Hi,
we have a problem with RAM upgrading. we have 16GB on server and want to upgrade to 32GB, but if you replace 16 to 32, so the server is very very very slow.
Can you help? where can be a problem?
THX JfK
what if you unplug the ram from the physical host? still that slow?
how you mean unplug?
16GB was unplugged from server and 32GB plugged.
I had an issue some time ago with a server, where the firmware/BIOS had issues with newly added RAM. The solution was to upgrade the firmware to the latest available. It may also be an issue with the correct placement of the RAM modules. Some system have specific placement guidelines.
Does the new RAM have the same specifications regarding rank (2x, 4x, ...) and speed (800, 1066, 1333, ...)?
André
Hi
it is a server DELL PowerEdge R510 whit 2CPUs socket but we have only 1CPU in socket 1. it has 4 memory slots for each CPU. we have 4x4GB RAM (slots A1,A2,A3,A4) and we must upgrade to 4x8GB (absolutely the same type of memory, only larger). When I unplug the memory from sockets A1-A4 and replace it by the larger, BIOS sees 32GB. ESXi also sees 32GB, but VMs starts very very very slow.
new knowledge: when I run the VM in safe mode, then stops at loading Classpnp.sys and it is realy slow.
This is a question for Dell support. Run the Dell diagnostic CD and perhaps memtest. Dell will need you to run the diagnostic CD anyway. They will have a better understanding of what the issues might be.
I mean, its not hardware problem, BIOS see all 32 GB. can not be anything wrong set in ESXi? with technicians from Dell I've already talked about it, they bequeathed to me to VMware support. Its critical problem for our system.
Is the system slowing down the memory clock speed when that much is added? I have seen systems that go from a clock rate of 1333 to 800 to accomidate the additional RAM.
As an example that would significantly reduce transfer rate:
Friendly Name | Industry Name | Peak Transfer Rate | Data Transfers/Second (in Millions) |
DDR3-800 | PC3-6400 | 6400 MB/s | 800 |
DDR3-1066 | PC3-8500 | 8533 MB/s | 1066 |
DDR3-1333 | PC3-10600 | 10667 MB/s | 1333 |
If this is a critical problem an Dell can't help then I would make a call to VMware support. Just because the BIOS sees all the RAM and ESXi sees the RAM does not mean that this isn't a hardware issue. There are memory speed test utilities.
Our RAM = SAMSUNG M393B1K70CHD-CH9...speed is same as old RAM.
It isn't about the speed of the RAM it is the speed of the BUS. There may be a BUS speed change depending on the RAM configuration. It isn't normally a huge issue but it is common with most servers.
Open a suport call with VMware.
Worth a look...
http://www.techrepublic.com/blog/datacenter/review-dell-poweredge-r510-server/2553
Check out the 'What's Wrong' section. Especially the part about the chassis choice effecting what size memory modules the system can handle (up to 4GB modules or up to 16GB modules). This is just a review so, take it with a grain of salt but I would definitely go up the Dell support laddar to be sure.
It looks like Alceryes is correct. I went to the Dell site and priced out the 4 drive server. 1333MHz Dual Ranked RDIMMs display this error.
Errors: | |
R510 Chassis for Up to Four 3.5" Cabled Hard Drives does not support (2x8GB),(3x8GB),(4x8GB),(6x8GB),(8x8GB),(16x8GB),(3x16GB),(6x16GB) DIMMS structured with Memory option selected. |
1066MHz Quad Ranked RDIMMs seem to be allowed. That does suggest that this might be the issues with the RAM. If the RAM was purchased from Dell you may be able to get them replaced.
we have POWEREDGE R510 RACK CHASSIS FOR UP TO 8X...