Reputation
6,725
Top tag
Next privilege 10,000 Rep.
Access moderator tools
Badges
19 56
Impact
~188k people reached

May
19
reviewed Close Outlook 2013 clean up mail on the server
May
19
reviewed Close Block brute force attacks on exim2
May
19
reviewed Approve Apache log to one file for multiple domains with logrotate
May
19
reviewed Approve Should a wildcard SSL certificate secure both the root domain as well as the sub-domains?
May
15
comment SAN performance too low?
If you come back with some details, I can refine my answer. There are a number of layers that could be choking you. I figured we'd start with the top.
May
14
answered SAN performance too low?
May
14
reviewed Approve google-cloud-sql tag wiki
May
14
reviewed Approve zywall tag wiki
May
14
reviewed Approve zywall tag wiki excerpt
May
12
answered When it makes sense not to setup quota on NetApp volumes?
Apr
29
awarded  Popular Question
Apr
15
comment during snapshot merging… are deleted transactions ignored?
It was explained to me the other way... basically reverting to snapshots involves no extra IO, but deleting a snapshot involves a bunch of writes as they merge the changes since the snapshot into the base VMDK.
Apr
7
revised vSphere Datastore for specific datatypes: NFS or iSCSI
removed an assumption I was making about how this would be like FC.
Apr
7
comment vSphere Datastore for specific datatypes: NFS or iSCSI
I thought it was like FC... I guess it's not.
Apr
7
answered vSphere Datastore for specific datatypes: NFS or iSCSI
Mar
12
comment Why is RAID 1+6 not a more common layout?
I've (briefly) configured something like a raid 6+1- a Netapp local syncmirror will create an identical copy of itself and multiplex reads across both plexes, while mirroring writes. It's mostly used for migrating Netapp V-Series to new backend LUNs, however if I wanted to double my reliability, I could do that with this.
Mar
9
comment Storage tiering and SLA models
It can be, especially with SSD making two workloads on the same hardware bottleneck at two completely different levels. You can get a blurry number by measuring latency under various types of loads. Alternatively, you can apply a number that's definitely way lower than the box is capable of spiking to but high enough for your clients. And when your phone rings, you'll be guaranteed it's because someone's hitting their QOS limit, not a box limit.
Mar
9
comment Storage tiering and SLA models
The benefit is that it prevents one app from making everyone else's latency higher. I would almost recommend capping all workload to maybe 75% what the box can do, just to ensure that spikes never hit anyone else.
Mar
7
comment Storage tiering and SLA models
QOS, at least on Netapp, works the other way. It limits performance, but doesn't guarantee it.
Mar
6
answered Storage tiering and SLA models