Storagezilla

eLab TechBooks for everyone

Backup and Recovery in a SAN TechBook - Provides information on traditional backup and recovery architecture, SAN-based backup and recovery technologies, and disk and tape backup and recovery. Some case studies are included.

Building Secure SANs TechBook - Identifies and exemplifies some common SAN security attacks, presents some built-in and bolt-on mechanisms to enhance SAN security, and provides some insight on how to implement various product-specific security mechanisms.

Extended Distance Technologies TechBook - Provides a basic understanding of distance extension technologies and information to consider when working with extended distance. IP-based distance extension solutions are also included.

Fibre Channel over Ethernet (FCoE): Data Center Bridging (DCB) Concepts and Protocols TechBook - Provides a high-level overview of Fibre Channel over Ethernet (FCoE), the supported configurations and features, and examples for incorporating FCoE into an existing data center. Discussion of Ethernet basics is included. Setup examples for Nexus Series, Virtual PortChannel (vPC), and MP-8000B are provided. EMC RecoverPoint and Celerra MPFS solutions are discussed. Basic troubleshooting techniques are also presented.

Fibre Channel SAN Topologies TechBook - Provides a high-level overview of Fibre Channel SAN topologies, discusses simple and complex Fibre Channel SAN topologies, and provides case studies for Brocade Virtual Fabrics and EMC RecoverPoint. FICON connectivity is also discussed.

iSCSI SAN Topologies TechBook - Provides a high-level overview of iSCSI SAN topologies and includes basic information about TCP/IP technologies and iSCSI solutions.

Networked Storage Concepts and Protocols TechBook - Provides fundamental information about Fibre Channel. It presents fabric design considerations, explains how SAN technology works, and describes IP SAN concepts. Extended distance technologies and solutions are also discussed.

Storage Virtualization and Replication Technologies TechBook - Provides a high-level overview of storage virtualization and replication, including EMC’s Invista, RecoverPoint, and VPLEX. It also provides information for SAN connectivity for Invista and RecoverPoint. Basic information on interswitch link hops, along with examples showing hop counts in four different fabric configurations is included.

WAN Optimization Controller Technologies TechBook - Provides a high-level overview of the WAN Optimization Controller (WOC), including network and deployment topologies, storage and replication application, FCIP configurations, and WAN Optimization Controller appliances.

via brasstacksblog.typepad.com

The eLab folks are a large part of the EMC engine room, doing the testing and writing the documentation to stand behind it.

Erik Smith's new Blog over at Brass Tacks is all about Connectivity and just as EMC eLab's role is changing inside EMC I'm sure Erik's blog content will reflect those changes as they occur.

EMC joins the Storage Performance Council

Proving that even after more than a decade you shouldn't take anything for granted, on Thursday EMC joined the SPC as an Associate Member. What happens next? Don't concern yourself with that. It's early days, so early the ink is still drying on the paperwork but if there is something to share you'll hear about it. EMC will make sure of that.

VG8+VNX? About half a million IOPS

New Year, New Platform, New Benchmark.

I’m looking at the recently published SPECsfs results for the VNX where a five XBlade, eight Disk Processor Unit configuration posted a competitor crushing 497623 IOPS and I’m thinking this is only the first benchmark on that config.

Right now it’s on top of the pile, even above the posted VMAX numbers but VMAX and VNX are different platforms.

That being said even with these numbers I’m convinced we could get more out of the VNX config. We’ll have more benchmarks later on in the year and the SPECsfs CIFS numbers will be published in the near future.

NetApp Bad Science

Vaughn Stewart makes me laugh. Now, I don’t think he’s a bad guy but from time to time he is guilty of a lot of the Bad Science he’ll rail against should someone say ‘God bless you’ in his presence.

Lets turn to Google cache.

Unfortunately i did not attend the webcast, and don't have Mr. Gelsinger's actual comments. However, I trust the accuracy of Chris's reporting and will work to obtain a transcript.

 

This says ‘I was either too busy or too lazy to check any of the facts before I wrote and posted what follows.’

I have no problem with Chris Mellor and was just last week marvelling at his lustrous page boy haircut at Cisco Live but I do disagree with his opinions from time to time. Like this one here..

“If EMC has misread the situation and NetApp's lead lies in its current feature superiority, such as block-level primary data deduplication, as well as in the general smoothness and integrated management and feature sets across the FAS array range...”

That’s Chris’s opinion but there are opinions from NetApp resellers which differ from that.

Now, EMC, in my opinion, is doing a terrific job with Unisphere, the tool is solid, and it keeps getting better and better (which is in stark contrast with their storage OSes that are more fragmented than ever before). and NetApp, which is doing great with their OnTap unified architecture on the storage side, should follow Unisphere in their approach to management consoles where they can also integrate with SanScreen (acquired with Onaro in 2008) which is already largely used by enterprises all around the world.

So even people who sell against EMC every day and take a pop at the purpose built strategy say that Unisphere is the right approach and something NetApp should follow. That’s a management lead right there.

It’s my opinion that based on customer feedback it’s not just a lead but a clear lead. And it hasn’t even been available for a year yet.

But back to Vaughn, who looks like he finally started doing some research when the heat was being turned up on him in various places and retrofitted the following onto the existing  post.

UPDATE 2/11/11: If you read the transcript of the webcast, Mr. Geslinger comments that NetApp has a lead which EMC will try to close and suggests that compression and deduplication are merely alternative means to achieve storage efficiencies. It appears he is to attempting to obscure the differences between these capabilities while also sending a message that is in stark contrast to his previous statement regarding how EMC engineering is working towards providing data deduplication capabilities.

And we’re back to more more spinsanity.

I know because I watched the webcast and you can watch the replay here. Pat stated primary block level dedupe was a feature which EMC will ship in the second half of this year. Before and after that he was hammering on all EMC competitors while promoting EMC leadership in the field of Virtual Storage. 

And you should watch the webcast, it’s quite interesting and as per usual Pat starts talking about futures in a way product managers wish he wouldn’t.

What we’re seeing in Vaughn’s post is another example of a bizarre corporate trait I term NetApp Bad Science. NetApp Bad Science is where if a bunch of NetApp employees believe something to be true amongst themselves then it must be true, regardless of any facts involved.

This year alone NetApp publically stated in various forums that  they were the leader in Flash sales as they shipped 3PBs of Flash. They had no other data points besides the self belief that this must be the case, but they said it anyway. Repeatedly. EMC then released it’s numbers where it showed it had shipped 10PBs of Flash and the first reaction?  ‘Well if that’s true..’

The idea that NetApp’s claims of leadership is verified by 39 minutes of video footage where Pat Gelsinger says otherwise is another example of NetApp Bad Science at work.

My friends, the world is not flat and vaccinating your children isn’t lobotomising them.

Avoid all forms of Bad Science even NetApp Bad Science.

Two divisions. Two videos. One EMC.

You can have 35 Trillion Gigabytes…

…and more than what’s in the box.

All from EMC.