Thread: Re: Hardware/OS recommendations for large databases ( 5TB)
> Merlin, > > > > just FYI: tyan makes a 8 socket motherboard (up to 16 cores!): > > > http://www.swt.com/vx50.html > > > > > > It can be loaded with up to 128 gb memory if all the sockets are > > > filled :). > > Another thought - I priced out a maxed out machine with 16 cores and > 128GB of RAM and 1.5TB of usable disk - $71,000. > > You could instead buy 8 machines that total 16 cores, 128GB RAM and 28TB > of disk for $48,000, and it would be 16 times faster in scan rate, which > is the most important factor for large databases. The size would be 16 > rack units instead of 5, and you'd have to add a GigE switch for $1500. It's hard to say what would be better. My gut says the 5u box would be a lot better at handling high cpu/high concurrency problems...like your typical business erp backend. This is pure speculation of course...I'll defer to the experts here. Merlin
Merlin,
On 11/15/05 7:20 AM, "Merlin Moncure" <merlin.moncure@rcsonline.com> wrote:
With Oracle RAC, which is optimized for OLTP and uses a shared memory caching model, maybe or maybe not. I’d put my money on the SMP in that case as you suggest, but what happens when the OS dies?
For data warehousing, OLAP and decision support applications, RAC and other shared memory/disk architectures don’t do you any good and the SMP machine is better by a bit.
However, if you have an MPP database, where disk and memory are not shared, then the SMP machine is tens or hundreds of times slower than the cluster of the same price.
- Luke
On 11/15/05 7:20 AM, "Merlin Moncure" <merlin.moncure@rcsonline.com> wrote:
It's hard to say what would be better. My gut says the 5u box would be
a lot better at handling high cpu/high concurrency problems...like your
typical business erp backend. This is pure speculation of course...I'll
defer to the experts here.
With Oracle RAC, which is optimized for OLTP and uses a shared memory caching model, maybe or maybe not. I’d put my money on the SMP in that case as you suggest, but what happens when the OS dies?
For data warehousing, OLAP and decision support applications, RAC and other shared memory/disk architectures don’t do you any good and the SMP machine is better by a bit.
However, if you have an MPP database, where disk and memory are not shared, then the SMP machine is tens or hundreds of times slower than the cluster of the same price.
- Luke
Merlin Moncure wrote: >>You could instead buy 8 machines that total 16 cores, 128GB RAM and > > It's hard to say what would be better. My gut says the 5u box would be > a lot better at handling high cpu/high concurrency problems...like your > typical business erp backend. This is pure speculation of course...I'll > defer to the experts here. In this specific case (data warehouse app), multiple machines is the better bet. Load data on 1 machine, copy to other servers and then use a middleman to spread out SQL statements to each machine. I was going to suggest pgpool as the middleman but I believe it's limited to 2 machines max at this time. I suppose you could daisy chain pgpools running on every machine.