Virtual Geek

A couple of EMC/VMware issues to know about….

Every day there are a bunch of “hot issues” that come up – most aren’t worth a broadcast (which can actually cause more damage than it helps) but these two are (IMO) worth drawing attention to for the sake of the customer.

• Emulex HBAs with vSphere (frankly – this isn’t an EMC issue per se, but since customers look to eLab for HBA firmware revs…)   It was discovered by Ralf Kastl (who is just a GREAT EMCer – thanks Ralf) during a customer case.

  “while working on SR XXXXXXXX we figured out that there’s an Issue with Emulex HBA  FW 2.0.0. A3 used by LPe12000 Adapters with ESX 4.x.

  Anytime the connection between the SAN switch and the CX SP’s were disconnected (i. e. during SP reboot), the ESX Servers are unable to reconnect automatically.

  This is caused by the used HBA FW Version, which is 2.0.0.A3 currently recommended by EMC (eLab Support Matrix)

  We’ve done excessive testing to isolate the problem and it’s definitively an issue caused by the HBA FW.

  The newer version 2.0.0.A4 is not affected by this problem, after implementing this version all seen problems are gone.”

I dug into this deeper with eLab and they got back to me lickety split.  The issue was reported to Emulex in March; we received a fix (2.0.0 A4) in late April not in time to retest and publish (in the eLab Support Matrix) in May.  However, the June 3rd ESM has been updated with the A4 FW rev.

SO – if you’re using the pretty common LPe12000 Adapters (which are great), make sure you’re on the A4 firmware, or you might have a bad day during an SP reboot (which also occurs during array OS updates for CX FLARE and VNX OS updates

UPDATE (July 20th, 2011, 10:48am ET) – A REALLY good post on HOWTO here (thank you Jon Kohler)… http://www.jmeow.com/blog/2011/6/19/managing-emulex-hbas-on-esxi.html

• We’re experiencing a couple VMware/EMC cases surrounding VAAI use.   Sometimes being the #1 choice when it comes to storage for VMware (including in some of the largest environments) means we uncover these things – I wish we didn’t, but we do.  The case counts I would still call rare, but we can reproduce the scenarios.   There is a VMAX epack planned.  On the VNX side, there’s only one VNX case, but reproduce-able – we’re working out the best way to resolve.  Stay tuned.

Cisco UCS–one person’s take 2 years in…

It’s been an interesting 2 years, and a milestone for many – we’re coming up on the 2 year anniversary of the Cisco UCS FCS, which will formally be on June the 20th.

I remember it like it was yesterday.   It was the early part of 2009 when I was pulled into a series of meetings with Soni, Ed, Prem, Luca, Mario and a small set of others who were part of the Nuova Systems startup (spin-out/spin-in) that developed the Unified Compute System.  That first meeting led to many others (see my notes on one of those trips here)

There were really two things that stood out in that first meeting:

1. The product itself was pretty cool.   The way they tightly coupled the server to the network, and a focus on “carrier-like” management models was very funky.   The memory expansion was also neat, but the “make it really easy to manage in a stateless way at scale” was the standout to me.  The hardware was also swank in the sense that I loved that the blade chassis was in essence a dumb part.
2. This was going to trigger all out war in the industry :-)

A lot of people were confused by that second point.  I heard several customers say “why couldn’t the status quo of me using HP for servers, Cisco for networking stay?”.  

The answer is that it was inevitable.   Sometimes in our day-to-day heads-down activities, we miss the “big picture”.   The “big picture” on this front is that “traditional” servers, network, and storage continue to be commoditized every day.   Virtualization and cloud are accelerants of this trend.  

If you’re Cisco, HP, IBM, EMC, or other big players – you see this as an existential threat to your existing revenue streams and business models.   At the same time, it’s a huge opportunity if you can rally around it.  

There are a couple ways to try to rally:

1. The path EMC is taking:  i) try to make your great, successful and existing cash cow products like VNX and VMAX adapt rapidly to fit the new trend – a huge task; ii) while simultaneously investing in organic and inorganic innovation in the adjacent markets – like Greenplum, VMware, Isilon and others); iii) expanding into different go-to-markets like SMB and service providers.
2. The path others are taking (and EMC is also participating in through the VCE Joint Venture) is to embrace the commoditization of the components by offering highly integrated stacks.   UCS is in a way a great manifestation of that idea, as it integrates directly two of the stack components (servers and network), and you can see that Cisco’s recent acquisitions of Tidal and Newscale highlight that Cisco continues down that path.

The virtualization and cloud waves are industry-level forces (like the shift to x86 and away from RISC, or towards ethernet for everything) that are bigger than the giants (including EMC, Cisco, HP and others).   As large as we are – if you don’t surf these waves, they crush you.

If you think HP wasn’t going to ramp up their core switching/routing business and go after Cisco’s bread and butter, you’re perhaps being naïve.

So, with all of that said, in the early days, the big server vendors marginalized Cisco UCS in the field saying (I literally heard this with customers) “they weren’t serious about the server market”… I remember thinking to myself: “wow, you can say a lot, but that one doesn’t make sense to me – they are clearly DEAD serious about it”.

So – 2 years later, Cisco is now officially (according to IDC), the #3 player in the blade market.

Anecdotally, many, many customers I talk to have UCS – in fact, north of 30% of the customers I engage with have some.    We have many customers (hundreds of Vblock, many many more of UCS with EMC in the traditional mix/match model).   

UPDATED May 31st, 2011, 2:52pm ET:  I’ve had many readers ask me about my take on the C-series (rackmount).   They have always been fine rackmount servers, but until recently not that differentiated from their peers.  HOWEVER, with the addition of the ability to use virtualized adapters in the C-series, and manage them via UCS Manager, they have inherited a lot of the mojo from the B-series.   My team uses C-Series servers extensively (they powered a lot of the Chad’s World Live demos at EMC World, and our mobile VMUG demo kit is all C-series.   In Vblocks, they use C-series for the AMP modules (standalone management modules that run vCenter, remote management and other functions).   I don’t see as much customer demand for them as I do the B-series, but frankly, think that’s as much because Cisco isn’t marketing/selling them as hard as they are the blades.   I there are a ton of use cases for rackmount – particularly in embedded/OEM scenarioes (look at how Cisco can use them for Unified Communications as an example).

The voice of the customer matters most to me.   Maple Leaf Foods.  Alcon.   Morris Hospital.  Interestingly, all the 7 customers on stage at Chad’s World Live were joint Cisco/EMC customers – and almost all had, or were considering getting UCS.  The common thread I hear from customers:

• UCS is simple.
• The integrated network/server idea WORKS – in every way.
• UCS performs very well.
• UCS is a rock.
• In spite of the FUD we’ve heard, UCS is less expensive than we expected, and is MORE than competitive.

Congrats Cisco!   While there’s a lot of continuous adaptation needed (for the big and the small) to survive and thrive (and yes, those changes can be scary and involve some pain) look at the results from the innovation that was and is Cisco UCS for inspiration of what future days at Cisco can bring!

BTW – while I’m mostly focused on VMware, it’s inevitable that I spend a ton of time with Cisco.   We’ve done a lot together, and will do more in the future.   A couple examples below: 1) a fun video from EMC World; 2)The Cisco write up of EMC as a customer using UCS for our most mission-critical workloads – yup we both use each other’s stuff very extensively; 3) a Cisco Validated Design (because not everyone wants the “product” that is Vblock, some want a recipe and make their own) around Cisco’s VXI initiative.  

You can expect to see more, and more, and more down these fronts in the years to come.

 

EMC World 2011–vLabs = aka the miracle in Las Vegas :-)

I’m going to do a first today, and do a guest post.  This post was authored by Chris Horn, who leads the vSpecialist Tech Enablement team – who were the folks who made the EMC World 2011 vLab hands-on-labs happen.

BTW – I know some folks question whether I do my own blog posts.   For what it’s worth, I write them ALL.   I produce a lot of the demos on my own lab systems.   Occasionally, someone else on the team produces a demo, and I will link to it (though in those cases, I always try to credit the source).

At EMC World, a lot of people ask me how I find the time to blog.  I mostly author posts in advance, and then just post at the moment.  

I do have a several advantages that I leverage.  

1. While at my core, I’m a geek through and through, I am also a senior executive at EMC and deeply plugged into EMC and VMware engineering, which means there’s long-window awareness of many things well in advance.  For example, my EMC Greenplum Hadoop and Project Lightning blog posts from last week were something I started on months in advance, and finalized the week before EMC World. 
2. There’s a huge volume of information to digest.   I’m lucky to be a speed reader, so I can suck it up pretty fast.
3. I usually write on the plane.  It’s ideal “downtime” to put down thoughts.  I capture screenshots, demos, etc, and put them up on Youtube and other things in advance, then compose while in the air.   Since I fly a LOT (500K miles travelled on Air Canada alone last year) – it gives me a lot of time to write.
4. My role enables me to do several days of “in the lab” time after a sprint of several weeks – I find that lab time to be a key part of keeping “whole” – both to return to a calm Zen state of mind, but also to keep my chops fresh.

I’m not complaining – but my deepest professional insecurity is loosing my technical edge as I move into more senior roles (today, there are several hundred folks on my staff).  So, I work hard to avoid atrophying technical muscles, and am perhaps hyper-sensitive to the possibility of becoming a talking head :-)

So – without further ado – except one more heartfelt THANK YOU to Chris, his core team (Tee, Simon, Nick), Nancy Wilga on the EMC events team and everyone else who did their part to support the EMC vLabs – here we go!

Read on for the “Chris Horn report” on BEHIND THE SCENES (including hardware/software, stats, logistics) of the EMC World vLabs!

VMworld 2011–Session Voting HOWTO and recommendations…

The ink is barely dry on EMC World (which was great – still need to do the vLab HoL and summary blog posts – aiming to have that all wrapped up by tomorrow)… and already VMworld 2011 planning is in full swing!

(it’s amazing to me how much my annual calendar is now built around Cisco Live, EMC World and VMworld – and equally amazing how much work they take…)

One of the things that happens earliest are the major product planning pieces, but RIGHT after comes the session submission and voting cycle.  The public voting closes out on May 18th – SO VOTE NOW.

The way this works within EMC is I beat the bushes for people across the company to make their own submissions – everyone from the field, to the product teams, to the services and support folks.   My rules are simple:

• DO: Remember that this is VMworld: the audience is technical VMware users.
• DO: ALWAYS include a customer, and ideally co-present with a key partner.
• DO: ALWAYS include a demo.
• DON’T: Make it an EMC product commercial – the audience is not there for that.
• DON’T: Use EMC product names in the session title or abstract – and a minimum in the session.   If possible, make it a multi-vendor session.

So – after a lot of work… not in beating the bushes to create demand – loads of people wanted to submit – rather Dave Henry (a great vSpecialist team member) and I helping people make sure they made their proposal the best they could, here’s where we ended up (notably right in the middle of EMC World!)… 

EMC as a whole made a total of 103 session proposals, aligned to the tracks VMware has outlined in the following breakdown.

1. Business Continuity: 14 proposals
2. Cloud Application Development: 2 proposals
3. End-User Computing: 14 proposals
4. Enterprise Applications: 12 proposals
5. Enterprise Hybrid Cloud: 11 proposals
6. Management and Operations: 21 proposals
7. Security and Compliance: 7 proposals
8. Technology and Architecture: 22 proposals

In the end, the selection process is pretty brutal.   Only a small subset will get selected by VMware – so every vote counts.

Here’s my ask and my commit to you in return, dear reader:

MY SIMPLE ASK: please go to the VMworld 2011 session listing here.  Create an account if you don’t already have one and click on the “Vote Now” button on the landing page.  Vote for the sessions YOU want.   IF you want to filter for these EMC sessions, you can filter by hitting “search options” and filtering for the keyword “EMC”.  When you vote for a session, you will see a “thumbs up” icon.

MY SIMPLE COMMIT: In years past, the sessions we have done have gotten quite good attendee rankings (which I follow diligently, and reward and show preference for their submissions the following year).   You have my commit that these sessions I will work non-stop to make sure they are the best possible for YOU. 

If you’re not already using the “Everything VMware at EMC” community here, and following Dave Henry, I would   You can see his post on this topic right here.   If you want a more detailed breakdown on my recommendations from the 103 sessions, read on…

EMC GreenPlum Hadoop

I remember when we first started to really look at the Big Data Analytics space 2 years ago – one of the first realizations was a simple, clear, and blunt observations: If we think we can provide value and compete in the data analytics space with a storage array, whether it be SAN, DAS, NAS – we’d be smoking something.

Don’t get me wrong – there’s a place for external storage, but man, it’s a tiny niche (for “not so big big data”).

In these markets – the software stack was and is the most important thing, and to offer it as pure software and as appliances.   Yes, people wanted it to be packaged in a variety of forms – some as integrated software/hardware appliances (think the GreenPlum appliance, Netezza, Exadata), but sometimes they wanted it as software and run it on their own hardware.

Hence – the acquisition of Greenplum which is the leader in the software-based scale-out shared nothing structured data analytics space, and then shortly afterwards, the first EMC Data Computing Division appliance – the GreenPlum Appliance.  

It’s also important to note that the Greenplum acquistion not only brought critical technology, but as importantly – a ton of new DNA into the EMC family.  That DNA around Big Data Analytics is captured in the brains and the hearts of the Greenplum folks.

In Pat Gelsinger’s keynote – Matt from Greenplum showed the power that can be delivered by the model – doing advanced queries on massive datasets (live) – in seconds, by leveraging hundreds of Intel Xeon CPUs, terabytes of RAM, a ton of SSDs – all in a shared nothing pile of servers running Greenplum.

But – right adjacent to the structured big-data use cases are the semi-structured use cases where Hadoop, and open-source software framework designed analytics with that type of data dominate.

So – with the announcement of GreenPlum Hadoop Enterprise and Community edition – EMC is now a huge player in this space.

Don’t take my word for it.

“All of a sudden, EMC looks like the strongest Hadoop company going, and everyone else needs to figure out in a hurry how they’ll counter today’s landscape-altering news”

It’s a big deal.    It also makes EMC’s Big Data Analytics stack more complete (I would argue one of the most complete and advanced in the industry)– and one appliance can accelerate analytics of both structured and semi-structured data.   

Furthermore – the EMC Hadoop distribution can deliver 2-5x the performance of what customers are doing today with Hadoop, and also adds a lot of other things that Enterprises are looking for (snapshots, replication and a lot more on both the technology and packaging/support front).

Within the community, response I’ve seen so far seems very positive.   A big part of participating in anything that at it’s core is open source is about partnering, and also contributing source code back into that community.    It’s important that’s what the Community Editions of BOTH the Greenplum database and the Greenplum HD are there for.  

 

 

Exciting stuff – and it’s amazing to see (and be a small part of) the ongoing evolution of EMC – I love how we continue to adapt, expand, change, integrate as the market changes.   If all we did was DAS/SAN/NAS, wow – it would be tough to have a serious conversation about the world of Big Data.

VNX/VNXe Updates – Bigger, Faster, Stronger

VNX and VNXe started shipping in volume in March, and demand has been overwhelming – which rocks!  

So – 2 months have passed since volume shipping, and 5 months since the announcement, which means it’s time for a TON of new stuff :-)

The VNXe has been a hit – customers, partners everyone loves the extreme simplicity, advanced functionality, multi-protocol block/NAS capabilities, VMware and application integration built in, tight packaging, and of course the low cost.

When the VNXe was being developed, we knew that the SMB market was something that EMC didn’t really have a strong track record in (Iomega notwithstanding), so we needed to really nail it.   But, we also suspected that there were a ton of OEM use cases that we had never addressed.  Since then I’ve been pulled into the coolest discussions about this.   VNXe’s in tanks, on boats (even using VPLEX to have stretched vSphere clusters across larger military vehicles so they can withstand substantial fire and damage – wow).   And that’s the just beginning of the OEM use cases.

So – perhaps it’s not a surprise to see an OEM application – Themis has partnered with EMC to create a hardened, ruggedized VNXe for a variety of military uses – pretty darn cool.

On the VNX front – the list is long.  

We haven’t been talking about it as much as we could, but the amount of bandwidth you can drive through a VNX is very impressive.  10GBps is possible on even the VNX5700.   That’s a lot.   That’s 80Gbps, or put another way, the bandwidth of 8 10Gbps intefaces, or about the bandwidth of a loaded UCS configuration.

For perspective, you can do about 15GBps on a fully populated VMAX (which of course is usually selected less for throughput and more for transactional IOps performance and scale), and of course, on a loaded Isilon cluster, you can do 50-80GBps.

But unlike the VMAX and Isilon which are built for purpose (“tactical nukes”), the VNX is simple, efficient, powerful utility/swiss army-knife storage – used for all sorts of blended use cases.  I think that 10Gbps range is pretty impressive.   One of the reasons was the jump to 4-lane 6 SAS on the backend.   But for some customers, even that isn’t enough, and there’s still headroom in the storage processors.   So, we’re introducing the “High Bandwidth” VNX configuration.   In this configuration, the already impressive backend bandwidth is jacked up by a factor of 2x.

Then, the 25 spindle 2.5” drive tray has started shipping.   These are cool, but if you ask me, have a surprisingly limited life – at least in the way that people are thinking about them.   

Most customer look at them and go “hey SAS disks are smaller (physically) and can be less expensive, and use less power – so that’s good”.   Well, yeah.

I REALLY think that before you know it, we’ll be in all SSD configurations (SLC and MLC), with magnetic media being used for very large capacity/$/W/sqft scenarioes (for which they will exist well through 2015 – again, IMO).

So with that in mind, it’s better thought of as a “25 SSD IO eating enclosure”

But – what these new enclosures represent are tip of the iceberg for very dense SSD scenarioes.    And yes, of course we’re looking at all sorts of stuff, including ones where potentially we leverage form factors that eschew traditional flash packaging (think of how Apple used the mSATA form factor in the MacBook Air).

BUT – sometimes it’s not about IO capacity, but is about how dense the platform can be when it comes to GB/array/floor tile.

So – we also continue to expand capabilities on that front – here you can see the new 60 spindle 4U enclosure.   This enclosure was codenamed “voyager” for EMC trivia buffs.   This lets you fit 60 3.5” disks in 4U, or put another way, 600 disks in a rack, or (coupled with the new 3TB spindles) 1.8PB raw per floor tile.   That’s a lot.

 

 

 

 

 

 

 

 

A quick “heads up” – these require a special “dense rack”.  They can be added to existing VNX systems, but not into existing VNX racks.    BTW – the majority of these dense configs literally configure the entire array that way.   Also – in case anyone is wondering about 2.5” disks – of course, we’re working on the next-gen “dense” configs.  For now, when you want dense GB configs – 2.5” disks are a bad choice, you get far more bang for your buck from 3.5” disks.

Also – there’s a pile of software updates that were announced.  Unisphere continues it’s rapid march of progress.   BTW – we’re finding many customers are selecting VNX after trying Unipshere.  Customer feedback is overwhelmingly positive (it’s not perfect, and we’ll continue to improve) .

Unisphere version 1.1 (remember from my VNX/VNXe launch posts about how the teams work together, driving innovation in a producer/consumer model to each other?) continues to get better.   VNX customers get the VNXe goodness of application-integration right in the UI itself on top of the vCenter API integration that was already there.  And yes, if you look closely, there is a VMAX tab in that Unisphere screenshot :-)

Unisphere is also getting key awesome sauce that is in EMC Replication Manager embedded directly in the platform.  EMC and NetApp have led the way amongst the storage vendors with the idea of “application-integrated” replication – EMC with Replication Manager and NetApp with the SnapManager family.

While customers dig those tools – one thing that we’ve really been focused on is making the “time to value” and overall solution simplicity better.   So – rather than just creating a Unipshere “snap-in” for Replication Manager, this project (again, for trivia buffs, this is code named “project Archway”) will EMBED the majority of those functions directly in the platform itself – making management cake.  this part is a bit of a preview of things two come.

BTW – in case you missed it, and you that no matter HOW easy we make storage management UIs like Unisphere… for the people who run the applications preference would be to NEVER LEAVE your application UIs.

Hence, the EMC Virtual Storage Integrator (VSI) which plugs into vCenter, as an example. 

The week before EMC World, we launched EMC Storage Integrator (ESI).  ESI is the equivalent of the Virtual Storage Integrator, but instead of vCenter integration (“home” for VMware admins), it integrates with the Microsoft MMC tools for Sharepoint, SQL Server and other apps.  We also announced our SCOM Pro pack.  

I’d highly recommend that any customer using any Microsoft technology (..ahem.. that means you, because it means everybody) check it out.   Like VSI, ESI is completely FREE.   You can read more about it at my colleague Adrian Simays’ site here.

Phew… Is there more?   YES!

We’ve launched the Cloud Tiering Appliance.   This enables VNX customers to take files and based on policy, to automatically extend the idea of FAST VP right out of the platform, and right out to

 

What else?   Well – we’ve integrated the VNX filesystem with API-level integration with the Google Search Appliance.   This is pretty cool – through this integration, finding your corporate info becomes much faster and easier.  

The Google search appliance doesn’t need to crawl the VNX filesystems, the VNX literally pushes the info to the appliance.  Cool.

2011 has been an EPIC year so far for EMC’s midrange and SMB customers – and… we’re only just getting started :-)

VPLEX Geo = Awesomesauce.

So – VPLEX Geo is now GA.   For the first time, it’s possible to have active-active geographically dispersed storage models across Async distances (~50ms).   It’s going to take some time for this to “sink in”.

The VPLEX Geo announcement is an the GA (and literally available now) of VPLEX GeoSynchronicity 5.0 (the software) and new VPLEX hardware (bigger, faster, stronger).

So – what does this mean in a nutshell?

1. VPLEX now has a 3rd site witness option.
2. The VPLEX Metro (synchronous) use cases with VMware have been hardened (partition behavior works much better now, and has been developed hand in hand with VMware folks working on the next version of vSphere).   BTW – stretched vSphere clusters over async distances are, as of right now, still not supported though VPLEX Geo is out.   Stay tuned for more on the topic.
3. There’s a new class of options – where certain workloads can be active-active across asynchronous distances.

BUT – beyond that (which on its own is awesome and extends the “category creator” nature of VPLEX), there’s a lot more.

The core technology (distributed, coherent transactional cache) behind VPLEX is what we think of as a “foundational technology”.   While it’s first manifestation is as a product (VPLEX) which virtualizes external storage and makes it active/active across geographic boundaries – it has MANY interesting applications.   BTW – if you look at the Project Lightning blog post, you can imagine how the technology inside VPLEX could also help there.

But – here’s another FASCINATING idea of how it could be used.   This is a real Proof-of-Concept developed by the EMC CTO office in conjunction with the VMware vmkernel team.   It’s a response to an interesting customer challenge…

This customer runs large HPC workloads, and like many in their position, one frustration is that sometimes a job runs for several days, and then, before it finishes, a new job comes that MUST run immediately, making the days of prior work wasted.

They asked: 1) Could we capture distributed system state as “one big consistency group” – both the compute, memory, and dataset all together; 2) could we “push those workloads to a public cloud”.   We said “sure, let us work on it”.   Then, they added a tough twist: “we don’t want to ACTUALLY move our dataset to the cloud provider – but some of them would suck over a WAN – can you help with that too”

So…  After developing a custom vSphere vmkernel that enabled capturing system state across many VMs with exact synchronicity and a dramatic acceleration of the snapshot and resumption process… we integrated in something REALLY cool – the ability to “push out” a virtual VPLEX (vVPLEX if you will) instance which presents the storage to the VMs as they are running in the cloud – without the dataset ever actually really existing at the service provider using VPLEX’s AccessAnywhere approach.

Check it out:

Download the high-rez version here in MOV and WMV.

So – here’s an interesting question dear readers… This customized vmkernel would expose this “consistency group accelerated snapshot via an API” and could be used for all sorts of stuff.   Right now, this may/may not ever make it in, the main question being “is this something customers would like, and partners would leverage, or is it just relegated to HPC?”  I can think of MANY use cases, a ton of which cross all sorts of use cases.

But – would you want this to become a general vSphere capability?

EMC UIM v2.1 Provisioning and Operations

4Today, we released UIM v2.1 Provisioning Center and Operations – adding important new capabilities – the abilities to see events and issues, correlate them with the service catalog and the vApps.   Also, Provisioning Center added core vCenter integration.

As a reminder, EMC isn’t trying to build “uber managers”, or “uber orchestration” – UIM Provisioning Center and Operations provides a simple, unified API for Vblocks.  You can use it directly if you want, but of course, it can integrate with your existing BMC, CA, or Tidal/Newscale tools – making the environment simpler for those folks to integrate to – one API for all the physical infrastructure.

The other thing I wanted to point out – UIM manages physical infrastructure – integrating with vCenter.  Sure – it provides a PHYSICAL service catalog (think “catalog of automated, standard vSphere cluster builds”), but it’s not intended to be an end-user self service portal.   Like Vblock, it makes getting to ITaaS easier – accelerating that process for customers, but you need things like vCloud Director or other orchestration/portal on top to really have ITaaS.

If you’re wondering how this works in the larger EMC/VMware management picture, this picture tells the story.   It’s a manifestation of our goal to help customers build Hybrid Clouds.

As is most often the case, SEEING is better than explaining – here’s a demonstration of how it all fits together, delivering ITaaS in a hybrid cloud use case…

You can download it in high-resolution MOV and WMV format.

So, What do you think?   Curteous comments/critiques always welcome!

Understanding Project “Lightning”

This is one of the bigger pieces of news coming out from EMC World 2011 in my opinion.

For those following the industry, it shouldn’t be too much of a surprise – but it strangely will be..  Flash is going to change everything, and exist everwhere.   It’s no different than any fundamental technology innovation.  When something is an order of magnitude (or more) faster, cheaper, smaller – it doesn’t get used in ONE way (as a tier, as a cache, as an extension of server memory), but tends to get used in MANY ways.

Flash is one of those examples.   It is literally 100x-1000x faster than magnetic media, and 10x-100x cheaper per unit capacity than DRAM.  

So – why would ANYONE think it would only replace high-performance magnetic media (like 10K and 15K RPM disks in the form of automated tiering like EMC FAST VP), or only as an extension of cache (like EMC FAST Cache), or only as an extension of host system memory (like Fusion/IO)?    We certainly don’t (and note how consistent we’ve been on this point – I’ve personally pointed it out over the last 2 years, here, here and here as examples).

I’ll quote myself (just to point out – we work/think about these things for a long time, so when we put something out there early, like Lightning, there’s a LOT behind it)

“Now you can see why I go on and on about Flash being transformative, and how it will be applied EVERYWHERE.   You’ll see it in servers, you’ll see it in array caches, you’ll see it in non-volatile storage tiers, you’ll see it in laptops, desktops, mobile devices – EVERYWHERE.   Just like “protocol passionistas”, I would tend to not listen to people who say “one way always”.   It’s just as likely that it’s simply what suits them.”

This is likely something everyone is going to need to do (like auto-tiering, like flash-assisted caches, like dedupe, compression, etc…)

So – what have we announced?  Well beyond making our broader flash strategy more clear (including: all-flash VMAX and VNX configurations; that we are going to do MLC in addition to SLC; and more) – the big news is outing the fact that for many moons, we’ve been quietly working on a project to not only deliver a server-side flash-based PCIe card, but more importantly – an end-to-end solution around applying the idea of “tiering” from the server all the way to the array.

If you’re interested, read on for more…

Twas the night before EMC World 2011

EMC World is about to start…

Remember – follow online using the #EMCworld hashtag, and on www.emc.com.  

If you’re an attendee – you can download the EMC Folio app on the appstore (just search for EMC Folio)

Here’s my “the night before” walkabout….