Archives for March 2009

Cisco Nexus 1000v details..finally!

For eveyone that has been waiting for some juicy technical details on the Cisco Nexus 1000v, Cisco and VMware have broken the silence and published a document describing many of the new features. It consists mostly of a table comparing the various features that the built-in VMswitch has verus the Cisco Nexus 1000v. I’m expecting a lot more documentation as the product nears release, but this is a good start.

If you aren’t familar with what the Cisco Nexus 1000v is, it’s a software embodyment of a physical Cisco Nexus ethernet switch which is a whole sale replacement for the built-in VMware ESX virtual switch. Among many benefits, the network team now has full visbility right into the NICs for each VM and is no longer left in the dark. It only works with ESX “4.0”, which isn’t out yet, so it may be a few more months before it is publically available.

You can download the document here.

Exchange 2007 SP1 Rollup 7

Microsoft has released rollup 7 for Exchange 2007 SP1. To read all of the fixes in this rollup, see this link. As always, Microsoft recommends you update your systems to the latest RU, after thorough lab testing.

New ATI Windows 7 Drivers

According to this article at Extremtech, ATI will release updated video drivers for Vista and Windows 7 on a monthly basis. To download the new drivers go to this link. New drivers were released today, and I’m downloading them as I type this.

Major Intel CPU chip flaw exploit to be released

Check out this article for a very scary flaw in current Intel CPUs which could render an exploit virtually impossible to detect. The exploit was discovered by the same researchers that developed the Blue Pill hypervisor rootkit.

Hopefully Intel and motherboard manufacturers can release firmware updates to patch this bug.

Storage Questions Part 2

1. Snaphots
a. What is the maximum number of read-only and read-write snapshots per LUN?
b. Is a separate pool (reserve) of disk space required for snapshot use?
c. Are snapshots space efficient?
d. If a LUN has 10 read/write snapshots, and if the source LUN is modified, are 10 copies of that data made or do all snapshots reference the single new block?
e. Does the array support consistency groups? Maximum limits of the groups?
f. Must the snapshot of a volume use the same RAID level?
g. Can snapshots be scheduled, and at what intervals?
h. Can snapshots be set to automatically expire?

2. Thin provisioning
a. Does the array support thin provisioning?
b. Will the array reclaim previously allocated but now unused space? For example, a 1TB LUN had 700GB of physical data which was reduced to 100GB. What happens to that 600GB of allocated storage?

3. Performance/QoS
a. Does the array differentiate between various performance ‘zones’ of a disk, such as outer tracks and inner tracks?
b. What performance tiers can be created? (RAID1, 15K FC disks, outer tracks vs. RAID5, SATA, inner tracks, etc.)
c. Can LUNs be migrated non-disruptively between tiers?
d. Is there any automated data progression between performance and protection tiers?
e. If so, is this at the LUN level or more granular such as block level? For example, could a single LUN have some blocks on high performance disks while other less used blocks are on nearline disks?
f. Can LUNs be assigned performance parameters based on workload (Exchange, SQL, VMware, etc.)?
g. Are there any published SPC-1 benchmarks?