Originally Posted by hdawg
nice post easy-v!
DISKS DISKS DISKS = GOOD GOOD GOOD
9 out of 10 performance problems I've seen with Exchange is disk i/o related
For redundancy ... you could use software from a company like neverfail to "cluster" your BES servers ... you could also add a 2nd BES, share a configuration database and balance the users between the two. In the event that one fails, you drag and drop users from one BES to the other and they're back in action within minutes.
Monitoring ... it really depends on how granular you want to get. Do you want to just monitor service availability? Do you want to monitor health? Performance? Boxtone has some great products as does Zenprise. ... but this really all depends on your requirements.
We just completed an ExRap from MS and have some recommendations based on that, that we are looking to implement.
More info on my environment:
Exhange is on a SAN. The SAN has reached is I/O limits already and we are suffering performance issue. We'll be changing the SAN ina few months
and I hope to add another exchange server and balance user load.
We use HPOV to monitor our servers. I am aware there is a SPI for BES by a thrid party (cant remember name)
But i do like the pretty pictures that BoxTone have on their web site. Also, their reporting seems to be really good.
I like the redundancy idea. Its still has a SPOF but at least its the SQL guys problem and not mine
What concerns me the most is how the additional 700+ BBs will affect my exchange server. I've read a 1:4 to 1:8 BB vs Outlook load.
With those numbers, how are companies dealing with the huge jump in I/Os?
I appreacite the info guys.