3 ms·
Devils advocate. Workload dependence... average user... not significant: A lot of users probably see "30%" and will assume the worst. There really isn't a lot
by thezilch 9y ago
Devils advocate.
Workload dependence... average user... not significant:
A lot of users probably see "30%" and will assume the worst. There really isn't a lot of evidence of how much an average
consumer will be affected. Pretty sure your average consumer isn't running pgbench all day. Hell, I'm willing to bargain 90% of servers (I really want to say more) are over-provisioned by 30% or more.
Mitigated over time:
Kernels are getting out safe patches ASAP, performance be damned. With more time to be careful, performance can be clawed back. And/or CPUs are upgrade and features such as PCID mitigate perf losses.
- teilo 9y agoThe problem is that "average users" pointedly does not include server and DB admins. I don't know about anyone else, but that's what I'm sweating over. We already know high I/O is what takes the biggest hit. I want to know what kind of a hit I can expect on our MSSQL VMs, and I am pretty sure the answer is not "minimal."
- deadshift 9y agoIf you're running SQL in a VM ... I'm pretty sure I know how to improve your performance by 30%.
- teilo 9y agoIf you mean run on bare metal, not happening, due to MSFT's outrageous core pricing model. They force you to license every core available to the OS, whether you need the capacity or not. Running in a VM is the only way of getting around that requirement. If you mean run something other than MSSQL — also not happening. Vendor lock in.
- jsteemann 9y agoGiven the potential slowdowns to be expected by the kernel fixes, I am not expecting a jolly January. I work for a database vendor.