The IO load is not perfectly balanced, but is pretty well distributed: when WAFL writes to disk, it will build the data map in memory and dump it to disk in large chunks - part of that process for multiple Raid Groups is to dump chunks of data to each in a round robin fashion (chunk size used to be about 64MB per RG, unsure what it is now). This does a pretty good job of making sure the write _and_ read workload is distributed unless you are hitting a very small chunk of data over and over again.
-----Original Message----- From: owner-toasters@mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Page, Jeremy Sent: Monday, April 07, 2008 8:18 AM To: toasters@mathworks.com Subject: balancing load across Aggregates
If I have 4 RAID groups in an aggregate does the IO normally get balanced across them? I trying to figure out what the best way to present ~15 TB of NFS space to my VMware environment is. I have 5 shelves of 500 gig drives so my plan was to create 4 RAID DP stripes down the shelves (so 4 RAID groups of 10 drives each, counting parity) and then stick all 4 raid groups into an aggregate. Anyone seen issues with this? It's on a 3070, I'll be connecting via 10 gigE and NFS will carry most of the traffic.
I apologize if this is a repost, I thought I sent this last week but don't see it on the list.
This message (including any attachments) contains confidential and/or proprietary information intended only for the addressee. Any unauthorized disclosure, copying, distribution or reliance on the contents of this information is strictly prohibited and may constitute a violation of law. If you are not the intended recipient, please notify the sender immediately by responding to this e-mail, and delete the message from your system. If you have any questions about this e-mail please notify the sender immediately.