process_physical_memory_low is 1 in sys.dm_os_process_memory

Question

process_physical_memory_low is 1 in sys.dm_os_process_memory

Robin35

SSCertifiable

Points: 6471
More actions
May 2, 2017 at 9:04 am

#337696

Hello,
On our secondary node (Always On ), i'm seeing the following. Is this a concern ? I dont see lot of documentation on how to remediate this.
can anyone please shed some light on this ?
SELECT physical_memory_in_use_kb/1024 AS [SQL Server Memory Usage (MB)],
   large_page_allocations_kb, locked_page_allocations_kb, page_fault_count,
     memory_utilization_percentage, available_commit_limit_kb,
     process_physical_memory_low, process_virtual_memory_low
FROM sys.dm_os_process_memory WITH (NOLOCK) OPTION (RECOMPILE);
Output :
memory_utilization_percentage
99process_physical_memory_low
1

Viewing 5 posts - 1 through 4 (of 4 total)

You must be logged in to reply to this topic. Login to reply

SQLAssAS SSCertifiable Points: 7246 More actions · Answer 1

Hi

Is this constantly returning 1 or was this a one off? I would set up a job to capture this figure every 5 minutes or so to see if it is constant if you dont have a proper monitoring tool.
Is page life expectancy low? Do you have memory grants pending constantly?

Many many factors can lead to poor performance relating to memory, but first of all get a good picture of how memory is being used.

Perry Whittle SSC Guru Points: 233897 More actions · Answer 2

Robin35 - Tuesday, May 2, 2017 9:04 AM
Hello,
On our secondary node (Always On ), i'm seeing the following. Is this a concern ? I dont see lot of documentation on how to remediate this.
can anyone please shed some light on this ?
SELECT physical_memory_in_use_kb/1024 AS [SQL Server Memory Usage (MB)],
   large_page_allocations_kb, locked_page_allocations_kb, page_fault_count,
     memory_utilization_percentage, available_commit_limit_kb,
     process_physical_memory_low, process_virtual_memory_low
FROM sys.dm_os_process_memory WITH (NOLOCK) OPTION (RECOMPILE);
Output :
memory_utilization_percentage
99process_physical_memory_low
1

How physical RAM does the server have installed?
Whats the max server memory for this instance?
How many instances on the server?

-----------------------------------------------------------------------------------------------------------

"Ya can't make an omelette without breaking just a few eggs" 😉

Robin35 SSCertifiable Points: 6471 More actions · Answer 3

SQLAssAS - Monday, May 8, 2017 3:59 AM
Hi
Is this constantly returning 1 or was this a one off? I would set up a job to capture this figure every 5 minutes or so to see if it is constant if you dont have a proper monitoring tool.
Is page life expectancy low? Do you have memory grants pending constantly?
Many many factors can lead to poor performance relating to memory, but first of all get a good picture of how memory is being used.

Hi,

i ran it for 5 times in different intervals, it returned value 1. i will have to setup a job to capture the value. I checked the PLE value, it's around 5000, which is fine and no memory grants pending.

Robin35 SSCertifiable Points: 6471 More actions · Answer 4

Perry Whittle - Monday, May 8, 2017 11:34 AM
Robin35 - Tuesday, May 2, 2017 9:04 AM
Hello,
On our secondary node (Always On ), i'm seeing the following. Is this a concern ? I dont see lot of documentation on how to remediate this.
can anyone please shed some light on this ?
SELECT physical_memory_in_use_kb/1024 AS [SQL Server Memory Usage (MB)],
   large_page_allocations_kb, locked_page_allocations_kb, page_fault_count,
     memory_utilization_percentage, available_commit_limit_kb,
     process_physical_memory_low, process_virtual_memory_low
FROM sys.dm_os_process_memory WITH (NOLOCK) OPTION (RECOMPILE);
Output :
memory_utilization_percentage
99process_physical_memory_low
1
How physical RAM does the server have installed?
Whats the max server memory for this instance?
How many instances on the server?

total RAM 512GB , two sql instances, instance1 350GB and inst2 140GB...Instance1 is having this issue, instance2 is fine. i'm thinking to increase the memory for instance1. both instances are secondary replicas for Alwayson nodes. this is three node always on cluster, 2 primary nodes with each instance and 3rd node is for secondary replicas with 2 instances in single node.