InstantForum.NET v2.9.0SQLServerCentralhttp://www.sqlservercentral.com/Forums/notifications@sqlservercentral.comSat, 25 May 2013 11:20:34 GMT20http://www.sqlservercentral.com/Forums/Topic1451602-377-1.aspxWithin a VM environment that is using one massive array of hard disks. is it really beneficial to create 2 VM Drives out of the same array to separate your MDF and LDF files?it would seem if I am carving out 2 drives from the same array of disks, it sort of defeats the purpose of separation. the array will still have to do sequential and random read/writes on the same physical disk, i.e. no performance gain.or am I totally missing the concept of VM Environments and how carving out disk drives work?Fri, 10 May 2013 07:53:55 GMTSQL_Padrehttp://www.sqlservercentral.com/Forums/Topic1421899-377-1.aspxGreetings to all -- I have been tasked with evaluating storage options for our server, and the LSI Nytro NWD-BLP4-800 PCIe SSD card was mentioned. Does anyone have any experience with this make/model? Thanks in advance --sqlnycTue, 19 Feb 2013 17:23:37 GMTsqlnychttp://www.sqlservercentral.com/Forums/Topic1416762-377-1.aspxWe have a very large client that will have 5000+ stores that will be connecting in on a web app and I'm looking for the best architectural design to make everything run super efficiently. Putting aside how it's coded, etc. what are some recommendations for best performance when considering the high number of incoming connections? The application is 80% reads and about 20% writes/updates. We batch insert new data in at night or through the day at regular intervals. We have thought about possibly doing a partitioned approach (10 servers) using a load balancer to point certain store number ranges to certain servers (ie - at the network level (reading packet to get first digit of the store number or something like that). Or we could think about letting all connections come in to one server for reads and another for writes and then doing merge or peer to peer replication between them. Or do we just keep everything in one really beefy server and use snapshot isolation?Any recommendations?Wed, 06 Feb 2013 16:21:08 GMTSQLJockyhttp://www.sqlservercentral.com/Forums/Topic1411273-377-1.aspxHi,Looking for some opinions/links. If a SQL server (OLTP) is restricted to a single box with max 8 drives, would four RAID1 arrays (OS,TEMPDB,LOGS,DATA) be the way to go?Or a RAID1 for OS and a 6 disk RAID10 for the rest?Or RAID1 OS, RAID1 TEMPDB, RAID10 for the rest?ThanksThu, 24 Jan 2013 11:15:04 GMTipgsqlhttp://www.sqlservercentral.com/Forums/Topic1399050-377-1.aspxWe're setting up a new database server for our OLTP and smaller website databases, and disk performance is a big area of concern. We have 3 disk arrays available to us, and 64GB of RAM.The 3 storage systems are:Inline SAS HDD (2-disks 15K)DAS SAS HDD (4-disks 15K)DAS SAS SSD (8-disks)On these 3 storage arrays, we have 6 main types of storage.System Data Files (master, msdb, model, distribution)TempDb Data Files (8 data files for 8 cores)OLTP Data FilesOLTP Data Indexes (separate file)Website Data Files (much lower use than OLTP)All The Log Files (System, OLTP, Web, etc)We planned on distributing our files on the storage systems like this:Inline HDD (OS) – Don’t want much here because it houses each of the OS’s - System Data Files - TempDb Data File (1 of 8 data files)MD HDD (4-disk) - All The Log Files - TempDb Data Files (3 of 8 data files)MD SDD (8-disk) - OLTP Data Files - OLTP Data Indexes - Website Data Files - TempDb Data Files (4 of 8 data files)Analyzing our current system, the largest I/O requirements are for TempDb, followed by OLTP Data files - so those are our main focus. We are looking into the option of allocating 8GB more RAM to the server to create a RAMDisk (not taking away any of the 64GB already allocated to SQL). The RAMDisk would be used entirely for 4 of the 8 TempDb data files, isolating most of the TempDb work in RAM. Then rearrange the rest a bit to better isolate TempDb and OLTP data files:Inline HDD (OS) – Don’t want much here because it houses each of the OS’s - System Data FilesRAMDisk (8GB) - TempDb Data Files (4 of 8 data files - 2GB each)MD HDD (4-disk) - All The Log Files - Website Data FilesMD SSD (8-disk) - OLTP Data Files - OLTP Data Indexes - TempDb Data Files (4 of 8 data files - 2GB each)To me, this solution looks much better, and from what I read we should get some major performance boosts as a result. There are lots of guides out there showing how to use RAMDisks for TempDb performance improvements, but other experts say to stick to the SQL Recommended Best Practices which doesn't include RAMDisks. My question is, has anyone here used RAMDisks for TempDb, and what are their thoughts and experience?Thu, 20 Dec 2012 10:19:58 GMTkevin.kembelhttp://www.sqlservercentral.com/Forums/Topic1400976-377-1.aspxHey Folks,I'm a bit confused regarding the Latency numbers reported by SQL Server Virtual File Stats (sys.dm_io_virtual_file_stats) compared to the Latency reported by the SAN Itself.According to my Systems Engineers, the SAN is only reporting "avg read\write latency" of ~20 whereas teh Virtual File Stats Read Latency of 93 and Write Latency of 87.How are we getting such a vastly different number and how can I get an apples to apples comparison?the query I'm using is below (the Systems Engineers are using the Tools Built into the SAN to get their figures). The SAN is a NetApp 2040.\SELECT sample_ms,DB_NAME(vfs.database_id) AS database_name ,vfs.database_id ,vfs.FILE_ID ,io_stall_read_ms / NULLIF(num_of_reads, 0) AS avg_read_latency ,io_stall_write_ms / NULLIF(num_of_writes, 0)AS avg_write_latency ,io_stall / NULLIF(num_of_reads + num_of_writes, 0)AS avg_total_latency ,num_of_bytes_read / NULLIF(num_of_reads, 0)AS avg_bytes_per_read ,num_of_bytes_written / NULLIF(num_of_writes, 0)AS avg_bytes_per_write ,vfs.io_stall ,vfs.num_of_reads ,vfs.num_of_bytes_read ,vfs.io_stall_read_ms ,vfs.num_of_writes ,vfs.num_of_bytes_written ,vfs.io_stall_write_ms ,size_on_disk_bytes / 1024 / 1024. AS size_on_disk_mbytes ,physical_nameFROM sys.dm_io_virtual_file_stats(NULL, NULL) AS vfsJOIN sys.master_files AS mf ON vfs.database_id = mf.database_idAND vfs.FILE_ID = mf.FILE_IDwhere vfs.database_id =5 ORDER BY avg_total_latency DESCThanksGreg JacksonFri, 28 Dec 2012 10:36:44 GMTGregoryAJacksonhttp://www.sqlservercentral.com/Forums/Topic1400124-377-1.aspxI hear it is best to have at least three disk controllers for these items:-SQL Server main installation (master database and other databases)-tempdb -tlogsIs there any other portion of SQL server that should go somewhere else? With SQL Server on a virtual server, how can these be put on separate hardware when a SAN supports the physical storage?Tue, 25 Dec 2012 19:36:27 GMTGolfer22http://www.sqlservercentral.com/Forums/Topic1394820-377-1.aspxMy first post so hello to everyone for the first time :)I work for a company who make a large ERP product and my role is to spec hardware for the onsite Managed Service i.e. we manage from hardware up but in the customers data centre. There are around 10 non DB Server VM's and 2 SQL Server VM's for Production and a Dev that hosts multiples of Production database.Currently we use the following systemsVMWare ESXi 5.1 + vCenterHP DL380 x 2 | 64GB | Quad Core Xeons x 2 HP iSCSI P2000 | 10k for VM OS Drives and 15k for SQL Server 2008GB Cisco Switches x 2I am now working on an IBM solution and for a similar cost I can get the same spec with a few less disks in the SAN and [b]1 x 365 GB Fusion I/O[/b] card for one of the hosts. I have a 765GB one on loan at the moment and wow its fast but that's just in ATTO and IOMeter tests..Some of our customers have 200GB DB's plus BLOB tables stored on slower disks.Do you think it would be better to fill the VM Hosts with loads more memory and allocate this to the virtualised DB server or to stick with 64GB and use a Fusion I/O card to host the database?1 x 8GB DIMM = £751 x 16GB DIMM = £1601 x 365GB Fusion I/O card = £3508If my memory allocation is bigger than my database what are my bottlenecks? CPU and Disk Writes?If you have any ideas on hardware HP or IBM then please let me have them :) Go wild...Mon, 10 Dec 2012 16:56:21 GMTParaffinhttp://www.sqlservercentral.com/Forums/Topic1368992-377-1.aspxContinuing the topic http://www.sqlservercentral.com/Forums/Topic1352495-377-1.aspxHey Guys,I keep having the same problem, but now I have some useful informations, see below:First, we have changed the MAX DEGREE OF PARALLELISM (http://msdn.microsoft.com/en-us/library/ms181007(v=sql.105).aspx) parameter from 0 to 4, but we have some problems in maintenance then we return to 0, but when this parameter was on 4, we had the same CPU problema.Second, we find the "auto update statistics asynchronously" option that was on default "false", we change to "true", because we had the suspect that when a lot of queries are using the same tables, the SQL was unable to update the statistics, but 1 hour after change this parameter, we had the same CPU problem.But, when we was looking the SQL, we find the same query a lot of times, this query is used multiple times by a lot of threads from the Weblogic server in the same time, like 150 executions simultaneously on the same table.So, we start to question about the OUTPUT and INSERTED impact.UPDATE nextEntryIn SET STATUS = @P1 , DATE_TIME_PROCESSING = getDate() OUTPUT INSERTED.COD_QUEUE_IN, INSERTED.PRIORITY, .... few more fields FROM (SELECT TOP 5 COD_QUEUE_IN, PRIORITY, .... the same few more fields FROM QUEUE_IN WHERE STATUS = @P2 AND CONSUMER = @P3 AND PRIORITY = @P4 ORDER BY COD_QUEUE_IN ASC) AS nextEntryInAgain, when the CPU goes high, we make the rebuild on this tables and the data of companies table.Thanks for any help.Fri, 05 Oct 2012 06:44:45 GMTvladimir.guerreirohttp://www.sqlservercentral.com/Forums/Topic1352495-377-1.aspxHi people,I have a Windows Cluster with 3 instances of SQL Servers, conected true a Fiber Channel network to a HP Storage 4400 and 6400.All my instances are running ok, with a normal e acceptable CPU, but in one instance we have a big problem.In one instance, that runs a relational application, this server is exclusive to this instance and 95% of utilization is for just one base that have 200gb.My server is a HP blade with 132gb RAM, 2 Intel Xeon E5540 2,53 Gb, 16 cores.Normally, this server is has 15% CPU average, every sunday I run a rebuild and update statistics of all index.We have an application on a Weblogic server running against this base and 3 Webmethods server with many services connecting in this base, but, despite this, the applications run very well, but, eventually, the CPU goes on top of server, and stays on 100% for ever.This base have one table with around 3 billions of lines of the last 6 months that is related with other table with registration data of companies with around 160k lines.Normally, when the problem happens, the only solution we have is to rebuild the primary key index of the data of companies table, doing this action, the CPU turns back to 10% and the applications back to running normally.Have anyone a problem relative with this? This problem are causing a big nuisance to our company.Thanks.Thu, 30 Aug 2012 14:39:40 GMTvladimir.guerreirohttp://www.sqlservercentral.com/Forums/Topic1358756-377-1.aspxHi all,At present our DB servers and applications is running on 32-bit environment.We are planing to migrate DB servers from 32-bit to 64-bit on fail over cluster.But as all the servers are in live we can't stop users from accessing the solution.So we need to install 64-bit on one server and then migrate all the applications to that server and then Install 64-bit on another server and bring them to cluster and should be same as 32-bit environment.Please suggest me the best way to proceed further.Thu, 13 Sep 2012 11:49:38 GMTKoteswara http://www.sqlservercentral.com/Forums/Topic1310336-377-1.aspxHi All,I'm working on a new SQL Server landscape for the region I'm responsible for, today we have 14 MS SQL Server installations spread across 7 different physical sites.The sites are all connected trough our Corporate WAN, which for most sites gives us only 512kbit to 1Mbit connection speed.Now we want to centralised some of the SQL Server databases so that we can reduce the total number to 7 SQL Server installations, most of them will be centralized into 1 big SQL Server in the main site of the region (which currently has a 4Mbit connection).Connection speed is not that fast, but it is sufficient today. After centralization we will need to upgrade some lines, but actually I would like to look at other solutions which could improve SQL Server communication.Does anybody have any experiences with this?http://www.nitrosphere.net/If it works it sound like a pretty good solution, and not that expensive. The only problem is that their website doesn't really give me confidence...Does anyone have any other solutions?Thanks,KoenMon, 04 Jun 2012 01:25:10 GMTKoen Wuyshttp://www.sqlservercentral.com/Forums/Topic1283839-377-1.aspxI would like to bui a SQL server from scrat.I need some guidance on what softwa I neeed.Mentioned below are my requiremen.I) planning to build 2 node sql2008 cluster.I need to know what all softwares I need apart from sql server.This would be all on one physical box .I would pref microsoft softwares as I have license for all of those.Sun, 15 Apr 2012 16:52:14 GMTcurious_sqldbahttp://www.sqlservercentral.com/Forums/Topic1333350-377-1.aspxGreetings fellow DBAs --Just started a new DBA gig, and have unfortunately inherited a textbook version of how NOT to use SQL Server. Client is running 2008R2 Enterprise, with a db size of ~100GB. Due to massive performance problems they have already moved to SSD on PCI Express, and enabled Row Committed Snapshot Isolation, because of frequent deadlocks. The SQL environment is not run in-house, but instead is hosted. They are getting quotes on moving to a new hosting provider, and the specs I saw mentioned internal 15k magnetic drives. I warned them that this would likely be a mistake. Now they want to see a bench mark that would determine the potential performance hit if making this kind of switch. I realize that an apples-to-apples comparison like this might not have been done, so I'll phrase it another way. Is there a way to configure 15k drives either internal to the server or DAS that would perform in the ballpark of SSD PCI Express drives? Thanks very much in advance for any advice. Sat, 21 Jul 2012 08:14:52 GMTsqlnychttp://www.sqlservercentral.com/Forums/Topic1153930-377-1.aspxplan to upgrade to Windows 7, but which version (home premium, pro or ultimate) will run 2008 Express? Thanks, JerryWed, 03 Aug 2011 16:33:33 GMTjerry-986350http://www.sqlservercentral.com/Forums/Topic1233108-377-1.aspxHi,In my company, we have a SQL Server 2008 R2 Cluster, a few days ago, we doubled the CPU and memory of one cluster node.We have 1 HP BL680C G7 with 2 Intel Xeon X7550 2Ghz and 64Gb RAM.We doubled this hardware, now we have 64 Logical Processors and 131Gb RAM.The problem is, my server is VERY slowly, I have to set SQL Server to use just 2 processors and 60Gb RAM, and this way, it works.My doubt is, we need to configure or execute something to make SQL use all new hardware?Thanks.Guerreiro.Tue, 10 Jan 2012 06:08:31 GMTvladimir.guerreirohttp://www.sqlservercentral.com/Forums/Topic1149678-377-1.aspxHi,Helo helpCan i install sql server 2008 on Windows 2000 server system,.Please anyone advise methanksJerryWed, 27 Jul 2011 14:56:52 GMTsolomon.jernashttp://www.sqlservercentral.com/Forums/Topic1211348-377-1.aspxLets say ive run extensive tests with SQLIO and found for my particular workload the ideal blocksize is 64kb.I ran the command [i]fsutil fsinfo ntfsinfo T: [/i] and can see the drive currently has a cluster size of 4kb.Should i now reformat the drive with a cluster size of 64kb?Im pretty sure thats what i do, but i just want to be certain.Wed, 23 Nov 2011 14:06:15 GMTwinston Smithhttp://www.sqlservercentral.com/Forums/Topic1218259-377-1.aspxHi folks,it's time for a (partial) hardware replacement and our company decided to provide a SAN as a replacemenet of the dedicated drives of let's say 10 server with two drives each, originally. To be specific: there'll be only one rather small drive per server just to hold the OS and the pagefile, eventually. (I say "eventually" since the the folks responsible for the sizing decided to use a standard config of less than 50GB, regardless of the servers memory. My question how they would provide a full memory dump for a server with 256GB memory in case of a support request to MS is "still under investigation"... but that's a different story).The SAN will have 15 spindles in total and RAID0, RAID1, and RAID5 groups.There will be about 50 SQL Server DBs (ranging from just a few MB up to 80GB) as well as several applications.If I'm not completely wrong there'll be only max. 7 spindles net to hold all the LUNs (based on the RAID level and the physical number of spindles).Even though the total size of storage capacity is enough to meet our needs, the number of physical spindles seems to be too low. From my perspective the better disc performance (e.g. 15krpm vs. currently 10krpm) will not compensate the parallel access from several server to the same spindle(s). My question "Why not reduce the disc size in favor of more spindles?" has been answered with "The discs we've chosen are based on the lowest cost per GB and the best performance." I suggested to at least increase the local disc size to have TempDb stored locally (on a separate logical drive), but so far without any luck...Am I worried for no reason?Edit: I'm planning to "attack" the SAN using SQLIO concurrently from several servers prior to going into production to verify the performance metrics they've told us.Wed, 07 Dec 2011 15:05:27 GMTLutzMhttp://www.sqlservercentral.com/Forums/Topic1203152-377-1.aspxGood Day All -We are looking to replace our current SQL Server box and are trying to determine the best subset of components for this box. We will be running SQL 2008 R2 on this box with a handful of production databases that total roughly 250GB. We at any given time will have in roughly 400 process instances running in the Activity Monitor.Any insight that anyone could provide would be very helpful in our decision making. Please let me know if you need more information.ThanksWed, 09 Nov 2011 12:24:43 GMTja_yeagerhttp://www.sqlservercentral.com/Forums/Topic1135861-377-1.aspxIs there a tool/application to calculate space requirement for a database in SQL Server? I know sp_spaceused can be used, but it calculates the actual space being taken up at that moment. What I was looking for is a tool that connects to the database and the user can interactively provide input like average number of rows and get the space requirement. This can be used for planning for future. The tool should be able to connect to a SQL Server instance/database, Get data size for one row for each table User should be able to put estimated row counts for all the tables The tool should be able to estimate the size requirementWith some more input like transactions per day etc, we can estimate the size of the log etc.You can see this here too:[url=http://stackoverflow.com/questions/6264827/is-there-any-standard-tool-to-estimate-disk-space-requirement-for-sql-server]http://stackoverflow.com/questions/6264827/is-there-any-standard-tool-to-estimate-disk-space-requirement-for-sql-server[/url]Mon, 04 Jul 2011 05:10:29 GMTKangkan Goswamihttp://www.sqlservercentral.com/Forums/Topic1121108-377-1.aspxI am researching this memory setting and trying to gather as much info as possible. I want to be sure that my sql min/max memory settings are adjusted appropriately for this change. Also, how will this play out with my large paging file that has a dedicated physical disk? My total used memory on the (SQL only) box is at 99% and SQL uses 92%.SQL BOX:Win Server 2008 2008 R2 EnterpriseMemory - 144GBPagefile - 300gbPagefile local disk - 418gbSQL Server 2008 Enterprise SQL min server memory 0 MBSQL max server memory 143872 MBUse AWE = no[u]Lock pages in memory[/u]This security setting determines which accounts can use a process to keep data in physical memory, which prevents the system from paging the data to virtual memory on disk. Exercising this privilege could significantly affect system performance by decreasing the amount of available random access memory (RAM).[url=http://support.microsoft.com/KB/918483/EN-US]http://support.microsoft.com/KB/918483/EN-US[/url]Want to be sure that I am getting the most out of my hardware - Would appreciate any advice...Tue, 07 Jun 2011 09:27:25 GMTSkyBoxhttp://www.sqlservercentral.com/Forums/Topic1093591-377-1.aspxCan anyone recommend a laptop to run sql server 2008 and all the BI tools? I've used Dell in the past happily but currently have an HP Pavilion which is running at a snail's pace and I hate the keyboard.Thu, 14 Apr 2011 08:50:59 GMTLouise-604184http://www.sqlservercentral.com/Forums/Topic1108761-377-1.aspxHello. We are getting a new server with the following specs:Dell PowerEdge R71012GB Memory 1333MHZ(2) Xeon E5503, 2.0GHZ, 4M Chache, 800MHZ, Max Mem ProcessorsWe need to run SQL 2000 and possibly SQL 2005. I am planning on turning this into a virtual server. I would like to get the most robust SQL environment and trying to decide on SQL Enterprise vs Standard. My question revolves around the processors.If I creat a Virtual server with: -64 Bit Server 2003 -6 GB Ram -2 ProcessorsWhat should I do with SQL? Should I get enterprise version so it can use more than 2gb of RAM? I noticed that enterprise is only licensed for 1 processor. Am I correct in assuming that SQL 2000 Standard will use both processors, but only 2gb of RAM. But, SQL Enterprise will use only one processor but more than 2gb of RAM? Do I have to decide on Processor power vs memory power?Any other words of wisdom related to my setup would be greatly appreciated.Fri, 13 May 2011 17:14:11 GMTbcmathttp://www.sqlservercentral.com/Forums/Topic1108736-377-1.aspxI would like to lab this out but i do not have the hardware available at the moment.but lets say hypothetically i had the following setup.1 SQL 2008R2 server with the following disk configuration.- OS and SQL binaries on servers internal Disks Raid-1- Raid Card 1 with external disks raid-0 for tempdb file1.mdf- Raid Card 2 with external disks raid-0 for tempdb file2.ndf- Other disks for Database and log files (not important for this question).the question is... let say the server is running fie and a disk fails that is storing the tempdb file2.ndfwhat happens?does SQL see that it can continue to run using the TempDB files that are still available?does SQL get weird then shortly recover on its own running off of one temp file?does SQL jut go downuntill the issue is fixed?what about if the file1.mdf is not available is that a perment failure if the ndf is still available?any insights on this would be great.Fri, 13 May 2011 15:14:47 GMTleroy-1092048http://www.sqlservercentral.com/Forums/Topic972163-377-1.aspxHello,So I have a SQL server that is experiencing some memory bottlenecks. O/S Memory Utilization is over 90%, Buffer cache hit ratios are tanking to below 50% and Procedure cache hit ratio is consistently at around 40%.After going through queries and optimizing them the best I could, I have decided to try the /3GB switch.My server is on Windows Server 2003 Enterprise and is a SQL 2005 Standard Edition. I have 8GB RAM in it right now and it is running 2 instances.I have a few questions about using the switch.1. How exactly is the virtual memory allocated to user/kernel mode operations? For example, I have 8GB and I want to use /3GB switch. The system would reserve 4GB blocks of virtual memory for each instance of SQL server. 3GB would be reserved for one SQL and 1GB would be reserved for kernel mode. In the other block, 3GB would be reserved for that SQL and 1GB would be for kernel mode. Does this mean that the kernel mode operations get 2GB of memory or 1GB?2. How much memory is sufficient to run kernel operations? Should I use the /USERVA switch to allocate a bit more to kernel mode?3. If I wanted to add a third instance with the switch, how much more RAM would I need?Assuming the case is true where each user program will reserve 1GB for the kernel mode operations could I for instance have 10GB of RAM with two of the SQL servers set to use 3GB max and the third to use 2GB? This would leave 2GB for kernel mode.ThanksEDIT: I also would like to know if SQL Server has the /LARGEADDRESSAWARE linker enabled by default or if I have to set that manually.Thu, 19 Aug 2010 14:13:22 GMTntran777http://www.sqlservercentral.com/Forums/Topic973003-377-1.aspxHi,I am tuning a disk sub system for our OLTP datebase server thesedays. This is DAS disks. I now came into a question.The server has 8 148GB SAS drives. To have fault tolerance, we want tohave RAID mirroring, so RAID 1 or 10. That leaves us half of thewhole capacity, which is 4 * 148GB. As this box is only used fordatabase, we want to have partitions for OS, data file, log file andtemp db. Now I there is a dilemma:I know for best practise, temp db should on its own physical disk. If I set up four partition(logical drives), one for OS, one for datafiles, one for logs and one for temp db, then I can only have RAID 1on each of the disk array. But the advantage of this is temp db isonly its own disk.Or the second choice I can have is to partition the disks into threearray (logical drives), namely OS, data files and logs. In this case Ican have RAID 1+0 (striping and mirroring) on data file partition andRAID 1 on OS  and log.Could you give me some idea, which one is better? Anydisscussions are welcome. Thank you very much.Sat, 21 Aug 2010 15:07:04 GMTBlue Mancunianhttp://www.sqlservercentral.com/Forums/Topic1047539-377-1.aspxHey folks, a quickie question about the drive space assignment mechanics.We'll simplify things. Working on a single drive in a LUN, said drive is a 1Tera drive. I need 300 gigs, so I get a nice simple 300 gig LUN assignment off the single spindle, leaving 700 of it free.I will eventually need all the space, but for now I drop 50 gigs into my LUN. The 700 gigs eventually gets assigned to something else, and it sporadically writes to the drive.As I add files (or grow my .mdf), I use up more and more of my 300 gigs.My 300 gigs looks contiguous from the perspective of the OS which can see into the LUN... however, is the physical spindle still contiguous, or does it interleave with the other 700 gigs assigned to a different LUN?Thu, 13 Jan 2011 13:00:17 GMTEvil Kraig Fhttp://www.sqlservercentral.com/Forums/Topic1037393-377-1.aspxHello experts,we are looking for a server solution for the next three years - at least. We collected 6TB in the last 3 years. We are using two direct attached disk boxes with 25 disks each connected to a dual socket quad core server. Currently RAID 5 due to space requirements. Storage and performance needs are growing with an estimate target capacity of 50TB ... 100 TB in the next 3 years. Any query should take same or less time than now, using the complete history of collected data.We are skeptic about a SAN solution using a NetApp filer since we dont believe that a SAN solution is able to deliver the same performance as direct attached storage. We also fear that a netapp SAN system could be 5 to 10 times more expensive. Whats yor opinion and experience with SAN performance? How good do SANs scale wrt. performance and capacity? SAN backup is extremely expensive due to strange political internal contracts (imho). Expert opinions wanted!Mon, 20 Dec 2010 14:03:49 GMTmlbauerhttp://www.sqlservercentral.com/Forums/Topic1052953-377-1.aspxHello,i want to make a small poll about your experiences with SAN systems.Here are a few topics of interest:- manufacturer- capacity- practical performance- used software: OS, DB- hardware and number of connected servers- daily problems- scaling experience- ease of administration- cost compared to direct attached storage- experiences with SQL server clusters- backupThank you for participating!Tue, 25 Jan 2011 02:48:49 GMTmlbauerhttp://www.sqlservercentral.com/Forums/Topic874827-377-1.aspxDoes anyone have any stats showing Equallogic performance. They claim it does as good or better than any other SAN. What I can't figure out is how it has better performance for SQL if(big if) the databases are set up correctly.Equallogic takes 14 drives and set them up in one RAID group.For SQL performance you really need the log on a different RAID group than the database and ideally tempdb to be on at least one other RAID group if not more.I believe some people would see performance gains, because the databases were never set up properly, so the optimization that Equallogic does would be much better than what they had before.Does anyone have any experience with this?Mon, 01 Mar 2010 19:53:47 GMTMark R1http://www.sqlservercentral.com/Forums/Topic1040220-377-1.aspxHello,We have a typical RAID 10 (2+2) configuration with 4 HDD on one of our production database server where I observed in perfmon for the drive which is configured with RAID 10 configuration.Avg. Disk Read Queue Length - 12.51Avg. Disk Write Queue Length - 0.06Somewhere I read , I should divide these figures with the no of spindles. Now the question is , should I divide these values by 4 or 2 and what should I interprete from these values ?Thanks,HemantWed, 29 Dec 2010 06:02:49 GMThemant.chaurasia-942319http://www.sqlservercentral.com/Forums/Topic983366-377-1.aspxDoes the SQL engine work as compiler or interpreter or hybrid ?Thu, 09 Sep 2010 12:20:43 GMTmoh_samir_86http://www.sqlservercentral.com/Forums/Topic1006979-377-1.aspxOk, we all know that we can change SQL Server settings to improve performance and database tuning to improve specific databases..... Now what do we do with the servers and the OS themselves? Dedicated SQL Server (new box)Windows 2008 64bit Enterprise EditionSQL Server 2008 64bit Enterprise EditionSAN attached drives128GB memorygoing to be be around 600-700GBs of DBs transactional. My list so far is (all require researching and testing): pagefile - 1.5x physical memory still the base standard? obviously I dont want SQL paging power management background services hyperthreading limiting the max memory of SQL serverWhat else should be added to my list?Tue, 19 Oct 2010 08:02:53 GMTgrahamchttp://www.sqlservercentral.com/Forums/Topic1021159-377-1.aspxI posted this up in the general SQL Server 2008, but thought I might have better luck here:Hi All,We are looking to migrate our to a new server with SQL 2008. Currently, our biggest DB is about 50 GB in size and supports with heavy a decent amount of reads/writes between automated processes and about 20ish users on it. The server is without a doubt a mission critical sever as it houses our CRM db on it as well as a few other smaller databases.The current server has 3.93 GB of ram and is running a 4.5 GB page file. We are a small business, so budget is far from unlimited, but what would you recommend in terms of hardware?Thanks for the help.Mon, 15 Nov 2010 17:10:06 GMTAdam Hart-452651http://www.sqlservercentral.com/Forums/Topic1021195-377-1.aspxi have to figure out the best configuration for 2 Servers - identical to each other1. SQL Server 20082. Cold Fusion for Intranet website3. 3rd Party BI (not ssrs or ssas) tool with integrated web#3 suggests that it should have it's own server, and if you put #3 on the same server as #2, you have a conflict on port 80, so you would need to do a bit of configuration to get them to work togetherso given #1 and #3 have best practices to run on their own server, where should I put #2 the cold fusion??? does anyone know if cold fusion is a memory hog or it's affect of putting cold fusion on the same box as sql server.thanks in advanceMon, 15 Nov 2010 21:23:32 GMTnawilliahttp://www.sqlservercentral.com/Forums/Topic979200-377-1.aspxWe are lookng to consolidate a few existing apps so they would hit the same common single database. The size of the database initially would be about 125 GB. Need to set up a brand new production environment. Expecting 1500 to 2000 concurrent users hitting the box, although the total number of possible users will be much higher. For the backend, looking to buy three high end servers. One server would be the OLTP box, the second needs to be a Reporting server that will be read-only and will be replicated from the OLTP within seconds. The third (DR) would be placed off-site and would serve as the DR/backup server in case the primary site goes down completely, and so will also need data replicated from the primary site with minimum delay.We would need some kind of SAN to SAN replication which is very fast.If cost was not a very big consideration, and obviously we don't want to buy a supercomputer or spend more than what we have to; can you please give some recommendations for what has worked for you if you have had a configuration similar to the one we are looking to have.We would prefer HP products, but would be nice to know your experiences with other servers as well. If you have had EMC or other third party vendors for setting up replication please comment on what you have used and what is good and bad about it. Thanks,Bvip.Wed, 01 Sep 2010 16:08:33 GMTBharatviphttp://www.sqlservercentral.com/Forums/Topic975239-377-1.aspxHello. My team is tasked to go through all the host servers we have SQL instances on to verify that write caching is disabled on all disks hosting active DB files. We've been researching WMI queries to see if we can identify the property that indicates the current write caching status but we haven't had any luck. Has anyone come across a utility or script that will will capture physical disk info across multiple servers?Thank you,MattWed, 25 Aug 2010 15:26:00 GMTmaz9022http://www.sqlservercentral.com/Forums/Topic973791-377-1.aspxIn the process of commissioning a new server and thought I'd see what everyone thought of this configuration.The server will be used to host a 1 reasonably large OLTP database at around 40GB and several smaller ones <2gb. The main DB is fairly heavy on both reads and writes.My questions are1. am I losing anything by going RAID1+0 on the data and log vol for the main database?2. would I be better off using with just either the data of log from the main db on the RAID1+0, and placing the tlog and random back ups on the other RAID vol?3. Ditch RAID1+0 entirely and have a separate vols for the radom db's log and data files.The random db's are largely web apps with little writes here and there, fair bit of reading, but nothing huge.2x Disk RAID1 LogicalDisk1 – Local Server Disk Sysvol C: - Partition1 Sysvol2 D: - Partition2 4x Disk RAID1+0 (10) LogicalDisk2 – Array Disk Data File E:4x Disk RAID1+0 (10) LogicalDisk3 – Array Disk Log File F:2x Disk RAID1 LogicalDisk4 – Array Disk TempDB G:2x Disk RAID1 LogicalDisk5 – Array Disk Transaction Log backups/ Random DBs(mdf/ldf) H:Mon, 23 Aug 2010 18:09:55 GMTMat-381415http://www.sqlservercentral.com/Forums/Topic941945-377-1.aspxI have always heard that the only thing that should be running on a SQL Server database server (or any make perhaps?) is SQL Server. No Microsoft Office, No Firewalls, No Antivirus. no other software period.We have a Windows 2003 Enterprise server that we run our SQL Server 2008 Enterprise db on. It sits on a RAID 5 disk subsystem. It powers our intranet and is not accessible from the outside world as it is in a heavily fortified DMZ. There are no files that originate from a client machine being written to the server with the exception of small photographs of adoptable children that are moved from one or two desktop machines within our organization that are also running mcafee. All other files are written to a directory on the server from stored procedures. We push these to other machines within our intranet periodically. About a year ago our IT department installed the McAfee VirusScan Enterprise 8.7.0i client on the database server. It performs an on access scan whenever a file is read or written to. Performance slowed to a crawl and eventually it was discovered that we had to exclude certain DB files from being scanned. Once that was done performance improved. We then upgraded to 8.7.0 and performance once again improved.I would like to hear from other professionals what their preferred method is for virus scanning a db server such as ours, one that never gets input from outside the intranet and rarely from within. Is it really necessary for this to be on our database server? Couldn't the one directory that has files orginating on a desktop be scanned remotely rather than having the app installed? Is it better to have this sitting on the DB Server itself? Does it provide a benefit that a remotely hosted scan can't provide? Can you think of any issues one way or the other?Thanks for your input!Wed, 23 Jun 2010 10:59:44 GMTMothInTheMachine