range of streaming rates, while other drives have a single rate, the same principles apply to all commonly used backup tape drives. 10 March 2009 page 4
Data Protector Tape Backup Performance Ø “Performance” can mean a number. Ø “Performance” can mean efficiency. Ø “Performance” can be subjective. o A customer who upgrades from DLT 8000 to LTO 3 might be pleased with the faster backups, and not realize the performance is poor compared to the capability of the LTO 3. o Or that customer might be very displeased because the LTO 3 backup could be slower than the DLT 8000 backup. We’ll see why. 10 March 2009 page 5
Data Protector Tape Backup Performance Is There a Performance Issue? Ø Calculate: Bytes backed up Divided by time Divided by number of drives Ø Equals performance: Ø how many bytes per second/minute/hour per drive. Ø Is this value within the STREAMING range for the drive? Ø No – a performance issue. Ø An issue for HP DP Support? Sometimes. 10 March 2009 page 6
Data Protector Tape Backup Performance An issue for HP DP Support? Things To Check Ø Ø Recent changes? Patches? Upgrade? New drive? Patch – does backing out the patch help? Upgrade – to 6. 00? 6. 10? o writedb/readdb to defragment the Filenames tablespace, then keep an eye on it (omnidbutil –info). Ø New drive? Customer claims “no compression”? o Probably not a Data Protector issue. More likely the backup is not streaming, which takes longer and reduces tape capacity. 10 March 2009 page 7
Data Protector Tape Backup Performance Borderline cases Ø Added an object to the backup and now it takes twice as long. Ø Two backups of same size, one takes much longer. Ø “Same” backup in another Cell is twice as fast. Ø In cases like these, HP DP Support probably will explain streaming, give advice, URLs, proviso (slide 10). 10 March 2009 page 8
Data Protector Tape Backup Performance Proviso (1) Ø AFTER conversation with customer, understanding the issue and agreeing to provide some guidance for a performance issue, HP DP Support will send advice along with a proviso. 10 March 2009 page 9
Data Protector Tape Backup Performance Proviso (2) For Customers _____________________________________ Please note that, in the absence of errors, performance analysis and tuning is not a Response Center service for Data Protector. However, I am familiar with some of the issues and may be able to offer some advice. If my advice does not help, you may wish to obtain the services of a performance specialist. There is excellent performance guidance at this website: HP Surestore and Storage. Works - Performance Troubleshooting and Using Performance Assessment Tools http: //h 20000. www 2. hp. com/bizsupport/Tech. Support/Document. jsp? object. ID=lpg 50460 This whitepaper for the Ultrium 960 discusses performance issues. http: //h 71028. www 7. hp. com/ERC/downloads/5982 -9971 EN. pdf If email puts a space in the URL, just remove it. There are no spaces in these URLs. _____________________________________ 10 March 2009 page 10
Data Protector Tape Backup Performance Why HP DP Support Does Not “do” performance Ø HP DP Support supports the Data Protector product, not its entire environment. Ø HP DP Support does not control the environment that Data Protector operates in. Ø HP DP Support personnel are not and cannot be sufficiently familiar with customer’s Data Protector environment. Ø There are many factors outside Data Protector that affect performance. Ø Performance analysis and tuning is its own specialty. 10 March 2009 page 11
Data Protector Tape Backup Performance Why HP DP Support DOES help with performance HP DP Support personnel: • Know some of the factors that customer might not know. • Know some of the common problems and solutions. • Know of Data Protector issues. • Know the options in Data Protector. 10 March 2009 page 12
Data Protector Tape Backup Performance S T R E A M I N G The obscure, misunderstood, esoteric, recondite, abstruse, arcane, shadowy, unseen, unnoticed component in tape backup performance. 10 March 2009 page 13
Data Protector Tape Backup Performance Streaming – The Secret to High Performance Ø Streaming: moving the tape continuously during a backup Never stop/start. Ø Streaming requires that data be delivered to the drive above the minimum streaming rate. Ø Tape drives perform best when they are streaming. Ø Tapes hold the most data when streaming – 100% of capacity. Ø Performance degrades severely and precipitously when streaming is not achieved (“stream-fail”). Ø Stream-fail dramatically reduces capacity. 10 March 2009 page 14
Data Protector Tape Backup Performance Streaming Rate Varies with Compression(1) Ø No compression, the drive writes one byte of data for each one byte received. One byte in – one byte out. Ø 2: 1 compression, the drive writes one byte for each two bytes received. Two bytes in – one byte out. Ø 4: 1 compression, the drive writes one byte for each four bytes received. Four bytes in – one byte out. Ø 8: 1 compression, the drive writes one byte for each eight bytes received. Eight bytes in – one byte out. Ø The greater the compression, the faster the data must be delivered to write that one byte. 10 March 2009 page 15
Data Protector Tape Backup Performance Streaming Rate Varies with Compression (2) Ø LTO 1 minimum streaming rate with no compression is 6 MB/second. Ø LTO 1 minimum streaming rate with 2: 1 compression is 12 MB/second. Ø LTO 1 minimum streaming rate with 4: 1 compression is 24 MB/second. Ø LTO 1 minimum streaming rate with 8: 1 compression is 48 MB/second. Ø LTO 4 minimum streaming rate with 8: 1 compression is 320 MB/second! Ø Data must be delivered at these rates so the drive can stream the data onto the tape. Ø Lower data delivery rate results in stream-fail. 10 March 2009 page 16
Data Protector Tape Backup Performance S T R E A M I N G Big deal! Of course modern tape drives are fast. So what? 10 March 2009 page 17
Data Protector Tape Backup Performance Ø People naturally expect that when the rate of data delivery slows down the backup will slow down at about the same rate. 10 March 2009 page 18
Data Protector Tape Backup Performance It’s much worse than that! Ø Performance drops precipitously, down to less than 1% of best performance, because of repeated repositioning (0. 25 – 3 seconds each time). 10 March 2009 page 19
Data Protector Tape Backup Performance Stream-Fail – The Secret to DISMAL Performance Ø Stream-fail o Costs Time o Costs Capacity Ø When the rate of data delivery drops below the streaming rate, the tape repositions – stop, reverse, forward, stop – known as the “shoe-shining” effect. This is “stream-fail”. Ø Each stream-fail costs 0. 25 - 3 seconds. Ø Each stream-fail leaves 10 -100 MB of position-markers on the tape between data blocks. Ø Stream-fail can reduce performance to less than 1% and capacity to 8% or less! We Want STREAMING. L We Don’t Want “S T R E A M – F A I L” L bad bad bad bad 10 March 2009 page 20
Data Protector Tape Backup Performance True Story (1) Ø RCE said customer’s new LTO 3 was not compressing. Ø How did he know? How did he conclude this? Ø The tape was full at 50 GB and it took 8 hours to get full! Ø Hmmmm. LTO 3 at 2: 1 compression can put 800 GB on a tape and will put 50 GB on that tape in 6 -18 minutes, but needs to receive 54240 MB/second to do it. Ø NOTE that the compression factor is stated. It is needed to calculate the streaming rate for compressed data. L He was backing up one filesystem from a simple disk. L Reading the filesystem at perhaps 6 MB/s – far short of 54 MB/s. L NOT STREAMING! NOT EVEN CLOSE! L He had 50 GB of data intermixed with 750 GB of non-data gaps from stream-fails! 10 March 2009 page 21
Data Protector Tape Backup Performance True Story (2) Stream-Fail Reduces Capacity Streamed tape ------------------------------------------------------------------------------------------------------------------ datadatadatadatadatadatadatadatadata ------------------------------------------------------------------------------------------------------------------ J data = 100% of tape J J Elapsed time = 6 -18 minutes J Stream-failed tape -------------------------------------------------------------------------------------------------------------------- data. L s p a c e Ldata -------------------------------------------------------------------------------------------------------------------- L data = 8% of tape L L non-data. L = 92% of tape L L Elapsed time = 480 minutes L L = 8 hours L L 0. 8% of best performance L 10 March 2009 page 22
Data Protector Tape Backup Performance Did It Stream? 1 TB in ten hours @2: 1 (1) 1 x LTO 1 (40 -100 GB/hour) 2 x LTO 1 3 x LTO 1 1 x LTO 2 (67 -200 GB/hour) 2 x LTO 2 3 x LTO 2 1 x LTO 3 (190 -560 GB/hour) 1 x LTO 4 (290 -860 GB/hour) 10 March 2009 - Streaming at top speed! - ? ? ? - Not - ? ? ? - ? ? ? page 23
Data Protector Tape Backup Performance Did It Stream? 1 TB in ten hours @2: 1 (2) 1 x LTO 1 (40 -100 GB/hour) 2 x LTO 1 3 x LTO 1 1 x LTO 2 (67 -200 GB/hour) 2 x LTO 2 3 x LTO 2 1 x LTO 3 (190 -576 GB/hour) 1 x LTO 4 (288 -864 GB/hour) 10 March 2009 - Streaming at top speed! - Streaming at 50 GB/hour. - Not! 33 GB/hour. - Streaming at 100 GB/hour. - Not! 50 GB/hour. - Not! 33 GB/hour. - No way! 100 GB/hour. - Sorry! 100 GB/hour. page 24
Data Protector Tape Backup Performance DATA COLLECTION 10 March 2009 page 25
Data Protector Tape Backup Performance Data from Customer ¯ How many tape drives? ¯ How many tapes? ¯ How much data was backed up? ¯ How long did it take? ¯ What kind of tape drive(s)? ¯ What generation cartridges? (LTO 2 cartridge in LTO 3 drive? ) ¯Needed to calculate whether the backup streamed. Ø Network backup or local backup? Ø What is the network speed? Ø How many Objects? (Filesystem on UNIX; Disk on Windows) Ø Recent changes? 10 March 2009 page 26
Data Protector Tape Backup Performance NETWORK BACKUP OR LOCAL BACKUP? Cell Console Network Backup Local Backup Disk Agent TCP/IP Cell Manager TCP/IP Session Manager Media Agent 10 March 2009 Shared Memory Scheduler TCP/IP Session Manager TCP/IP Media Agent page 27
Data Protector Tape Backup Performance Data in DP Session Ø Session Report from GUI or CLI Ø Command line omnidb –session 2005/05/21 -3 –report Ø Reports bytes per backup. Ø Use session_devices report to get bytes per drive. For capacity, you still must determine bytes per tape if multiple tapes. 10 March 2009 [Normal] From: BMA@server 2. com "drive_5" Time: 05/21/05 03: 01: 23 STARTING Media Agent "drive_5" [Normal] From: BMA@server 2. com "drive_5" Time: 05/21/05 03: 01: 25 Loading medium from slot 50 to device /dev/rmt/2 mn [Normal] From: OB 2 BAR@server 1. com “server 1" Time: 05/21/05 03: 02: 39 Starting OB 2 BAR Backup: /dbs 01/0 (dbspace) [Normal] From: OB 2 BAR@server 1. com “server 1" Time: 05/21/05 03: 18: 47 Completed OB 2 BAR Backup: /dbs 01/0 (dbspace) [Normal] From: BMA@server 2. com "drive_5" Time: 05/21/05 12: 22: 47 COMPLETED Media Agent "drive_5“ Backup Statistics: Session Queuing Time (hours) --------------------Completed Disk Agents. . . . Failed Disk Agents. . . Aborted Disk Agents. . -------------------- 0. 00 62 0 0 Disk Agents Total . . . 62 ==================== Completed Media Agents. . . . 1 Failed Media Agents. . 0 Aborted Media Agents. . 0 -------------------- Media Agents Total . . 1 ==================== Mbytes Total. . . 270983 MB Used Media Total. . . 1 Disk Agent Errors Total. . . 0 page 28
Data Protector Tape Backup Performance “Display Statistical Information”(1) Options tab Other tab Display statistical info Advanced options 10 March 2009 page 29
Data Protector Tape Backup Performance “Display Statistical Information” (2) Ø Enable “Display Statistical Info” in the backup specification (Backup GUI, select backup, Options tab, Filesystem Options, Advanced button, Other tab) Ø Statistical information is displayed for each Disk Agent in the Session Report. [Normal] From: OB 2 BAR@server 1. com “server 1" Time: 05/21/05 03: 02: 39 Starting OB 2 BAR Backup: /dbs 01/0 (dbspace) Directories……… Regular Files…. . ------------------------Objects Total…. . 0 1 1 Kbytes Total…. . 6181958 Note that this is KB, not MB. ----- At completion of Disk Agent: [Normal] From: OB 2 BAR@server 1. com “server 1" Time: 05/21/05 03: 18: 47 Backup Profile: Run Time. . . 0: 16: 08 Backup Speed. . . . 6386. 32 (KB/s) 10 March 2009 Note that this is KB, not MB. ----- page 30
Data Protector Tape Backup Performance Session Objects Report Ø To see each object’s performance omnirpt -report session_objects -session <sess_ID> omnirpt –report session_objects -session 2005/05/21 -3 Session Objects Report Cell Manager: server 1. com Creation Date: 05/23/05 11: 29: 03 Object Type Client Mountpoint Description Status Mode Start Time Duration [hh: mm] Size [k. B] # Files Performance [MB/min] Protection # Errors # Warnings Device ________________________________________ BAR server 1. com /dbs 01/0 Informix Completed full 05/21/05 03: 01: 48 0: 16 6181958 1 373. 81 07/16/05 03: 01: 48 0 0 drive_5 BAR server 1. com /dbs 02/0 Informix Completed full 05/21/05 03: 17: 57 0: 14 6187042 1 405. 05 07/16/05 03: 17: 57 0 0 drive_5 10 March 2009 page 31
Data Protector Tape Backup Performance Session Devices Report Ø To see a tape drive’s write rate omnirpt -report session_devices -session <sess_ID> omnirpt -report session_devices -session 2005/05/21 -3 Session Devices Report Cell Manager: server 1. com Creation Date: 05/23/05 11: 28: 18 Device Start End Duration GB Written Perf [GB/h] # Objects # Media ________________________________________ drive_5 05/21/05 03: 01: 28 05/21/05 12: 19: 28 9: 18 264. 31 28. 42 62 1 This is per drive, not per tape. 10 March 2009 page 32
Data Protector Tape Backup Performance Ob 2 Tape. Statistics(1) Ø global file option Ø Disabled by default # Ob 2 Tape. Statistics=0 or 1 # default: 0 # If enabled, this option allows tape statistics logging into # media. log file. 10 March 2009 page 33
Data Protector Tape Backup Performance Ob 2 Tape. Statistics(2) Ø Set in global file on Cell Manager o Windows C: Program FilesOmni. BackConfigServerOptionsglobal C: Program. DataOmni. BackConfigServerOptionsglobal o UNIX /etc/opt/omni/server/options/global Ø Logged to media. log on Cell Manager o Windows C: Program FilesOmni. BacklogServermedia. log C: Program. DataOmni. BacklogServermedia. log o UNIX /var/opt/omni/server/log/media. log 10 March 2009 page 34
Data Protector Tape Backup Performance Ob 2 Tape. Statistics(3) Ø Entries are placed in media. log upon close of media 05/21/05 12: 21: 45 cf 98 e 98 f: 403 a 4182: 2 e 50: 0001 "[DKL 002] INFORMIX_60" [TAPE WRITE STATISTICS] logical drive=drive_5 errsubdel=59586 errposdel=0 total=744 toterrcorr=744 totcorralgproc=0 totb=97714108800 totuncorrerr=0 Ø Ø Ø errsubdel errposdel total toterrcorr totcorralgproc totb o o o = errors corrected with substantial delays = errors corrected with possible delays = total number of re-writes = total errors corrected = total number of times correction algorithm processed = total blocks processed, after compression. This field has different units for different drive types. For many drives it’s bytes. For LTO it’s the number of datasets, which is the Data Protector data within the data blocks. For LTO 1/2 the size of the dataset is 403, 884 compressed bytes. Slightly larger for LTO 3/4. This value divided into the bytes sent to the drive (see Session report or Session Devices report) gives the compression ratio. totb x 403884 bytes = data-bytes-written-to-tape bytes-sent-to-drive-in-session / (data-bytes-written-to-tape 1 + data-bytes-written-to-tape 2 +. . . ) = compression ratio NOTE: NEED TO SUM totb FOR ALL THE TAPES WRITTEN BY THE BACKUP BECAUSE WE DO NOT HAVE SESSION-DATA-TO-DRIVE TOTALS FOR INDIVIDUAL TAPES. Ø totuncorrerr = total uncorrected errors q Usually error counts will be very low. High error rates degrade performance and Merit a hardware call. 10 March 2009 page 35
Data Protector Tape Backup Performance A N A L Y S I S 10 March 2009 page 36
Data Protector Tape Backup Performance Measuring Backup Performance Ø What is the performance of each Disk Agent? o Are there enough DAs to achieve data delivery at the minimum streaming rate? Ø Compute overall compression ratio for the session. o Find all the tapes in media. log and add up “totb” for them, divide into bytes backed up. Ø Calculate what would be a streaming number of tapes used. (For LTO 3, one tape per 800 GB at 2: 1) Ø Overall, how many bytes were sent to each drive? bytes-to-drives / session duration / # of drives = bytes per time per drive. o 1. 6 TB per 2 hours per 2 drives = 800 GB per hour per 2 drives = 400 GB per hour per drive o Is that a streaming rate for LTO 3 with LTO 3 cartridge at 2: 1 compression? Ø Is this within streaming range for the drive? 10 March 2009 page 37
Data Protector Tape Backup Performance Troubleshooting Media Agent Performance Ø Performance is: bytes written per time period per drive. Ø Use “omnirpt -report session_devices -session <sess_ID>” plus Ob 2 Tape. Statistics to calculate compression rate. Ø Must total the statistics for every tape written to all the drives that the backup used. Ø It is possible to calculate per drive only if you can determine which drive used which tapes. Ø UNFORTUNATELY, if a backup uses two or more drives, the only way to determine which tapes were used in which drives is to use the Backup Session report and list the storage slots used by the BMAs, and to know which tapes were in those slots at the time of the backup. That slot-tape information is not available after a tape has been moved. Ø Test a drive’s performance independently of Data Protector with LTT - Library and Tape Tools http: //h 18006. www 1. hp. com/products/storageworks/ltt/index. html? jumpid=re g_R 1002_USEN 10 March 2009 page 38
Data Protector Tape Backup Performance Troubleshooting Disk Agent (disk) performance Ø Use “session_objects” report Ø Enable DP’s “Display Statistical Info” o Shows object Backup Profile/Statistics in session report Ø Run VBDA (Volume Backup Disk Agent) standalone o The standalone test shows what the Disk Agent actually can do without being slowed down by tape drive repositioning. o Run test backup to /dev/null (Unix) or C: nul (Windows). o “-profile” means “Display statistical info”. /opt/omni/lbin/vbda –vol /opt –trees /opt/omni –out /dev/null –profile Kbytes Total ………. 774347 Run Time …………. . 0: 01: 10 Backup Speed ……. 11062. 10 (KB/s) 10 March 2009 page 39
Data Protector Tape Backup Performance local backup Troubleshooting Network Performance network-based backup Cell manager SAN (LAN-free) backup IDB disk agent media agent via NDMP disk agent 10 March 2009 tape page 40 40 Page
Data Protector Tape Backup Performance Troubleshooting Network performance Ø Network backup = DA and MA on different hosts. Ø Network must have sufficient capacity to move data at high rates. o 10 Base. T (<1 MB/sec) and 100 Base. T (<10 MB/sec) are slow compared to current tape drive transfer rates. o 1000 Base. T (<100 MB/sec) can handle some network backups. o 1000 Base. T is insufficient for LTO 3/LTO 4 much above their minimum streaming rates (54/80 MB/s) for no more than one of those drives. o Better performance is possible using SAN. Ø TEST: ftp large or many files between DA and MA hosts. Ø TEST: Run test backup to a Data Protector drive defined with /dev/null (Unix) or nul (Windows). Ø TEST: Use PAT - Performance Analysis Tools http: //www. hp. com/support/pat 10 March 2009 page 41
Data Protector Tape Backup Performance - OPTIONS - SOLUTIONS - TUNING 10 March 2009 page 42
Data Protector Tape Backup Performance Tuning for Performance (1) Ø Ensure current patches, firmware, drivers are installed o Data Protector, Operating System, Drives, Tape Library, NSR, SAN Switch, etcetera. Ø Software Compression o Don’t use it - causes high CPU overhead. o If used, except for Ultrium/LTO, disable hardware compression, otherwise non-LTO drive will produce larger blocks and run slower. Ø Hardware Compression o On by default. 10 March 2009 page 43
Data Protector Tape Backup Performance Tuning for Performance (2) Data Protector Settings Ø Block Size o o Equivalent to data transfer size. Should be at least 64 KB. Use the Data Protector default setting, except for LTO/Ultrium. See “Whitepaper for the Ultrium 960” link in Reference section. Ø Segment Size o Defines the amount of data DP writes to tape before a catalog segment is written. o Increasing this parameter will improve the importing speed of tapes and may improve write performance. o Uses more memory on the MA host. 10 March 2009 page 44
Data Protector Tape Backup Performance Tuning for Performance (3) Data Protector Settings Ø Disk Agent buffers o Set in the Data Protector tape drive definition. o It’s the number of buffers set up in memory for Disk Agents on both the Disk Agent host(s) and the BMA host. o The memory is shared if both the DAs and BMA are on the same host. o Default of 8 is reasonable. o Might help to increase number of buffers. Usually not helpful. But in one case, Media Copy that took 24 hours was reduced to 4 hours by increasing DA buffers to 8 from 32. o Memory used on the MA host • Block Size x DA Buffers x Concurrency 10 March 2009 page 45
Data Protector Tape Backup Performance Tuning for Performance (4) Data Protector Settings Ø Concurrency o Specifies the number of Disk Agents writing simultaneously to a Media Agent. o Range of 1 -32, default is 4. o Has negative effect on restore performance when not doing complete restore of the backup. o Keep to minimum needed for tape drive streaming. Ø Data Protector IDB logging level o Limit logging to level required for restore. 10 March 2009 page 46
Data Protector Tape Backup Performance Tuning for Performance (5) Data Protector Settings Ø CRC checking o More useful with older, less robust tape & drive technology. o Allows discovery of data corruption AFTER the backup, during Verify or Restore. o High CPU usage. • See “HP Data Protector software performance white paper” in Reference section: In tests with LTO 3, enabling CRC reduced backup performance by 20%. 10 March 2009 page 47
Data Protector Tape Backup Performance Tuning for Performance (6) Data Protector Settings Ø Software Compression o Helps with limited capacity LAN. o Except for LTO/Ultrium drives, disable hardware compression, otherwise, drive performance will drop and data blocks will expand. • UNIX - device file • Windows - N at the end of the SCSI path specification o High CPU • See “HP Data Protector software performance white paper” link, page 54, in Reference section: Figure 36 shows that enabling the software compression increased the CPU load from 13% to 99%. The CPU load was very high because Data Protector compressed five file systems in parallel. 10 March 2009 page 48
Data Protector Tape Backup Performance Tuning for Performance (7) Data Protector Settings Ø Set Object Order by Size Backups typically run slower, often dramatically slower, near the end. o Concurrency effectively drops because there are no additional Disk Agents to replace a completed Disk Agent so fewer and fewer Disk Agents are running. o With reduced Concurrency, the data delivery rate drops below the drive’s minimum streaming rate. This causes a precipitous drop in performance. o Rearrange the order in which Objects will be backed up so that Objects of about the same size are grouped together and backed up concurrently. • On the Summary tab for a Backup Specification, or on the Summary page of the wizard for a new Backup Specification, change the Object order by right-click, “Move up” or “Move down”. • The Objects stay listed in the same order but the new Object Order is shown on the far right, so either stretch the window or scroll right to see the ordering. • The Object listing can be rearranged by clicking on the “Order” column heading. 10 March 2009 page 49
Data Protector Tape Backup Performance Tuning: Data Protector Tape Drive Options (1) Settings Sizes Devices & Media 256 KB for Ultrium 960 and newer. See “Whitepaper for the Ultrium 960” in Reference section Advanced options 10 March 2009 page 50
Data Protector Tape Backup Performance Tuning: Data Protector Tape Drive Options (2) Settings Concurrency setting Devices & Media CRC Check Advanced options 10 March 2009 page 51
Data Protector Tape Backup Performance REFERENCE 10 March 2009 page 52
Data Protector Tape Backup Performance LTO/Ultrium Best Data Transfer Rates (1) No Compression LTO/Ultrium-1 230 LTO/Ultrium-2 460 LTO/Ultrium-3 960 LTO/Ultrium-4 1840 15 MB/s Ultrium-1 media 30 MB/s Ultrium-2 media 80 MB/s Ultrium-3 media 120 MB/s Ultrium-4 media ~50 GB/hour ~100 GB/hour ~288 GB/hour ~430 GB/hour 15 MB/s Ultrium 1 media 30 MB/s Ultrium-2 media 80 MB/s Ultrium-3 media 20 MB/s Ultrium-1 media 30 MB/s Ultrium-2 media 2: 1 Compression LTO/Ultrium-1 230 LTO/Ultrium-2 460 LTO/Ultrium-3 960 LTO/Ultrium-4 1840 30 MB/s Ultrium-1 media 60 MB/s Ultrium-2 media 160 MB/s Ultrium-3 media 240 MB/s Ultrium-4 media ~100 GB/hour ~200 GB/hour ~576 GB/hour ~860 GB/hour 40 MB/s Ultrium-1 media 60 MB/s Ultrium-2 media 160 MB/s Ultrium-3 media 40 MB/s Ultrium-1 media 60 MB/s Ultrium-2 media 10 March 2009 page 53
Data Protector Tape Backup Performance LTO/Ultrium Best Data Transfer Rates (2) 4: 1 Compression LTO/Ultrium-1 230 LTO/Ultrium-2 460 LTO/Ultrium-3 960 LTO/Ultrium-4 1840 60 MB/s Ultrium-1 media 120 MB/s Ultrium-2 media 320 MB/s Ultrium-3 media 480 MB/s Ultrium-4 media ~200 GB/hour ~400 GB/hour ~1, 152 GB/hour ~1, 728 GB/hour 80 MB/s Ultrium 1 media 120 MB/s Ultrium-2 media 320 MB/s Ultrium-3 media 80 MB/s Ultrium-1 media 120 MB/s Ultrium-2 media 8: 1 Compression LTO/Ultrium-1 230 LTO/Ultrium-2 460 LTO/Ultrium-3 960 LTO/Ultrium-4 1840 120 MB/s Ultrium-1 media 240 MB/s Ultrium-2 media 640 MB/s Ultrium-3 media 960 MB/s Ultrium-4 media ~200 GB/hour ~800 GB/hour ~2, 304 GB/hour ~3, 456 GB/hour 160 MB/s Ultrium-1 media 240 MB/s Ultrium-2 media 640 MB/s Ultrium-3 media 160 MB/s Ultrium-1 media 240 MB/s Ultrium-2 media 10 March 2009 page 54
Data Protector Tape Backup Performance LTO/Ultrium Data Rate Matching – Streaming Range LTO/Ultrium-1 230 LTO/Ultrium-2 460 LTO/Ultrium-3 960 LTO/Ultrium-4 1840 No Compression 6 -15 MB/s ~20 -50 GB/hour 10 -30 MB/s ~35 -100 GB/hour 27 -80 MB/s ~95 -288 GB/hour 40 -120 MB/s ~144 -432 GB/hour 2: 1 Compressed 12 -30 MB/s ~40 -100 GB/hour 20 -60 MB/s ~70 -200 GB/hour 54 -160 MB/s ~190 -576 GB/hour 80 -240 MB/s ~288 -864 GB/hour 4: 1 Compressed 24 -60 MB/s ~80 -200 GB/hour 40 -120 MB/s ~140 -400 GB/hour 108 -320 MB/s ~380 -1, 152 GB/hour 160 -480 MB/s ~576 -1, 728 GB/hour 8: 1 Compressed 48 -120 MB/s ~160 -400 GB/hour 80 -240 MB/s ~280 -800 GB/hour 216 -640 MB/s ~760 -2, 304 GB/hour 320 -960 MB/s ~1, 152 -3, 456 GB/hour Assumes same generation cartridge 10 March 2009 page 55
Data Protector Tape Backup Performance LTO/Ultrium Tape Capacity When Streaming End-to-End LTO/Ultrium-1 230 LTO/Ultrium-2 460 LTO/Ultrium-3 960 LTO/Ultrium-4 1840 No Compression 100 GB 200 GB 400 GB 800 GB 1. 6 TB 3. 2 TB 6. 4 TB 2: 1 Compression 200 GB 4: 1 Compression 400 GB 8: 1 Compression 800 GB 1. 6 TB Assumes same generation cartridge These capacities apply for any streaming rate. 10 March 2009 page 56
Data Protector Tape Backup Performance TYPICAL FILE COMPRESSION RATIOS Data type Typical compression CAD 3. 8: 1 Spreadsheet/word processing 2. 5: 1 Typical file/print server 2. 0: 1 Lotus Notes DB 1. 6: 1 Microsoft Exchange/SQL DB 1. 4: 1 Oracle/SAP DB 1. 2: 1 From: Enterprise Backup solution Design Guide (Ninth edition: February 2008), page 73 http: //h 20000. www 2. hp. com/bc/docs/support/Support. Manual/c 00775232. pdf NOTE: 10 March 2009 In limited sampling from customers, filesystem backups of Oracle databases compress 8: 1 (probably due to unused blocks)! 8: 1 requires VERY high data delivery rate to achieve streaming! page 57
Data Protector Tape Backup Performance Analysis and Tuning (1) Ø Performance Topics in Data Protector’s Online Help • DP GUI, Help menu, Help Topics • Contents tab • Expand “Backup” (click on [+]) • Expand”Backup Performance” (click on [+]) Ø 16 Articles, some brief, some a bit old but valid 10 March 2009 page 58
Data Protector Tape Backup Performance Analysis and Tuning (2) Ø HP Data Protector software performance white paper http: //h 71028. www 7. hp. com/ERC/downloads/4 AA 13836 ENW. pdf o Page 4 • This white paper provides performance-related information for HP Data Protector software 6. 0 together with some typical examples. The emphasis is on backup servers and two common backup and restore performance questions: – Why are backups so slow? – Why are restores so slow? 10 March 2009 page 59
Data Protector Tape Backup Performance Analysis and Tuning (3) Ø Introduction to Performance Tuning for HP-UX UPERFKBAN 00000726 (updated 2006) http: //saw. cce. hp. com/km/saw/view. do? doc. Id=emr_nac 00958823 Ø Troubleshooting Backup Performance - OVKBRC 00006135 (updated 2004) http: //saw. cce. hp. com/km/saw/view. do? doc. Id=ucr_na. KMNOVKBRC 00006135 10 March 2009 page 60
Data Protector Tape Backup Performance Analysis and Tuning (4) Ø Whitepaper for the Ultrium 960 o Discusses performance issues. http: //h 71028. www 7. hp. com/ERC/downloads/59829971 EN. pdf o Recommends 256 KB blocksize for Ultrium 960. o Data Protector default setting is 64 KB for Ultrium. o On Windows, for blocksize greater than 64 KB, Maximum. SGList registry entry for the HBA may need to be adjusted. Ø HP Storage. Works Enterprise Backup Solution design guide o Chapter 8 Performance: Finding bottlenecks http: //h 20000. www 2. hp. com/bc/docs/support/Support. Man ual/c 00775232. pdf 10 March 2009 page 61
Data Protector Tape Backup Performance Analysis Tools (1) Ø PAT - Performance Analysis Tools http: //www. hp. com/support/pat Ø LTT - Library and Tape Tools http: //h 18006. www 1. hp. com/products/storageworks/ltt/in dex. html? jumpid=reg_R 1002_USEN 10 March 2009 page 62
Data Protector Tape Backup Performance LTO / Ultrium (1) Ø LTO - Linear Tape-Open http: //en. wikipedia. org/wiki/Linear_Tape-Open Ø Quick. Specs HP Storage. Works Ultrium http: //h 18006. www 1. hp. com/products/quickspecs/11739_ div/11739_div. pdf Ø “Testing LTO-3 vs. LTO-2 and SDLT” o Info. Stor magazine article (2005) o Detailed discussion of testing and of performance factors. http: //www. infostor. com/article_display/testing-lto-3 -vslto-2 -and-sdlt/223974/s-articles/s-infostor/s-volume-9/sissue-3/s-lab-review/s-1. html 10 March 2009 page 63
Data Protector Tape Backup Performance LTO / Ultrium (2) Data Rate Matching Ø HP Ultrium tape drive, DRM - Data Rate Matching http: //www. sunddslto. com/upload. Links/DRM_Technical_Overview_LTO 4. pdf 10 March 2009 page 64
Data Protector Tape Backup Performance LTO / Ultrium (3) Ø User Guide: HP Storage. Works Ultrium http: //bizsupport. austin. hp. com/bc/docs/support/Support Manual/c 01141242. pdf Ø Includes very basic performance guidance o Pages 58 -59 Can your system deliver the required performance? 10 March 2009 page 65
Data Protector Tape Backup Performance LTO / Ultrium (4) Ø IBM: LTO: A better format for mid-range tape – great description of the robust Ultrium datablock design http: //www. research. ibm. com/journal/rd/474/jaquette. html Ø HP Storage. Works Ultrium 960 tape drive technical white paper http: //h 71028. www 7. hp. com/ERC/downloads/59830148 EN. pdf 10 March 2009 page 66
Data Protector Tape Backup Performance Consulting - HP Services Ø HP Consulting o HP Valu. Pack Services http: //h 20219. www 2. hp. com/services/cache/109579 -0 -0 -225 -121. html#services Ø “HP Services Valupak Consulting” o 2 -page brochure ftp: //ftp. hp. com/pub/services/perevent/valupack/vpbro chure. pdf 10 March 2009 page 67
- Slides: 68