Virtual Geek

Celerra Virtual Appliance HOWTO 301 - Replicating Between Two Celerra VSAs

Glad to see that people are having downloading and success with the Celerra VM on VMware Workstation based on this original post.  I wanted to provide a quick "HOWTO" to help customers Replicate between two VSAs, for many reasons - one of the most fun being VMware Site Recovery Manager.

Celerra Replicator (I hope the Celerra Product team doesn't read this) is a RIDICULOUSLY inexpensive (particularly when compared with the competition) but very sophisticated remote replication capability.   It can support cool stuff like:

• 1:N and also N:1 replication fan-in/fan-out replication relationships (like with any technology there are operating envelopes and parameters)
• It can support cascading topologies (i.e. site one replicates to site 2 with a given frequency, and then site 2 replicates the data from site 1 to site 3 at a different frequency
• It has pretty sophisticated QoS mechanisms (i.e. you can setup different bandwidth use for different parts of the day
• You can replicate all sorts of configs - CIFS/NFS/iSCSI, and fully supports thin provisioning at the source or the target
• And of course - most interestingly of all - it's fully integrated with VMware Site Recovery Manager (so you can use this to build your own SRM playground.

Of course - SRM doesn't support NFS yet, so I focused this HOWTO on iSCSI - but it's very easy to see how you would do CIFS/NFS (just hit the other radio button in step 3)

Before you do this - make sure you have a working VSA - just follow the Celerra VSA HOWTO 101 here...

I've updated the OVF - tightened it up a bit.   You can get the new one here

Ok - have you got it?   Have you followed the 101?    Want to add more storage - just follow the Celerra VSA HOWTO 201 here.  

Ok... then read on...

Quickly then here - in this HOWTO:

1. Configuring NTP
2. Correcting the Replication Database (post Clone or deploy from OVF the Celerra VM's serial number is changed) - BIG thanks to Himabindu Tummala and Santosh PasulaReddy in Celerra Engineering for helping me figure this step out
3. Configuring an iSCSI LUN that will be used as a replication target
4. Configuring Celerra Replicator

One of our elite-delta squad VMware Specialists is putting the finishing touches on an "all in" doc and pair of Celerra VSAs to make setting up SRM with two Celerra VSAs a breeze, but you have every ingredient you need now with these HOWTOs (and SRM itself is a walk in the park).   

One quick note - remember that if you are looking for support - post on http://forums.emc.com, in the Celerra, Celerra Simulator forum. 

Next up to plate is the HOWTO on Avamar Virtual Edition, and I'm also going to do a HOWTO post on how to use the super-cool Replication Manager with the VSA.   Rep Manager - aka "RM" can actually be used with all EMC's primary storage platforms, for all sorts of replication/test and dev use cases - for Exchange, SQL Server, Oracle, and also VMware VMs and VMFS datastores - NFS datastores being supported shortly.

Read on......

Some pretty fascinating statistics

First of all - happy Thanksgiving to my American colleagues!

Second, some crazy stats.   Read on if you're interested....

Version 4 of the PowerVDI tool

This tool continues to get updated (thank you Jeff Purcell who did SO much on this and the previous release, Dan Baskette and the Celerra team!).   You can get the latest version here:

http://powervdi.vcloud9.net/

This version is much faster - at least 2-3 times faster than the old one.   In terms of speed, I think it would be hard to get faster than this - in our "playing around", it's at least 10x times faster than the upcoming View Composer approach.

So - does that mean we think the array is the place to do this rather than the ESX server (aka VMware View Composer) long term or in all cases? NO. 

There are so many things that can be done re: image preparation and handling and management better at the VMware layer than the storage array layer - which are both as important as speed.   All the approaches have similar cost savings.

We're embracing the upcoming stuff (stay tuned - another blog post coming on this in a couple weeks).   Over time, we're going to work to merge them as much as possible (i.e. use the VMware View tools to manage, but where speed of mass distribution is the key - you can choose to do it at the array layer rather than at the VMware layer (but it's managed at the VMware layer).

Celerra Virtual Appliance HOWTO 201 - Adding More Physical Storage

Glad to see that people are having downloading and success with the Celerra VM on VMware Workstation based on this original post.  I wanted to provide a quick "HOWTO" to help customers add more physical storage, so they have more to play with!

Before you do this - make sure you have a working VSA - just follow the Celerra VSA HOWTO 101 here...

I've updated the OVF - tightened it up a bit.   You can get the new one here

Ok - have you got it?   Have you followed the 101?    Ok... then read on...

Quickly then here - in this HOWTO:

1. Adding physical storage to the VM
2. Configuring the additional storage to the Celerra VSA (and how to remove storage)

Next up to plate are the 301 (setting up Celerra Replicator and SRM), and the Avamar Virtual Edition.

Read on......

Wow - all I can say is THANK YOU.

http://communities.vmware.com/blogs/ManualAutomation/2008/10/07/site-recovery-manager-is-a-hit

There's no better source of input than a customer's own voice.   You'll notice that we're not perfect - so you know he's sincere :-)

We are seeing great success with SRM and EMC - customers really like it.   There are gaps - to be sure - and I'm in Palo Alto right now and we're working furiously to continue to innnovate together (and yes, we're working on those things you've pointed out and more)

Virtual_JTW - thank you for being an EMC and VMware customer!

What's the relationship between VMware vCloud and EMC Atmos?

First, it's useful to understand where they are not related.   Atmos is NOT a vCloud. 

vCloud is an initiative.   vCloud defines a set of standards for compute objects (VMs using the OVF standard), and APIs (using REST/SOAP) to manage VMs running in a cloud compute model,  APIs to communicate SLA requirements to the underlying layers (vApp - which applies similar logic to internal clouds), structures to get VMs in and out of the cloud, and the security requirements to make that model work.  A lot of really, really cool stuff.   It's primary benefit is that it makes the same objects that are your primary compute management model (Virtual Machines) transparently and natively run in a hosted model - which bypasses the barriers that stop certain workloads from being SaaS-able (or at least really hard to make work that way).   It also makes the distributed, dynamic datacenter expand past the walls of the datacenter itself (for example - for spot or seasonal workloads).

Atmos is a product.   Atmos' design goal is to be the best design for storing information in the cloud use case - cloud-optimized storage (COS).   The use case contains by definition several design implications.   It implies:

• many-petabyte (even exabyte) scaling and a cost model to match (which in turn demanded a scale-out storage model using low-cost storage).
• object counts in the billions  
• multi-tenancy
• global namespace
• a wide variety of access methods (REST/SOAP and more traditional CIFS/NFS/IFS)
• most importantly - an object-based management and information distribution model.

We've been working hard on this for a while - and in fact have been shipping this since June quietly.   It's definitely something new so we took our traditionally conservative model of product deployment (internally we do a "phased product release" process for all major products) and applied it to our innovative new idea, which is why you might have heard of "Maui" before.

Could Atmos be used for internal vClouds (i.e. customers deploying their own VMware Infrastructure with a vCloud inspired access and management model)?  Sure.   But, at the same time, customers however need to understand it's design goal is for internet scale, and internet-connected access to information.   

So - where is the intersection?   The intersection is a shared long-term view and goals.   Both EMC and VMware view this cloud model - both for computing and for storing information as a new emerging force in IT.   VMware is working to make a standard compute container model that makes clouds applicable to any workload - unchanged.    EMC is trying to make storing information for cloud use cases simpler and lower the bar for new entrants into this market.    We think that Atmos is the ideal platform for storing the information accessed  by cloud compute layers - including, we're confident, partners who plan on building those based on the vCloud model.   But, of course, when you have two large vendors like EMC and VMware working together, you can expect to hear more from us on this topic soon enough.

Read on if you want to understand Atmos better....

VMworld 2008 - EMC Wrap-up - Part II

Ok, here's the final wrap-up (a month late :-)  - I FINALLY got the video of our keynote - it's below in it's entirety.   For the ultimate "being there" experience (HA!), watch the video with the slides at the same time (the camera work was wierd, no slides behind us, they were off to the sides).   But, kinda neat regardless.   I liked the "how many folks here..." bits at the beginning, and you can see some of the physical demonstrations we did with the CX4 and the iomega StorCenter ix2 (our $329 NFS server/iSCSI target)

UPDATE (Feb 2, 2009) - The SS4200E is a fine choice, and so is the ix2.   If you can wait a little bit... I would hold on.   iSCSI work is in the lifeline codebase, but still pushing to make it available in all the platforms - which have varying CPUs, and varying iSCSI performance.   for now, consider the shipping SS4200E and ix2 as NAS-only until I give a firm update.   There are other reasons also to wait for about 1 month....

If you're wondering what that big tattoo on my hand is... no, I'm not that cool :-)   It's a big "I love VMware" temporary tattoo.

I also cleaned up the PDF (export of the straight PPT was a bit mangled due to powerpoint builds, and the one on the VMworld site is pretty weak quality), consider this the one to use if you want to look at it.   It's useful to look at this as you watch the video (see note above).

Read on for details....

VMTN Podcast - I'm on the roundtable today...

Folks - if you don't listen to these, I would highly recommend it....   They cover broad topics, and tend to be not the usual marketeering schlock, but rather a frank dialog between technical folks who work with VMware every day.

Every Wednesday, at Noon PST....

http://blogs.vmware.com/vmtn/podcasts/

Connect info is here. 

Today, I'm on the roundtable.   While they are unscripted and free flowing (who KNOWS where that goes - but that's part of the fun)

Topic that are on my mind below.   If these are interesting, dial in.

• What VMware, EMC and Cisco are doing together trying to design the NGDC
• What's coming in vStorage
• What's brewing in the reference architectures labs for Tier 1 applications like Exchange, SQL Server, Sharepoint
• What we're working on next (i.e. taking SRM to the next level) around Disaster Recovery for VMware
• Storage Best Practices for VMware - what we're seeing work well.
• And most importantly... anything you want to ask me!!!

A couple "By the way" comments:

1. A really good post on single initiator zoning on Yellow-Bricks... This has been in all the EMC VMware Academy training sessions we've been doing for more than a year.  While we run our "best of the best" through VCP, everyone who is a pre-sales technical resource needs to do the VTSP and these "Academies" which compress the "Install and Configure" course into 2 days (with EMC-specific content on storage, backup and DR you don't get in the VCP training).  Glad to see VMware saying the same thing.   Soft vs hard zoning is subjective (i.e. up to your preferences, and debatable), but single initiator zoning is not.   Do yourself a favor and don't create a single zone for all your ESX servers and VCB proxies.
2. I've also been up to my eyeballs over the last few weeks (last week alone I worked with customers who in aggregate have more than 20K VMs deployed - wow), but have some free time today, so am hoping to push out a couple of blog updates (including some long awaited Celerra Sim 201 stuff) I've been working on for a while.
3. Customers running VMware on NFS (on NetApp or Celerra) - make sure you read Scott's post on NFS.lock.disable (hint, leave it at the default of 0) here.  Note that this isn't a NetApp technical issue per se, except for in the sense that they published their best practices with this incorrectly in the past).   The same thing would occur on any NFS server (including the Celerra).   Moral of the story to me?  
• Locking isn't inherently a bad thing (it's often incorrectly bandied about as a core "VMFS issue" when people are FUD'ing VMFS scaling incorrectly).   Locking (at the SCSI on block devices and file level for VMFS and NFS) is an intrinsic and important property of ESX - and critical for ensuring no split-brain condition occurs.
• It's very important for all of us to prioritize  best practices around production use cases, not anciliary use cases.   This may make me a bit of a fuddy-duddy, but I think it's the "right thing to do".   While the VMware issue that caused ESX snapshots to be slow on NFS datastores was VMware's - not NetApp's (fix in ESX350-200808401-BG),  I personally think it's a mistake to recommend a workaround that compromises availability (since on NFS **and** on VMFS - where the file locks are maintained in the VMFS metadata area - the file locks are an important governator on VM's being booted on multiple ESX servers).   Again, people in glass houses shouldn't cast stones (i.e. we've made that mistake too in the past!) - but one of those things to consider as your vendors all bend over backwards to compete.    
4. Did you see EMC and Cisco buy more VMware shares?   For big companies - the amounts are small - but the message is clear - Cisco and EMC are doubling down here. 
5. I'm pretty pumped with the Obama win last night.   Some of my best friends and colleagues, smart and whom I respect are Republicans.  It is perfectly possible to disagree without being disagreeable.   But that said, I do feel like it's a fresh start, and is a great testament to the United States' ability to remake itself and adapt.