Nigel Dessau

The Next Logical Step

Welcome to my last blog as the “Storage Marketing Guy” at Sun. Just over two years after coming with the StorageTek team to Sun, I am heading to a role in the Strategic Alliances and Licensing team. I am really excited by the opportunity and will blog more about it when I know which way is up!

For the Storage team there is also change.

David Kenyon will be taking over as VP of Storage Marketing at Sun. David was with us at StorageTek and has been running the products management and business operations team for a while. Perfect guy for the job.

It was also announced today that we are combining the Storage team into the Systems team. Or, more realistically, we're combining the Server team and the Storage team into a Systems team. It may sound like semantics, but it's really not. If you look around at all the major “systems” companies you find that the server team and the storage team are always under the same umbrella organization. Given that Sun's focus is to be a Systems company, probably a higher corporate priority for Sun than any of the other major “systems” companies, that's a pretty good place to be. And it will also create some synergies.

Now as I learned through the STK acquisition – synergies can go both ways. While many people think they mean lay-offs, in the Sun-STK action they actually meant revenue and development benefits. Let me give you two examples:

• The Sun StorageTek VTL-V solution runs on a Sun X4500 (aka Thumper), running Solaris with ZFS. It is a Virtual Tape Library that can backup Windows, Linux, HP-UX, AIX and Solaris - it just happens to use Solaris as its OS. In STK days we would have had to license, buy, rent or invent nearly all of that for ourselves. Clearly the StorageTek solutions are better as part of Sun.


• The storage market is changing rapidly – files over block, iSCSI, SAS, Objects, Integrated Search, File Systems – I could go on. More and more these sound like a cross between an OS, a server and a storage device. They are and there is no point developing the same thing twice.

So, it is clear to me this is good for Sun, but is it good for our customers?

When I talk to customers they really all ask for the same things: quality products that deliver high value functions at a low cost and with ever improving 'eco' characteristics. I think the integrated development team will accelerate us along that path and will therefore deliver what the customers want.

Here's how:

• quality products require good design, good development and good test. Each of these areas is better off with a larger 'talent pool' than a smaller one. The Sun StorageTek solutions will get better as we share more with the Systems team. From ASIC design to physical packaging – there is much that can get better.


• high value function is more and more being delivered as part of an OS. OK, it may be delivered as a appliance but it is function that's going into the OS or the file system. Think of the VTL-V examples and what ZFS can do to offer a RAID capability without the need for the hardware and with no real performance impact.


• lower cost can come from common components, shared resources and volume. All come with the larger group.


• Eco – if you saw my blog of a couple of weeks ago, you will know that our 'Systems-like' approach is one of the key elements that differentiates the Sun Storage Eco story – so we have more to build off.

I think the other interesting aspect of this, starting to think about my new role, is to ask the “off platform question”.

What happens to Sun Storage sales to the customers that don't connect to Solaris?

The obvious answer is that we will of course continue to sell tape and disk to the those customers – all of whom will get better solutions and faster in this new world. But let's explore what “off platform” means in at Sun today.

At Sun we can help you with your Solaris needs, but we can also help you with you Linux needs and now we can help you with you Windows needs – so there is a lot of stuff that's not off platform anymore! Moreover, with millions of Solaris licenses being deployed on x86 on HP, DELL and IBM servers and partners like IBM committing to Solaris in public – what is the “platform” we're talking about? A chip or an operating system? It's a different world out there and our solutions and organizations need to change to match them.

But what if our customer's platform is an IBM mainframe?

No problem – happy to do business with you. Did we mention that we have over 80% share in the market for tape libraries over 1000 slots (i.e., tapes)? Did we tell you we have a new mainframe mid-range library coming next year? What about two new tape drives in next 6 moths? An update to our encryption solution just weeks away? The new ST9990V and ST9985V mainframe disk solutions already announced? Sounds like a pretty healthy portfolio to me and it is only getting better. Oh, we will use Solaris to help us make it even better. Should that worry our customers? No, if IBM's VTS can run AIX then we can use Solaris. I know which I would prefer!

So will customers worry? Of course, none of us likes changes within our suppliers – it has the potential to add risk. But in the long run I believe this move actually could do the opposite. It could add longevity and certainty where some would say it has been missing.

Good luck to my friends and colleagues in the storage world. I think the best is yet to come.

To my mainframe friends let me just say this, from me, for the moment, z_eod.

Posted at 01:06PM Oct 01, 2007 by Nigel Dessau in Sun Storage  |  Comments[1]

StorEcology

Someone just asked me to highlight the storage messages from the Sun Eco Innovation Initiative. I thought it sounded like a great blog (decide for yourself!) ...

We believe that Sun Storage differentiates itself in the Eco area around three things:

1. A balanced approach to storage
2. Building better products
3. Being part of the Sun Systems family.

Let's take them one by one.

A Balanced Approach

In a world where customers want to use as little power and cooling as they can – you can't get better than $0. That's fundamentally the cost of keeping a piece of data on a piece of tape. Given we can store over 70,000 1TB tapes in a SL8500 library, it's great on space too. Clearly you have to use power to read, write and move the tapes – but you get the point.

A possible problem with this approach is that you may want some data faster than you can read it from a tape drive. It is interesting to note that Sun has a unique product in the tape market called the 9840 that's designed for fast access – so we give you capacity and speed. Where millisecond access is required, disk is king. Sometimes it's very fast enterprise type disk, like our ST9990V, and sometimes it's SATA type performance, like our ST6000 family of modular storage.

In between disk and tape is virtual tape – whether it is for the mainframe, with our VSM, or for open systems, with our VTL solutions. What's interesting about our VSM and VTL solutions, unlike most of the competition, it's not about eliminating tape, it's about optimizing price performance and Eco through a tiered storage architecture.

Building Better Products

The balanced approach is vital but it is not enough. We have to be able to help our customers with better products at every stage. I am not going to use this blog to make a competitive point-by-point argument for each of our products but here are three eco-examples:

1. Our new low-end VTL-V solutions, based on a X4500 (aka Thumper) with Solaris and ZFS, uses 60% less power and 66% less space than EMC's DL210
2. In terms of power alone, the SL8500 is 57% more “Power Efficient” than the IBM TS3500
3. If value for watt is an issue, in $/W/GB the ST6540 is twice the value of the IBM DS4800 (which are built from some of the same parts)

The 6540 example is an important one because it makes the point that Sun has been thinking about this for sometime. Both the ST6540 and the DS4800 have the same LSI controller in them and probably even use some of the same disks – but we packaged it very differently. So not only will the ST6540 win price and performance benchmarks but Eco ones too.

Another set of considerations are functions like 'thin provisioning' – announced on our ST9990V. Using this technique, customer can increase the utilization of their enterprise disk solutions from under 40% to over 60% - for some from under 30% to over 70%. Doubling utilization can mean you need to buy less boxes, which means less power, less heat, less people to manage, etc.

Keeping It in the Family

Finally, it is about being part of a systems company.

More and more customers buy solutions, i.e. combinations of servers, software and storage, to solve specific business problems – they are not just buying for their infrastructure. In Sun the Systems, Software, Storage and Services teams work together to make sure we can put the best solutions into the hands of our customers. From High Performance Computing to running your ERP systems – we design Eco in, not tack it on afterwards.

Together at Last

So, put these three reasons together and you don't just have a good Eco story – you may just have the best story in the industry.

Posted at 09:43AM Sep 13, 2007 by Nigel Dessau in Sun Storage  |  Comments[0]

Getting Thin for the Fall – Twice the Utilization at No Extra Cost!

Often I am asked what our 'virtualization strategy' is. Like it should be one product or one solution. The reality is that we all have the same strategy – to help customers mask the complexity of managing multiple instances and increase the utilization of the assets, at the same time. For Sun Storage this means – disk virtualization, tape virtualization and with our partners, fabric based virtualization.

It's worth remembering that Sun has an incredible portfolio of virtualization across Systems and Software – which combined with the Storage really provides an unparalleled solution design. It is also very Eco friendly – more in the next blog.

We have just added another part to our disk virtualization portfolio - the Sun StorageTek 9985V system. Like its bigger sibling the ST9990V , the ST9985V has the ability to virtualize multiple storage devices from all vendors and present them to the application as a single pool of resources. It also has the ability to let the application see more storage than actually exists on the physical disks within the system.

So what?

In the past, the number one priority was to ensure that the application had access to enough storage - no matter what. This meant that data center managers had to allocate extra capacity in case there was a spike in demand for the application, and to ensure that there was enough space for all the data copies needed for business continuity or to recover from a disaster. In addition, adding storage when a new application was deployed lead to lots of storage devices that had to be managed independently, that couldn't access each other's resources, and that all had a lot of overhead. There had to be a better way.

Technologies such as the virtualization and thin provisioning capabilities in the ST9985V can help alleviate this inefficiency by giving all devices the same management interface and by reducing the amount of physical capacity required to support the application. More importantly, these technologies can help data center managers move towards a tiered storage architecture where data is stored on a device according to a pre-assigned value set by the organization. Since both of these considerations change over time, it's important that the tiered architecture have the appropriate tiers (primary disk, secondary disk, virtual tape, tape archive, etc.) and you have the ability to move the data from tier to tier according to policy.

Bottom line: Thin provisioning can increase your disk utilization from under 30% to over 60%. That's more power to your bottom line not your next disk array!

Posted at 09:00AM Sep 10, 2007 by Nigel Dessau in Sun Storage  |  Comments[4]

Not Just Being Diligent?

Recently, Sun signed a standard reseller agreement with Diligent, which allows Sun to resell Diligent's ProtecTIER as a field-integrated solution, combining Sun hardware and Diligent software. This has caused some excitement around our partnership with FalconStor and our strategy (which hasn't changed). I thought it was worth explaining what we have done and why.

We've added the Diligent-branded product to our Professional Services 3rd party software price list. This means we can sell Diligent software as part of an integrated Sun solution. In this model, Sun will support its components and Diligent will support its software. We do this with lots of software and it is really designed to support deals in the marketplace.

The Diligent relationship does not change Sun's commitment to its relationship with FalconStor or the current Sun StorageTek-branded VTL roadmap. The Sun VTL roadmap integrates FalconStor base software with Sun software and hardware IP into an end-to-end Sun-tested and manufactured solution. The agreement with Diligent also does not change Sun's commitment to integrate FalconStor's de-duplication software (SIR) into the roadmap which we are working on with FalconStor.

Sun StorageTek VTL has significant differentiation from other VTLs in the market, based on superior testing and integration, integration of tape management capabilities, Sun Solaris operating system, and Sun's server roadmap. Because Sun is the single source for all the components of the solution - software, OS, server, disk, AND tape - Sun is uniquely able to support all components of the solution, resolving all problems with the customer without hand-offs to third parties.

A recent de-dup market study by The 451 Group showed that 28% of non de-dup users planned to buy de-dup in the next 6 months. The Diligent agreement provides Sun with expanded access to market opportunities for de-dup until we have SIR available on our price list and supporting Sun's VTL solution. Diligent's de-dup architecture is 'inline'. FalconStor's approach with SIR is 'post-processing'. The jury is out on which architecture will end up dominating and, as with any emerging market, our enterprise customers will demand a choice.

The Diligent agreement is also in alignment with our strategy to support the applications our customers need. No one would expect us to just support one ERP vendor – why should we just support one storage software vendor? For example, for backup software we work with Veritas, Legato, TSM and BakBone – so is life.

The Diligent agreement is also part of a bigger story to create an application ecosystem around x4500 (aka Thumper). Today we are working with Luminex for mainframe VTL-type solution, Greenplum for data Warehousing, ipConfigure for security/surveillance, etc. The more people who use our x4500 the better!

So in the end it is about choice not about change.

Posted at 02:52PM Sep 05, 2007 by Nigel Dessau in Sun Storage  |  Comments[1]

pNFS: Imagining the future and delivering on it.

Lots of buzz about the first pNFS Open Solaris code drop. For storage people the things that strike me are:

Unlimited Horizontal Scale: pNFS delivers a global namespace for file systems - think of the problem with NetApp, when you fill up one NetApp box, you get another, etc. and then you have to deal with remembering which files are on which box. With pNFS you get one file namespace, regardless of the number of servers.

It achieves this via an industry standard protocol (that's what the NFS v4.1 spec is defining, the over-the-wire protocol used between the clients needing to access a file, and the server that stores that file). This means that vendors can now start delivering interoperable, horizontally scalable solutions. In all other "clustered" file system solutions today, the implementation is proprietary – you have to get all the parts of the solution from one vendor.

The next interesting part is that we are making this available via open source, in order to continue to propagate the standard to ensure customers get high quality, interoperable, next gen solutions for file.

Bottom line: true seamless horizontal scale across multiple NAS boxes or file servers. A desktop application will no longer have to remember which NAS or server is storing their file ... imagine that!

Posted at 06:10PM Jul 16, 2007 by Nigel Dessau in Sun Storage  |  Comments[2]

A Short Blog Entry

In my last blog I referenced Randy Chalfant. He now has his own blog. Have a look and send him a comment.

Posted at 06:01AM May 31, 2007 by Nigel Dessau in Sun Storage  |  Comments[0]

Time to Get Thin (Provisioning That Is)

Utilization in storage is a big issue.

A bank told one of our team, Randy Chalfant, that their Unix based database gave them a 4% utilization efficiency. Unix will commonly have a 40% allocation efficiency. So here is the math for that bank... if they had a 100GB database that was 40% allocated with 4% utilization, then they are effectively using only 1.6GB of the 100 they bought. Not a great return!

Every customer I talk with says they need to do more with less, data is growing exponentially, budgets are tight, and yet they are forced to over-provision storage to make sure they don't run out of capacity. This week we are launching our new high end disk system with thin provisioning features. This plays well into Sun's overall virtualization strategy and offers market-driven advances in ROI and massively improved utilization rates. Improved utilization also means less boxes – which means less power – which means less CO₂ – which means, eco-friendly!

It surprises some, but we've enjoyed some great wins over the five year history of the high end disk portfolio, the Sun StorageTek 9900 family: with over 2700 machines installed and over $2b in total revenues to Sun. The good news is that we'll be rolling the product out officially in early July and naming it StorageTek 9990V. It'll be just what the customers want – a high performance solution packed with advanced data services features that are actually pretty cool and clearly differentiate it from the competition in the market.

Posted at 06:00AM May 22, 2007 by Nigel Dessau in Sun Storage  |  Comments[0]

Reality or Spin?

In a recent blog I talked about our Key Management Strategy. Two comments – one positive and one negative. Thanks for both. Let me deal with the negative. You can read it for yourself but here are a couple of key sentences:

“Do you really expect that Decru will back off OpenKey or RSA will give up years of investment to come make Sun the King of KMS? No more likely than Sun suddenly adopting OpenKey! Until Decru, RSA, NeoScale, Sun, IBM, Vormetric, et al. come to the table and agree to deprecate your proprietary APIs and use a common standard (which doesn't yet exist), this is really just vendor posturing for positive press. It means nothing to us in industry.”

Obviously I wasn't clear or this was a competitor trying to justify their stance (anyone want to vote?). So, let me try again.

• We would love a standard.
• We would love a standard today.
• We would love a standard today that we all agreed to.
• We would love it today not in three years.
• We will do anything we need to do to make it happen.

My only point was that if giving our KMS technology to our partners would move this along I would be happy to do that. Let's think about our customers not about our proprietary solutions.

It's not about the industry – it's about helping customers manage their data explosion better.

Hope that helps! (I doubt it.)

Posted at 07:29AM May 11, 2007 by Nigel Dessau in Sun Storage  |  Comments[0]

Sun's Mainframe Commitment (In German)

I am in Germany this week and have had the request from our local team to re-post my earlier blog entry about our commitment to the mainframe - but in German. So here we go.

By the way, I don't speak German so I hope I'm not ordering Pizzas for the whole company!

Donnerstag, den 15. Februar 2007

Im Laufe der Woche bin ich an dieser Stelle auf einige Fragen eingegangen, die mir auf der Sun-Analysten-Konferenz gestellt wurden. Meine Antworten waren Anlass für weitere Fragen, von denen sich viele mit unserem Engagement für Mainframe-Produkte beschäftigen. Die FUD-Maschinerie von IBM (war ich mal ein Teil davon?) versucht immer noch, einige Kunden davon zu überzeugen, dass Suns Verpflichtung gegenüber unseren Mainframe-Storage-Kunden nicht ernst gemeint wäre.

Viele, die mit mir zusammenarbeiten, wissen, dass ich eine große Leidenschaft für den Mainframe hege. In meinem ersten Job schrieb ich Testprogramme für eine Vector Facility unter VM auf einer 3090-Maschine; Mainframes begleiten mich seit 20 Jahren. Auch wenn ich heute für Sun arbeite, meine Tage als Mainframe-SE lange vorbei sind, und ein Sun Server vieles von dem leisten kann, was ein Mainframe tut, denke ich immer noch, dass Mainframes für unsere zahlreichen z/OS-Kunden und ihre daten- und transaktionsorientierten Umgebungen eine bedeutende Rolle spielen.

Aber ich verstehe, dass sich diese Kunden Sorgen machen. Sun hatte in der Vergangenheit eine negative Einstellung Mainframes gegenüber; ist also das Mainframe-Business von Storage Tek in den Händen von Sun gut aufgehoben? Die Antwort lautet JA. Und, wichtiger noch, wahrscheinlich besser aufgehoben als vor der Integration in den Händen von Storage Tek.

Für Jonathan Schwartz und das gesamte Management-Team ist es selbstverständlich, dass wir mit unseren Kunden zusammenarbeiten - auf allen Gebieten und mit allem, was sie heute installiert haben – Mainframes, Linux, Windows – was auch immer. Jonathan übermittelt diese Botschaft schon seit fast einem Jahr unseren Storage-Kunden, zum Beispiel auf dem Forum im letzten Oktober und erneut auf der Analysten-Konferenz von Sun in der letzten Woche.

Aber Taten sprechen eine deutlichere Sprache als Worte:

• Letztes Jahr haben wir über 8 Millionen Dollar in Mainframe-Hardware und Mainframe-Software investiert. Wir besitzen einen nagelneuen z9-Rechner mit der aktuellsten Kanal-Technologie; wie die übrige Welt tun wir alles, um auf dem neuesten z/OS-Stand zu bleiben. Fred Casanova hat mich gerade erst über den aktuellen Stand informiert. Nach dem Mainframe-Upgrade verfügen wir nun über 5269 MIPS statt über 2100 MIPS zuvor. Unsere LPARs laufen unter z/OS 1.4 bis z/OS 1.8; für die Version 1.6 stehen 2500 MIPS zur Verfügung.

(Nebenbei, Fred ist einer der Gründe, warum unsere Kunden Tape-Produkte von Sun kaufen sollten: Sie werden doch nicht installieren wollen, was Fred zuvor nicht ausgiebig getestet hat?)

• In der Entwicklung haben wir erneut den Head Count erhöht; es würde mich nicht überraschen, wenn nun bei Sun mehr Personen an der Entwicklung unserer Mainframe-Produkte arbeiten als früher bei Storage Tek.


• Auf dem Markt der Virtuellen Tape-Produkte sind wir seit letztem Jahr mit den beiden neuen Systemen VSM4e und VSM5 vertreten. Der Performance-Update für die SL8500 ist seit kurzem verfügbar und innerhalb der nächsten 90 Tagen werden wir die Partitionierung an Sie ausliefern. Gerade seit gestern ist die FICON-Version unseres Encryption-Laufwerkes T10000 allgemein verfügbar – das waren einige arbeitsreiche Monate.


• Voraussichtlich nächstes Jahr werden Sie VSM6 zu sehen bekommen, außerdem eine neue Library für mittlere Kapazitäten, weiterentwickelte Software sowie die nächsten Versionen unserer Laufwerke T10000 und 9840 – alle diese Neuerungen sind für den Mainframe.

Sie sehen, wir arbeiten kontinuierlich an der Weiterentwicklung der kompletten Produktreihe.

Ich verstehe, wie einfach es ist, uns zu fragen, ob wir es mit dem Mainframe ernst nehmen. Aber ich könnte auch fragen, wie gut Tape in der Verantwortung von IBM aufgehoben ist. (Ok, ich habe einiges an FUD verbreitet, aber ich kann jederzeit damit aufhören!)

Ich erinnere mich an das Jahr 1990, als ich die Ausbildung zum IBM-Verkäufer durchlief: 'Niemand braucht Tape. Nur Verrückte kaufen Tape-Silos. Tape hat keine Zukunft.' Dennoch hat das Storage Magazine letztes Jahr berichtet, dass mehr als die Hälfte aller großen Tape-Kunden davon ausgehen, in diesem Jahr mehr Geld für Tape-Storage auszugeben – und auch mehr für Disk. In einem späteren BLOG vielleicht mal mehr über die Zukunft von Tape – aus meiner Sicht lässt sich jedenfalls sagen, sie sieht recht gut aus.

Die Wahrheit ist, dass die meisten, wenn nicht sogar alle entscheidenden Innovationen auf dem Gebiet der Tape-Verarbeitung und der Tape-Automatisierung von STK kamen oder aber wegen des Wettbewerbs zu STK vorangetrieben wurden. Wettbewerb ist gut für jeden Markt, und wir haben nicht die Absicht, IBM den Tape-Markt oder den Mainframe-Tape-Markt alleine zu überlassen.

Posted at 07:28AM May 04, 2007 by Nigel Dessau in Sun Storage  |  Comments[0]

The Key Is the Key (Management) for Tape

So phase one of the great tape encryption debate is over.

Those who had to do it have done it. Most had to pick sides – us or IBM. Sun provides a solution that is 100% independent of the rest of the customer's data structure – hardware and software. IBM has a solution that fits snugly into an IBM environment. Sun StorageTek's customers chose us and IBM's customers chose them. Some moved from one to another. I am sure we will claim victory and so will IBM but if we are both honest – it's fundamentally a draw. Either way – not as many went as we thought they might. So why?

Well first, let's set some ground rules:

1. We believe in 3 years you will not be able to buy a storage device without encryption in it (whether you turn it on or not);
2. When you have encryption you might as well turn it on (when you left the house this morning did you lock all the door and windows or just the ones the bad people can see?);
3. There will not be one key management solution.

Sorry – what was the last one again?!

There will not be one key management solution in the world. Coming to that understanding is holding most customers up.

Despite what we or anyone else will tell you, we don't believe there will be one key management solution in the world. I suspect customers don't want too many but they don't want one either - unless you just want to be locked into IBM mainframes (through ICSF). It's a heterogeneous world and that means multiple key management solutions. If the world is going to have multiple solutions then we need to have a way to pass keys from system to system.

We believe that without this, the world will be stuck in phase one. Customers are asking about how we will share our keys and make life easier. In the short term, all providers of encryption have their own solution for key management – in response to the urgency of customers' needs. This is not ideal, but a natural part of the evolution of the encryption market.

At Sun StorageTek we have a KMS today. Long term we may not want to be in the KMS business and it would be nice to work with the other leading solutions that our customers have to make life easier. The issue is – no one wants to play nice.

So – here is my offer. If you have a solution that needs a key management solution, you can have ours for free!* Yes, we are willing to give our KMS away to partners who want to think about customers and not 'lock-ins'. We want to share and swap APIs so we can share and swap keys.

Now we need the rest of the industry to come and play nice too. Sun is working hard with other suppliers and even competitors to drive towards a universal language for key management that will get us to where we need to be. Until then – we like everyone else will ship our own solution.

* IP and legal team willing! Actually the team has asked me to point out that this means that we will freely share our APIs which are how the KMS talks to an encryption device. I assume that does not mean we are giving away free KMS appliances. Sorry.

Posted at 12:48PM May 01, 2007 by Nigel Dessau in Sun Storage  |  Comments[2]

SNW Ramblings

So after a week of recovering from the excitement of SNW - what did I learn?
Well, I think three trends emerged from the show.

The show floor at SNW was full of exciting and new storage devices. Most use general purpose servers (AMD or Intel – both at SNW this year) and an Open Source operating systems like Linux or Solaris. The problem with most of them (mostly the ones running Linux) is that while the suppliers are getting the economic benefit – the customers don't seem to be yet. Still lots of 70% margin gadgets on the floor.

I say most of them because there are some exceptions – mostly around Solaris and the Sun X4500 (aka Thumper). You could see Thumper on the show floor running Solaris and any number of applications, including:

Five great examples of how a customer can get the benefits of low cost hardware, open source software and storage applications.

I think we can mostly accept that storage virtualization is NOW not tomorrow. I asked people why they thought this had 'tipped' this year and the answer: VMWare. So many virtual servers are causing a huge need for storage virtualization. Still, lots of choices in the market and all will not survive. Time to pick your favorite.

We all know that the role of tape in our customers environments is changing – not going away, just changing. Despite that fact, every year there is another technology that claims to be the death of tape. Not sure I saw one technology this year that made that claim, but there is an emerging set that may further change the landscape.

Imagine this: hundreds of hard disks, that are interchangeable for bigger drives at any time, that can be bought in your local computer store with 5-9s availability and SW RAID, which can be powered up or powered down, where your data is de-duplicated and then backed off to tape. Sound interesting? Sound like deep archive on disk?

Maybe. Time will tell.

Posted at 08:22AM Apr 26, 2007 by Nigel Dessau in Sun Storage  |  Comments[0]

How Cool Is That?

This week is SNW and I think some interesting trends emerged. More next week. For this week - a funny sight. Lots of the demo kits are in the corridors of the Hyatt in San Diego. I noted some IBM and some Sun kits next to each other. I did like the new IBM cooling system.

Posted at 04:39PM Apr 20, 2007 by Nigel Dessau in Sun Storage  |  Comments[0]

That's one small step for Sun, one giant leap for the storage industry...

A couple of weeks ago I shared my thoughts on storage and Solaris. As I mentioned, one of the things we MUST do is grow the community of storage developers and Solaris users.

Today we announced the launch of the Storage community within OpenSolaris.

The storage industry has grown to be a roughly $77b/year industry from being a closed, proprietary business. Today's announcements by Sun, represent a small step in the company's journey to openness, but has huge implications to the storage industry...

Posted at 03:10PM Apr 10, 2007 by Nigel Dessau in Sun Storage  |  Comments[1]

The Next Step Towards New Economics for the Storage Industry

Sun made an announcement today around its Storage and Micro-electronic businesses. I really like the idea that we are taking Open Source economics to the last bastions of the closed worlds – storage and micros. Let me focus on what this means to our storage business.

First – we have a new leader. David Yen is head off to manage the micro business (and who better to do so!) and Jon Benson – who we all know well, is going to pick up the mantle for Storage at Sun.

David Yen did a great job focusing the team on the key thing we needed – getting our road-maps clear and viable. While we are sad to lose David, he has left us focused on our products and our value propositions – with road-maps that show us having a long and clear view of where we are going. Jon will be just the right guy to help us drive these through so we can deliver the products.

Second – we've shifted people to help accelerate the adoption of Solaris as the 'storage OS of choice.' See my last entry on the how's and why's.

Lisa Sieker, Aisling MacRunnels and my other peers around Sun are working closely to make sure we remain connected - but the truth is our teams talk daily anyway, so we are not sure we will see much difference. We all live in The Matrix!

But what does this mean to customers:

• Basically we are aligning the Sun organization with changes in the market to increase the speed of adoption of General Purpose Systems and their economics (for our storage portfolio)
  
• We can improve time to market by simplifying development processes through strong collaboration across all Sun development teams, and finally,
  
• We are maintaining a strong focus on Tape, Archive and long term data retention through a dedicated Business Unit run by one of the world's experts in the area.
  

  So next for me – the press calls. Should be interesting to see what the feedback is!
  

  Posted at 01:30PM Mar 27, 2007 by Nigel Dessau in Sun Storage  |  Comments[2]

The Sun Raises on Storage

In my haste to post the news earlier this week, I accidently deleted an earlier post. Due to its overall relevance, I am just reposting...

Over the last couple of weeks I have been visiting customers ... first in New York as part of an event we held for the financial sector and then in New Zealand at our Executive Advisory Committee. On both trips I also got to spend time with other folks from around Sun including our East Coast sales team, executives from services, software and our server group. All the conversations, both internal and external really signaled a shift in the perception (and understanding) of storage at Sun.

The most apparent way to illustrate this is our sales folks that recently went to the American Top Gun Sales Academy enthusiasm about Jonathan Schwartz' presentation on Sun's strategy and where storage plays. Now I don't say that because he's the boss – but because the single biggest challenge for most the team has been working out what the Solaris and Storage Strategy is and how it benefits them...and most importantly their customers. For many, the speech represented a light bulb moment of realization: Solaris is THE long term differentiator for Sun (and Sun Storage).

So, I have been accused of being dense (no voting required), and believe from personal experience that it is a three inoculation process to understand the strategy. On first blush, Solaris seems to be a distraction from the storage business...opening questions such as: is it all about Solaris attach? What about the STK business on mainframes and other platforms? Is it just Solaris inside?

What was made clear to the team at Top Gun, and will be made clearer to all in the next few months is four key points:

1) Sun will leverage the power of Solaris in all our solutions.

Millions of dollars and R&D hours have been spent to make Solaris the best OS, and the investment continues at full speed, making it our competitive weapon (and our customer's too)!

Take for instance our VTL solution. A VTL solution is a system (probably x86), running an OS and an application. All VTL vendors – even those who sell appliances – need those three things. For us at Sun, working with application partners like FalconStor, we use our own systems and our own OS. In our case Solaris...getting the benefit of all that R&D (and rich functionality) for free.

A stand alone storage player (as StorageTek once was) or even some of the other systems companies out there, has to license an OS from someone or use a free Linux distribution. Both are costly and both distract from the main purpose­ producing the best solution for our customers. Today we are working on adding differentiated function to VTL in the shape of VTL-Enterprise. How can we do that? Well we have the storage skills and we don't have to pretend that we are a systems company...we are one! With the emergence of general purpose computing solutions that advantage gets even bigger.

2) Solaris is rapidly becoming the most rich storage OS out there.

Currently we can run all of our storage applications on Solaris...but some things can be done better as an integrated function of the OS. As an example let's think about file systems.

Many of the functions and applications we add to our storage systems are to compensate for what the file system can't do. Typical file systems don't provide RAID, don't provide de-duplication, can't do continuous data protection (CDP) etc. So we've decided to do that with ZFS. While some of this is futures, with ZFS today we offer:

• Enhanced data protection by allowing a ZFS pool to survive the failures of two disks
  
• More reliability and the ability to identify disks that could be used to replace a failed or faulted device in one or more storage pools
  
• More flexible administration and enable the replacement of an existing ZFS file system with a clone of that file system
  
• More data integrity by ensuring that Snapshot data is always from one consistent point in time and enabling users to recursively create snapshots for all descendant file systems

You see the point...the closer proximity to the OS enables the file system and the data protection solution to become faster and more reliable. Less pieces to integrate is better for everyone (especially the customer!). And file systems are just the beginning...

3) We must grow the Storage community using Solaris – and it's free!

Next problem is how we get our customers and partners to use (and improve) all the new integrated functions. Clearly the historical model is to sell you an appliance or solution using Solaris. But what happens if you are not a Sun customer or partner today? How do you get access to it? How do you join the dialog? Until recently – you couldn't.

Now you can go to a website and download all that code and knowledge and use it FREE! What's more, you can participate in Open Solaris and the growing Storage community – again for free.

The other great thing for customers is that they get to choose who's general purpose components, for example x86 server, they want to run that code on. Clearly we would like them to run it on Sun but if you want to try Solaris – run it on your Dell! Run it on your HP! Even run it on your IBM!

But what about running it on your NetApps box? Opps, sorry can't do that one – they don't sell systems they sell appliances only (and at a 70% plus margin).

Clearly this choice is good for the customer – the increased competition means a new economic reality! And we hope it's good for Sun as we expand Solaris users to potential Sun Solaris Storage customers.

4) We make money from serviceability, integration and selling the whole Sun solution

So you download free OpenSolaris and at some point our free NAS stack – do you need to pay Sun any money?

No you don't have to pay Sun any money...it's really free. But (there had to be a but) I have never met a customer who would run anything in their data center without service... Today, 70% of those who have downloaded Solaris have done so on non-Sun platforms. We were getting no revenue from them before so nothing lost if they don't take maintenance.

So why buy an appliance? Well we think we have the expertise to integrate the parts and tune the solution faster than most customers can do it for them selves. We think we can support it better as an appliance – because less options and configurations make easier support. Finally, customers may not want to become storage device specialists – they may want to leave it to us. But some customers might be happy to do all this themselves – in which case we are happy to support them.

Think about it this way – you can consult Emeril's cookbook Emeril's cookbook and buy the ingredients but can you cook like him? Some can – some can't. Many ways to solve the same problem.

So, what will be available from Sun and by when? More news soon (very soon!).

Posted at 02:25PM Mar 20, 2007 by Nigel Dessau in Sun Storage  |  Comments[1]