170 reputation
14
bio website compaspascal.blogspot.com
location Copenhagen, Denmark
age 43
visits member for 5 years, 1 month
seen Oct 25 '11 at 12:12

Mar
22
comment How to set dscp for incoming TCP connections to my server?
I was specifically asking os-agnostic because I didn't have a good idea about it - so this answer is great to get, because it helps me in my search. However, frankly, this was a Windows 2008 problem.
Mar
21
comment QoS for Firebird database connections - how?
I would prefer to see a gds32.dll api call or OS api call as a solution, that is why it is on stackoverflow.com, too - but I don't consider this likely. I think that a tcp tunnel solution or server configuration solution is most likely to be found first.
Mar
21
comment QoS for Firebird database connections - how?
Compressing is not the issue here.
Feb
23
comment rdate for Windows 2008
I want to point the server at pool.ntp.org or similar, but I want to be 100% sure that the system time does not change unless I ask it to. For our industrial purpose, this is very important.
Feb
22
comment rdate for Windows 2008
The Windows 2008 server is an industrial computer, that is on a firewalled network that does not have any PCs attached. It is not part of a domain or anything similar. Your link does not seem to solve the problem.
Dec
17
comment 2 virtual servers with DB on one SAN harddisk, one VM as spare for the other
How long time do you think the switchover would take?
Dec
6
comment Which database servers are not interrupted by server reboots? (Clusters?)
Many designes introduce latency for queries. For instance, the seek time of a disk is often a significant factor of latency, but network latency also adds to it. Bad SAN structures can cripple a database server, but often SAN only slows down things because the app is optimized towards being in charge of harddisk seeks.
Dec
6
comment Which database servers are not interrupted by server reboots? (Clusters?)
The human factor is usually the biggest downtime producer, either in operation or development of the app - I'm totally aware of this, but I still want info :-)
Dec
6
comment Which database servers are not interrupted by server reboots? (Clusters?)
@Seth: We're the app vendor. @warren: Please specify information about those requirements as an answer, since it is part of "How does it work?"
Dec
2
comment Which database servers are not interrupted by server reboots? (Clusters?)
Latency = time it takes to make one query.
Dec
2
comment Which database servers are not interrupted by server reboots? (Clusters?)
What are the latency implications of having an Oracle cluster, compared to an Oracle with a local disk?
Dec
2
comment Which database servers are not interrupted by server reboots? (Clusters?)
* Failover is not an option - it must be continuous operation. * VMware FT does not seem to support that the database server continues its operations while the virtualized (guest) OS is rebooting, am I correct on this? * Can you describe the MSSQL solution in details? I assume that the shared disc can be supplied by a group of servers that can also reboot without taking the disk offline. Please describe the performance implications of such a solution, compared to an MSSQL solution with a local disk.
Dec
2
comment Which database servers are not interrupted by server reboots? (Clusters?)
Note: I look for solutions that can keep the transaction open, even when the machine, on which the database server software is installed (Virtual or physical) is rebooting.
Dec
2
comment Which database servers are not interrupted by server reboots? (Clusters?)
How should VMware solve this? Normally I would put the OS on top of VMware, and the database server on top of the OS. So when I need to update the VM with patches, and reboot the VM, how do I avoid that the database transactions are interrupted?
Dec
2
comment Which database servers are not interrupted by server reboots? (Clusters?)
Note: I don't care about OS or vendor, all info is appreciated.
Nov
26
comment Copying 100 billion records per day to Oracle - how?
After a discussion with the customer, this will not be implemented.
Nov
26
comment Copying 100 billion records per day to Oracle - how?
Note, that the Oracle database will be read/only.
Nov
9
comment A few minutes of CPU time but on 1500 servers - where?
@womble: Solidity is good, but how much latency? Are we talking <1ms, 10ms, 100ms, 1000ms?
Nov
8
comment A few minutes of CPU time but on 1500 servers - where?
How do I get 1500 separate computers with rackspacecloud?
Nov
8
comment A few minutes of CPU time but on 1500 servers - where?
How much latency can be expected in EC2?