Author |
Topic |
andyl9063
Starting Member
19 Posts |
Posted - 2012-11-28 : 15:35:27
|
So I have two servers live. They have same processor, everything is the same except different amount of ram.Both have access to sanServer A: windows: 2008 standard sql: 2008Server B:windows: 2008 R2 sql: SQL 2008 R2Is it possible to have these two servers load balancing the traffic for the sql database?SQL Cluster will also be setup to replicate the databases across the two nodes.Any specific or best practice I should test? Also, any issues with the current setup?If you have any other recommendations, I would love to hear them. My first time setting up nlb fresh.Ultimate goal is to have two server hosting sql database that will spare the workload around. |
|
chadmat
The Chadinator
1974 Posts |
Posted - 2012-11-28 : 15:47:42
|
Not really. You could do replication, but that is as close as you could get. In SQL 2012, you can segregate Read and Write workloads using AlwaysOn Availability Groups.-Chad |
|
|
andyl9063
Starting Member
19 Posts |
Posted - 2012-11-28 : 15:57:06
|
that is the thing, our vendor software has not been tested in sql 2012.any other recommendations? I am sure other people run into this issue where their sql server is being overwork. |
|
|
robvolk
Most Valuable Yak
15732 Posts |
Posted - 2012-11-28 : 16:09:27
|
Besides replication, you could do log shipping and restore in Standby mode. You'll have more latency than you would with replication though. Are you using 2 different versions of SQL Server? That's going to complicate things. |
|
|
andyl9063
Starting Member
19 Posts |
Posted - 2012-11-28 : 16:52:49
|
no i can have two of the same sql server version.so no one has actually spread the load between two sql server? how do you cope with a busy sql server? |
|
|
chadmat
The Chadinator
1974 Posts |
Posted - 2012-11-28 : 16:54:55
|
How busy is your server? You can shard data across servers.-Chad |
|
|
andyl9063
Starting Member
19 Posts |
Posted - 2012-11-28 : 17:02:51
|
this is my servers. Server cpu spikes to 50% and 18%.Specs:tem Value OS Name Microsoft® Windows Server® 2008 Standard Version 6.0.6002 Service Pack 2 Build 6002 Other OS Description Not Available OS Manufacturer Microsoft Corporation System Name System Manufacturer HP System Model ProLiant DL360 G7 System Type x64-based PC Processor Intel(R) Xeon(R) CPU E5640 @ 2.67GHz, 2666 Mhz, 4 Core(s), 8 Logical Processor(s) BIOS Version/Date HP P68, 9/30/2010 SMBIOS Version 2.6 Windows Directory C:\Windows System Directory C:\Windows\system32 Boot Device \Device\HarddiskVolume1 Locale United States Hardware Abstraction Layer Version = "6.0.6002.18005" User Name Time Zone Eastern Standard Time Installed Physical Memory (RAM) 36.0 GB Total Physical Memory 32.0 GB Available Physical Memory 8.96 GB Total Virtual Memory 49.0 GB Available Virtual Memory 25.3 GB Page File Space 17.1 GB Page File C:\pagefile.sys |
|
|
chadmat
The Chadinator
1974 Posts |
Posted - 2012-11-28 : 17:10:11
|
How many Batch Requests\sec? That looks like a pretty standard server these days, if not under powered. If you are running into performance problems, I would focus on :1. Tune the workload/process2. Increase hardwarebefore trying to design some complicated load balancing scenario.-Chad |
|
|
andyl9063
Starting Member
19 Posts |
Posted - 2012-11-29 : 09:42:52
|
quote: Originally posted by chadmat How many Batch Requests\sec? That looks like a pretty standard server these days, if not under powered. If you are running into performance problems, I would focus on :1. Tune the workload/process2. Increase hardwarebefore trying to design some complicated load balancing scenario.-Chad
how can you find batch\secs? 1. not sure, not really a dbadmin2. I can upgrade processor and ram.Thank you for your reply |
|
|
robvolk
Most Valuable Yak
15732 Posts |
Posted - 2012-11-29 : 10:07:01
|
Run this query:select getdate(), cntr_value from sys.dm_os_performance_counters where counter_name = 'Batch Requests/sec' It's a cumulative counter, so you'd have to collect these stats and then query to get the difference between samples. Alternately you can use Performance Monitor to measure it, it's under SQL Server:SQL Statistics. This will do the difference calculation for you. |
|
|
andyl9063
Starting Member
19 Posts |
Posted - 2012-11-29 : 10:28:46
|
object_name counter_name instance_name cntr_value cntr_type Batch Requests/sec 106414453 272696576does that seem high? |
|
|
robvolk
Most Valuable Yak
15732 Posts |
Posted - 2012-11-29 : 11:12:19
|
It's better to run the counter from Perfmon, you'll get an instantaneous reading. You can't use the query I posted one time, you have to accumulate readings and compare them. |
|
|
andyl9063
Starting Member
19 Posts |
Posted - 2012-11-29 : 11:18:57
|
yea i just research a lot of perfomance counters.i got a whole html file with a lot of countersany thing on there I should look for specifically? |
|
|
andyl9063
Starting Member
19 Posts |
Posted - 2012-11-29 : 12:41:55
|
batch requests/sec average is 48.251Max is 436.087 |
|
|
robvolk
Most Valuable Yak
15732 Posts |
Posted - 2012-11-29 : 13:06:13
|
That sounds pretty low even for the hardware you currently have. I would still recommend upgrading the hardware if you can (RAM is really cheap). From what you posted earlier it doesn't sound like either box is particularly stressed, 50% CPU spike is nothing, and even 50% CPU sustained is not particularly worrisome. Nothing that requires load balancing.What kind of disks are you using? What RAID level? Are data and log files on separate spindles? |
|
|
andyl9063
Starting Member
19 Posts |
Posted - 2012-11-29 : 13:45:42
|
the actual data of the sql server are on the same server, however, some of the other data that its accessing are on sans drive. The disks on the server are 15k rpm sata disks.I dont think its a disk issue, i ran a lot of the performance counter and look at all the measurements, didn't see any issues or anything out of the ordinary. |
|
|
robvolk
Most Valuable Yak
15732 Posts |
Posted - 2012-11-29 : 13:57:36
|
My next suggestion is to look at the waits SQL Server is encountering, Glenn Berry has a nice query for that: http://sqlserverperformance.wordpress.com/2010/04/18/a-dmv-a-day-%E2%80%93-day-19/He also has a series of articles on using DMVs for performance analysis: http://sqlserverperformance.wordpress.com/2010/05/02/recap-of-april-2010-dmv-a-day-series/ |
|
|
andyl9063
Starting Member
19 Posts |
Posted - 2012-11-29 : 14:10:14
|
i ran the query and im getting latch-ex at 38.47 for wait_time_s.What are latch_ex ? I cannot find any good information on it. |
|
|
robvolk
Most Valuable Yak
15732 Posts |
Posted - 2012-11-29 : 14:13:10
|
http://www.sqlskills.com/BLOGS/PAUL/post/Wait-statistics-or-please-tell-me-where-it-hurts.aspxMore: http://www.sqlskills.com/blogs/paul/category/Latches.aspx |
|
|
|