« 2015 what to expect and This I Believe Part II. | Main | VMware PEX 2015 Bootcamp: Prepare to have faces melted 10,9,8 »

January 21, 2015


Feed You can follow this conversation by subscribing to the comment feed for this post.

Justin Warren

Nice to see the non-disruptive upgrade capability arrive!

I know it's a wikibon thing, but Solidfire is scale-out shared nothing, which is a bit different to v3 XtremIO which is scale-out full-mesh.

I assume XtremIO v4 will still be fully-meshed to guarantee the consistent low-latency? What's the ceiling on the number of nodes?

Vaughn Stewart

While I have no insight in to the situation that led to this post, I do believe competition can bring out the worst in some individuals.

You wrote, "distrust anyone who starts talking negative about the other guy" yet I just finished reading the EMC produced document titled, "Pure's Top Ten Lies: 10 things Pure will tell you that are not true"

See the irony? You are calling out the behavior of an individual who is following in the footsteps of the EMC corporate culture. Is it only OK if EMC slings the FUD?

How about publicly challenging the EMC marketing and competitive teams to act like an industry leader. Ask them to fulfill your vision and lead by example.

- cheers,


Well said Chad! That is also one of the best and most appropriate pictures I have seen to emphasize a point yet. I love that (most) EMCrs never go negative. There will always be bad choices made by some, but it should never be the rule.

Howard Wolowitz

Full disclosure: Current EMC-er, not with XtremIO, and trying to remain impartial here. Anonymous as I don't like personal search engine food.

Vaughn, I see where you're coming from, but in my eyes, correcting FUD isn't in the same boat as intentionally slinging it. That said, I've always seen Chad extolling the virtues of his manifesto to our pre-sales org, and discouraging any and all negative behavior toward colleagues, internal and external. People with negative attitudes don't tend to last long at EMC.

Perhaps we can do better with the marketing collateral? (I haven't read that doc) - No, not perhaps, we can *always* strive to do better in everything. I'm sure that Chad will follow up on your feedback to continue help shape the company for the better.

Mauro Ayala

Thanks a lot for the 4.0 update, it will be really useful.
I'm so happy to have our first XtremIO in my account in Chile, it was a competitive situation where we never go negative, we just talk about what XtremIO does and probe it in a productive system...... Go positive !!! and customer will trust you and hear you. Thanks Chad.

Chris M Evans

Vaughn, you are entirely correct. Customers need to make their own opinion from everything that is available, right or wrong. It's disingenuous to be claiming that everyone should play nice when EMC carries on producing material like the document you refer to.

Howard, with reference to "correcting FUD", EMC never corrected this page - http://www.xtremio.com/high-availability - which continues to claim XtremIO upgrades are non-disruptive. We know that's not true.

Chad, you may be trying to do the right thing, but people in glass houses shouldn't throw stones.


Chet Walters

Lots of good info here. I'm glad that what XtremIO will deliver in 4.x is what EMC has been advertising it as for the past 18 months.

As far as slinging poo, I'm with Vaughn on this one. The only manufacturers in my area sling negativity about others in my area is the EMC team.

Chad Sakac

@Vaughn - I hear you, but respectfully I don't see the irony. There would have been irony if I claimed EMC was innocent - I didn't.

Whilst the trigger for my post was an individual thing directed at EMC (that at least to my eyes was despicable as the doc you point directed at your employer from EMC) - my point wasn't **"EMC good, others bad!"**


I'll say it again, and direct it squarely at my employer: customers, partners - if an EMCer gives you something like the doc Vaughn points out, if I were you, I would employ the techniques I suggest in the post:

"- Distrust anyone (ANYONE) who starts talking negative about the other guy, particularly in describing what they do/don't do.
- FACT: It's hard enough to stay current on oneself, and focused on how to build solutions ourselves. The likelyhood that the person flinging poop is full of poop is high.
Ask your partners/vendors to describe what they think they do WELL, not what the other person does poorly.
- SHUN (absolutely SHUN) vendor created comparison tables (everyone has them, including EMC - I'm trying to make that a poorly-regarded practice). you'll recognize them by one vendor being all "check" and the other being all "exes". It's intellectual laziness. Sometimes the customers demand them. Please don't be an enabler :-)"

I got forwarded an example where the doc you point out went to a customer, and then the customer responded to your employer with a "give me a counter poop-list". I'm begging customers: "don't do that". Instead, respond negatively to the instigator.

I suggest that applies to the WHOLE vendor ecosystem. I don't think I'm the kettle calling the pot black.

I'm an individual who happens to be a leader at a company, and I'm suggesting that those practices are BAD regardless of source.

I'll say one more thing on this topic. I am constantly fighting furiously whenever I see any EMCer doing that stuff. If ever people have an example they can get me that I can action (I know it happens, but I can't action hearsay), let me know. Rest assured, I'm trying to action that doc, which I think is absolutely poop flinging. I think the arguments in the doc have merit (they reflect architectural differences), but the right way to do it is to emphasize how we do it, not how another vendor may make claims and go negative.

Vaughn - this isn't a company thing (at EMC or your employer) - it's a people thing. Some people play the game the right way. Some don't. We all need to do our part (me included - I'm not pointing at "others" - I'm pointing at all of us) to make this just a losing strategy.

What you are suggesting EMC does - I don't see generally, and when I DO - I absolutely try to correct.

@Chet - I'm glad you like XtremIO 4.x, and I'm sorry that anyone with an EMC badge every slings poop. If you can get me examples, I will always, ALWAYS work to stop it. It's not a company culture thing, it's a people culture thing.

@Justin, Mauro, vDirtybird, Howard - thanks, I'm glad you guys are excited, and please, please, PLEASE help me - keep being positive, and coach/encourage (and then escalate) when you see anyone wearing an EMC badge going negative. Teach them the better way. I'm glad that you all echo what I see, that as a whole we try to navigate the high ground.

@Chris - Chris, I don't know if there's anything I can do to change your mind. My sense is you have no love lost for EMC, and while I'm sad about that - I'm skeptical there's any argument I could make that would make you change your mind.

I've talked plenty about the XtremIO upgrade, and not sure what you're suggesting.

Should there be an asterisk on that marketing page that says "* while architected for NDU, and we've gone through 5 NDU upgrades, and one (2.4-3.0) was optional, but if you did it, it was DU in a way that we didn't anticipate when we created the original marketing of the product"?

How long should that be there in your opinion (we're going from 3.x to 4.x now for example).

Should this apply across the industry? Should all NetApp docs say "sorry, we are not NDU" because of the flexvol transition when ONTAP 7.0 was first released? Should ZFS implementations say "not NDU" because configuring ZIL in some cases? Should Tintri say "sorry we are not NDU" because of their path from non-HA to HA?

Those are all platforms designed for NDU, and do NDU as a standard course of business. I think they should all claim that they are NDU platforms. Periodically in their life, they failed to live up to that goal, but did it in a way that was a good trade-off for their customers. In each case, they offered the choice to the customer to defer the disruptive change. I genuinely think that's exactly the same situation with XtremIO.

Keep the dialog/debate going - I just am skeptical whether I'll ever persuade you.


One more observation - interesting to note how polarized the comments are...

It's a reflection that this post isn't about "EMC good, others bad". This post is about "XtremIO update" and "poop flinging - by any - is bad".

My experience reflects that of some of the comments: EMC of today is not the EMC of old (the 90's).

Brands take a long time to change. Negative examples are the exception rather than the rule (at EMC and others). The problem is that the bad examples echo LOUDLY, and persist for a LONG time.

There are bad apples, and there are also good apples that fall off the wagon (I fall off periodically - and always regret it).

Thanks for the comments all!

Jim Millard

Simply put: it takes _effort_ to follow Chad's manifesto. Worse, some of us are in "trusted advisor" roles and need to be able to articulate why we recommend one vendor's solution over another, and sometimes it feels faster and more efficient to explain why a given product was *not* chosen rather than reiterate why the recommended solution stands on its own merit. "Why didn't you recommend Y?" leaves the natural inclination to talk about Y instead of X.

One luxury that following the manifesto does do, however: you never have to worry about negative comments coming back to bite you, whether it's the next sales opportunity or the next job opportunity.

Mark Burgess

Hi Chad,

I have spent quite a bit of time looking at XtremIO and as a piece of technology I really like it, but for me it is just too enterprise focused and does not sit well outside of that space.

It would be great if it could scale-up as well as out and also get more performance out of the 32 cores and 6U of each X-Brick (which in turn would make scale-up more important) - on a per core and per U basis the performance can be easily beaten by products like NetApp's EF-Series (obviously it does not have any efficiency technologies, but it does go to show how much CPU resource are required to do de-dupe, etc.).

As you say in your post most non-enterprise customers want a "Swiss Army knife" of a product (i.e. VNX or FAS) and surely we are approaching a point whereby their flash capabilities are good enough for most organisations.

In the medium term does the all-flash array actually make sense?

The HDD is going to be with us for some time and even when it does disappear their will still be different tiers of Sold State storage (i.e. performance and capacity optimised) so all arrays will need to continue to handle multiple drive tiers.

This applies equally to Pure as well so it would be good to get Vaughn's comments on this.

I even wrote a blog on the subject at http://blog.snsltd.co.uk/does-the-all-flash-array-really-make-sense/

Your thoughts would be most appreciated.

Best regards

Chris M Evans


I agree on all the other examples you used and I believe I called out at least the NetApp issue when that happened. In fact I think I've made more comments about the whole disruption of 7-mode to C-mode migrations (which can be a nightmare) than anything else.

In terms of longevity, 4.0 isn't actually available so it's still true to say that upgrading to the current release is disruptive and destructive. When 4.0 is GA and available, then upgrades to the current release won't be disruptive from the previous release, but (if I understand correctly) any 2.4 upgrade will still be an issue. I guess there's a cutoff point as 2.4 deployments shrink where the upgrade issue simply becomes a footnote.

Fair play from you to call out the issue when you did, but the corporate message hasn't changed and I doubt the sales message on the doorstep hasn't either. I would speculate that the majority of your blog readers are technical focused so the honesty around the 2.4->3.0 upgrade was probably lost on those with the cheque books. This lack of transparency at a corporate level is more my issue, rather than with your honesty (which is not in question).

So as 4.0 is released, can we expect, for example that systems in the field will be capable of hardware expansion with no impact (outage or otherwise)? Will this be classed as non-disruptive? I'm looking forward to finding out in good time.



No beef with EMC or XtremIO here.... In fact I hear a lot of good about XtremIO

What I do have beef with is making the suggestion that negative viewpoints should not be believed. That's a very dangerous and wreckless thing to advocate. All view-points MUST be considered when any making big decision -


Yanni Mendleson

Not going negative at all!

XtremIO, the only scale-out architecture that doesn't scale...

Want to go from 1 X-Brick to 2? Let us ship you 2 X-Bricks, you get to migrate your data from the original X-Brick and ship us back the original X-Brick.

VPLEX, Recoverpoint and XtremIO, oh my!! (to simply replicate and protect my data?)

Will I run into Recoverpoint capacity licensing issues when trying to create volumes on XtremIO, or is that fixed now? Really caused me alot of pain!


Native Recoverpoint sounds great, but some of the information I have seen is conflicting. Will 4.0 support synchronous replication without a VPLEX or just asynchronous? For example, only asynchronous replication is mentioned at


The comments to this entry are closed.

  • BlogWithIntegrity.com


  • The opinions expressed here are my personal opinions. Content published here is not read or approved in advance by Dell Technologies and does not necessarily reflect the views and opinions of Dell Technologies or any part of Dell Technologies. This is my blog, it is not an Dell Technologies blog.