We’re proud to see one of our customers “going big” for the Super Bowl. Shazam has announced that it will have Shazam enabled content for several ads during The Game. You can read more about it in the New York Times.
http://mediadecoder.blogs.nytimes.com/2012/02/02/shazam-ties-into-the-sounds-of-super-bowl-spots/.
I have been recently reading about Amazon’s Storage Gateway:
The product is effectively an iSCSI target that has the Amazon S3 client on its back, to backup iSCSI LUNs to the local Amazon S3 storage. This arrangement has a number of immediate problems.
In general, disks are used either for file storage or for databases. In either case, there’s implicit or explicit transactional semantics that must be preserved when backing up storage. For raw files, this often is an open-to-close semantics whereby an open-for-write file is considered not to be “consistent” or even “visible” to other users until it is actually closed. For databases, update transactions have of course exact definition and scope.
Problem is, on the block target (low) level – there is no context and no information of those higher level transactions. It is, therefore, easy to imagine that Cloud Storage on the backend is being busily populated with LUN images that are internally inconsistent and cannot be used to restore the data.
When I said “being busily populated” above – I meant it. The other problem with AWS Storage Gateway can be described this way: disks are generally never idling – applications quite often are.
Operating systems and other applications that use block storage generate a lot of temporary and transitory content. File content and databases gets re-indexed, software reinstalled or upgraded, YouTube videos and other Internet junk uploaded into temporary folders—and so on.
However, at the block storage level, the information on the relative value of all this activity is totally missing. The only thing that a block target “sees” is plenty of new blocks that need to be snapshot-ed and backed up. Which it then is (yes, busily) executing. Garbage in, garbage out.
Finally, there is an iSCSI connection. I love iSCSI with all my heart but – the latency! The protocol is not very famous for its latency, let’s put it this way. And so, as a user, to take advantage local Cloud backup in the Amazon implementation, I’d now have to go iSCSI. One word of advice then – stress test it first, really well.
Today, we announced funding led by Menlo Ventures. Razor’s Edge and Sierra Ventures both participated, as did our existing investors.
A few quick points:
1. A hearty welcome to Pravin from Menlo and Tim from Sierra Ventures. I couldn’t be more proud that Menlo Ventures decided to lead our investment round. To be blunt – they had significantly better references than some of the other lead investors that we were considering. That sealed the deal for me.
2. The old cliché that the best time to raise money is when you do not need the money is simply true. Nexenta grew by 400% last year, largely funded by our own sales. We could have decided to not raise money, which made raising money easier.
3. So why did we raise a growth round? You don’t get to choose when these markets transition. And storage appears to be transitioning now. We named a few of our customers in the funding release. There are others with bigger names and tighter PR policies 
We are going to be using the funding to improve NexentaStor, scale our support, and further accelerate our innovation.
4. Thank you. If you are reading this, then you have something to do with Nexenta’s success. It is due to the strong pull from partners, customers, end users, and technical partners like Dell, Intel, LSI, and others. And that the lock-in-based business models of legacy storage are giving way to OpenStorage and our mission of “enterprise class storage for everyone”.
Evan
Another year end is upon us, and this always is a good time to look back on the past year, as well as look head to the year to come.
Here are my predictions for 2011, along with my score of how I did with my predictions from last year.
In April of 2011, the Register wrote an article suggesting that we may be the fastest growing storage company ever. Since then, our growth has accelerated.
What’s more is that companies from VMware to Cisco and Dell now are more actively supporting NexentaStor in the field.
You can read the 451 Group’s eye opening report on the momentum on display at the recent OpenStorage Summithere.
DR for the masses is starting thanks in large part to companies like Hosting.com that allow you to replicate your VMware VMs to their service. VDI, see below, needs to happen for DR to really proliferate. However it is coming.
Meanwhile, though, clouds themselves came under scrutiny in 2011; it turns out that Amazon and other clouds are not quite as good as a top enterprise IT organization at keeping the data safe. DR for the masses is on the way, but not here yet. With that in mind, I’ll give myself a C.
Certainly Russia, India, and China have shifted towards OpenStorage solutions. I project that our market share in India will shortly be in the top 5, well beyond our levels in North America. We’re making investments in China as well, including our recently announced plans to partner with Cloud Valley.
On the whole, though, I’d give myself a B as we could have done even more in 2011. My prediction is that this will be more of a story for 2012 and 2013.
It turns out that there are thousands of server resellers that are capable of selling servers, virtualization, storage, and more to create solid, total solutions for their customers. Storage is an expensive part of that equation and, therefore, an open solution that runs on their hardware can be quite attractive to these VARs. I predict that we will see more and more of this going forward. The velocity of the server channel selling storage will increase in 2012.
Today, VDI remains a niche application. If you already use XenApp to access major enterprise applications remotely, then XenDesktop is a great fit, especially as Citrix will make it worth your while. However, while we have major hosting providers like VESK, and health care customers like Delano Medical as customers, we have not seen a take-off in VDI deployments. The top reason, ironically enough, is storage.
All this changes in 2012. I predict that NexentaVDI fundamentally fixes the value proposition of VDI in a VMware environment.
Which is great; however, VDI in 2011 under-performed. So I’ll give myself a C on this one as well.
Microsoft’s Steve Balmer had it right when he shouted “Developers, Developer, Developer, Developers. (Seehttp://youtu.be/t6ZarKIKpSA for a laugh). Balmer, Gates, and Co. realized that the only way to have a successful platform is to have apps on it. The only way to get apps on your platform is for developers to write them. Ergo, Developers = Success. 2011 was a year of burgeoning platforms, from EC2 to companies like EngineYard and Heroku. There still is a lot of competition for winning the hearts and minds of developers, and I predict we’ll see some convergence in 2012 as the number of successful programming models consolidates.
I’ll give myself an A for this one.
Silent data correction, or bit rot, is one of the under reported problems of the storage industry. The disk manufacturers have done a great job at building cheap disks that are reliable. But as the amount of storage grows, if failure rates are constant as a function of storage size, then failures will go up. I predict we’ll start reading stories in the press about massive efforts to recover data lost to silent corruption.
While we had high profile cloud failures from Amazon and others, I did not see the press become aware of the fundamental importance of data corruption that passes unchecked through most of today’s storage systems.. I’ll give myself a C on this as I know that it is increasing in importance and I cannot control the press.
We’ll have more to say on this in the near future.
In a response to an article about NexentaStor being used as one of the storage solutions powering the incredibly intensive Hands on Lab at VMworld, EMC confirmed that they run on commodity hardware as well:
“Our hardware is based on commodity x86 as well (on all EMC platforms)”
http://www.theregister.co.uk/2011/09/22/vmworld_hol_sakac/
This quotation is a fundamental tipping point in the storage industry. For the first time, the premier priced solution is admitting that they are selling commodity gear and massive mark-ups. Are those mark-ups due to their software or are they due to their lock-in based business models that capture customers through a proprietary approach and then extract as much as possible from them through a high touch sales force with “flexible” pricing.?
Also, this year in my keynote at the OpenStorage Summit, I predicted that we will launch a storage system based on Intel’s Romley chipset next year that will achieve 1 million IOPS in a 2U form factor. I stand firm behind this prediction. And I also predict this will further alter the storage landscape.
So, I’ll give myself an A++ for this prediction. The commoditization of hardware is a fundamental trend in storage and, not surprisingly, not one that other storage prediction blogs tend to point out, perhaps because it destroys the premise of today’s storage industry.
Despite our best efforts to introduce pricing transparency while eliminating vendor lock-in, and the force of a troubled world economy, the storage industry continued to capture more and more wallet share , with Gartner and Goldman Sachs and other analysts indicating as much as a 12% increase in growth.
Recent reports from Gartner, Forester, IDC, and others have confirmed that the amount of data being stored is growing exponentially.
However, often overlooked is the observation that humans would store even more information and, thereby, be even better at solving problems, were it not for the legacy storage industry. To take one example, the New York Timesrecently reported that in 2012 it will cost approximately $1,000 to sequence the human genome; it will cost more than that to store each sequenced genome.
In short, as Jason Hoffman, CTO of Joyent, put it in his keynote at the OpenStorage Summit [http://www.nexenta.com/corp/newsflashes/86-2010/938-openstorage-summit-2011], there are a class of problems that cannot be solved with today’s storage industry.
Capacity deployed is increasing exponentially. However the true underlying demand is growing even more quickly. I’d give myself an A+ for this one.
All in all, 2011 was a great year. OpenStorage continues to gather momentum, bringing us ever closer to making our mission of “Enterprise class storage for everyone” a reality.
Happy holidays and Happy New Year to all.
Most readers of this blog already know that last week’s OSS11 was a tremendous success. Here’s a couple of things that jumped out at me at OSS11:
1. Be careful what you wish for. A late spike in registrations made us pretty nervous. 100 more people registered than we officially had room for. The rooms were crowded but not unsafe. Many of us spent the show in meetings and on the small show floor instead of in the sessions which helped make room for other attendees.
2. Distance isn’t dead. I love IT and the internet and free communications. We rely on it heavily. However, these shows / summits thrive because there is nothing as efficient as getting hundreds of people you ‘d like to chat with in one place. Nexenta team work in particular seemed to thrive at the show. With that in mind, we’re going to pull all Nexentians that do not need a visa into one place in January; those that are grounded thanks to visa complexities, I’m going to go visit.
3. The world isn’t flat. The world’s a lot smaller than it was, sure. However, our vision is to be globally local. That’s part of our secret sauce. There are real challenges, though, to being global: first, we have to recognize that we’re different. Americans are a minority of our employees and of our customers. Secondly, we have to be able to shift quickly for example from chatting with another American to discussing partnerships with our most important virtualization partner in Japan (they attended OSS11); at least for me, that shift is a lot of fun but also intensely challenging. Third, we have to understand that we are missing as much as we are receiving when we partner with customers and resellers around the world; humility does not come easily for many US start-ups or just many Americans for that matter, and yet IMHO being humble is a fundamental success factor to growing globally. And fourth, we have to hire locally; language skills are not enough, we require team members that can automatically pick up the high bandwidth signals that even Americans with lots of experience outside of the United States will miss.
4. Legacy storage cannot keep up. The legacy storage providers are, by all accounts, well run, high performing companies. So how is it that we’re able to out innovate them? I think it is as simple as the weight of their business models. By focusing only on software we’re able to sprint ahead into new, related use cases. We’ve just announced a new companion product, called NexentaVDI, that none of the legacy vendors could deliver in part because it is a software-only solution; it runs on ESX to enable massively faster and better optimized VDI desktop deployments. We’ll be sharing more news about this product soon, in the meantime you can see the announcement here. In addition to better addressing use cases with new software, we’re also sprinting ahead with support for tomorrow’s hardware. And tomorrow’s storage hardware will revolutionize the storage business. At OSS11 I made the prediction that next year we’ll see a Nexenta partner selling a solution pulling 1,000,000 IOPS and that we will see Nexenta partners selling 2PB racks of storage. I’ll bet anyone $1,000 that the 2PB rack will be a reality; you can buy a fully certified 1PB rack from Aberdeen today (please do: click here :)). And we have systems in the labs now that are capable of pulling 1 million IOPS. Please note that customers that buy NexentaStor will be able to upgrade their hardware to be able to achieve that performance simply by moving over their existing licenses.
What did I miss? What did you like best about OSS11? What didn’t work as well as you’d like (other than the bane of all conferences, hotel wi-fi)?
Anyone want to bet me $1,000? I’d be happy to pay it just to get some good comments to this blog!
Recently I reread some old blog posts in order, in part, to gain some inspiration for a new post. And it worked.
Here’s a post I wrote in June of 2010 about events at Nexenta:
What’s consistent since that time:
What’s new
I believe my job includes being able to be both our harshest critic and our biggest cheerleader. We’ve not done everything right. We continue to make mistakes. And we will make more. However – our pledge helps us to relentlessly focus on four things: customer success, VAR success, an aggressive and transparent culture, and, last but not least, fun.
All in all, people like me and other Nexentians live in part for what we are achieving here. To grow a company from a good idea in IRC chat to perhaps the leading independent competitor to the major legacy storage vendors in approximately four years is a (stressful, all consuming, out and out difficult) blast. For the right people, those that can seize opportunities while building sustainable processes – this is an opportunity of a life time.
And yes – we’re hiring pretty much across the board. We have 100+ recs that are open or will be shortly. Please give it your best shot. Email careers at nexenta.com to learn more. We’ll be launching a new careers site shortly as well, as a part of our spruced-up website.
Good news for fans of OpenStorage! While Gluster has never achieved the take off that NexentaStor and ZFS have, the brilliant and hardworking team there deserves a good outcome.
I’d like to welcome RedHat back to the enterprise open storage market. Their support of open approaches to storage can only accelerate the ongoing move of storage away from today’s dominant legacy, proprietary technologies and business models and towards the open future.
We remain committed to helping to lead that transition. To the extent that RedHat is becoming a competitor – may be the best product win!
Also, today Nexenta announced that NexentaStor is *again* powering the massive work load of VMware’s cloud hosted Hands on Lab. No other vendor other than EMC and NetApp have been invited to serve this massive load. We think this honor shows how OpenStorage and NexentaStor have earned their place at the table through superior performance and resiliency. You can read the press release here: http://www.businesswire.com/news/home/20111004005911/en
And you can see some slides providing detail on our success powering this VMware cloud at the Las Vegas show here: http://www.slideshare.net/NexentaWebinarSeries/nexenta-at-vmworld-handson-lab
As many commentators have pointed out, VMworld is becoming the critical show for the IT industry. Ironically, it occurred during Dreamforce which might be the other fastest growing show.
I think Nexentians really enjoyed our VMworld experience. The attendees at the show are generally fairly technical. Ask them about whether storage is too expensive on a per I/O basis, too difficult to use, or just too unsafe and they get it – immediately. Plus they ask all sorts of smart questions. Some of my favorite questions I heard when stopping by the booth included:
- Q: “What VAAI primatives do you support?” A: we support all four, however currently we only support them for iSCSI. NFS support is planned as well.
- Q: “Do you recommend the use of iSCSI or NFS for virtualization?” A: we tend to answer this by saying either works. I personally prefer NFS as it simplifies management and I think our global namespace cluster and NFS are hard or even impossible to beat. However – KT runs us with iSCSI and, as our 3.1 release indicates, we continue to invest in iSCSI and block performance.
- Q: “What is the performance you are getting when running the Hands on Lab?” A: Look here. We were running a real time feed of performance that we call the Aura interface.
Aura will be making its way into NexentaStor soon. It leverages the amazing data provided by DTrace. One reason we were not too concerned about showing this pre-released feature to our major competitors by showing it at VMworld is that no other major storage company has the level of performance granularity DTrace gives us. They may be able to emulate / rip-off the cool interface, but to match the quality of the information will require cracking open their kernel.
Somehow I think legacy storage has higher priorities, like fully supporting multi-core processors and the upcoming Sandy Bridge. Or building in-line dedupe. Or adjusting to the post hardware RAID world by delivering air tight data integrity and 256-bit checksums. Or, for that matter, offering infinite linked clones – great for VMware. All of these changes require changes to the fundamental architecture of their kernel and file systems, which is why it is proving so hard for legacy storage to catch up to us. They’re 32bit and we’re 128bit and that’s the way it is going to stay for a long time
In any case, here is the Aura interface that helped so many attendees understand the storage performance required by Hands on Lab:
Starting in the lower left and moving clockwise, this interface shows:
- Most frequently accessed files. In the case of the Hands on Lab, this showed which VMDKs were being accessed the most.
- Backend IOPS and then NFS operations. Pretty self explanatory. Yes, what data is presented is configurable.
- The circular area: the outer ring shows reads in blue and writes in green from each ESX server running the labs. The chords attaching these servers to NexentaStor, which is in orange, show the bandwidth utilized. The darker the color the greater the bandwidth.
The totals for the show are quite impressive – well over 2 billion operations were performed by the four NexentaStor instances running. The peak IOPS per system were several hundred thousand IOPS. Over 200,000 VMs were created and destroyed during the show.
And, here’s the kicker, NexentaStor was running on industry standard hardware from Supermicro with STEC drives for write and read cache and 7200 rpm SAS drives for capacity. Monday some DRAM on one of the four servers (two HA pairs) failed. And no end users noticed because of our HA cluster performed correctly and failed over. Meanwhile our load increased from a designed 33% to over 60% of the total load of the Hands on Lab due to unspecified issues with either NetApp or EMC. These vendors shifted to very heavy DRAM and flash based systems, and one vendor resorted to an a several million dollar DRAM and flash based array to keep up with the performance load. We estimate that our implementation was over 90% less expensive than the other solutions running and yet it was at least as fast and we believe it was the only system not to miss a beat as the load increased dramatically during the course of the show.
This blog is getting pretty long. So I’ll just mention a few other highlights of the show:
1. VDI in a box: Racktop Systems was demonstrating their VDI in a box solution, which they call the vBOX. This kind of system is exactly what the market needs IMHO. We have lots of happy VDI customers and are fast becoming the VDI solution of choice for both Citrix and VMware deployments.
Typically buyers and analysts focus on NexentaStor as providing IOPS and enterprise class reliability and data protection without the absurd pricing of legacy storage. That’s true. The other benefit, though, that various VMware S/Es and field engineers are telling us is the ability to scale the system as the VDI deployment progresses. This ability to scale an existing deployment is radically different than the legacy model, where if you start with a Chevrolet you simply cannot scale it up in terms of performance to the GMC loads required; to strain the analogy a little further, you have to start with a massive GMC if you believe you’ll eventually need the load. By comparison, you can start with a vBOX from Racktop with 20,000 IOPS and scale that to 100,000+ IOPS and then join that together into a massive name space cluster and so forth. This means you are not risking a multi-million dollar storage expenditure day one of the VDI deployment.
Racktop has shown that they understand the VDI use case extremely well. By including VMware View and NexentaStor in one box Racktop delivers a turn-key solution that can handle the random I/O and unique I/O patterns of VDI.
2. VDI at the Intel booth: We had a great experience in the Intel booth, demonstrating breakthrough cost-performance of VDI as measured by the Login VSI benchmarking solution. Many thanks to Marco Righini from Intel who is one of the industry’s leading minds in how to implement and operate VDI systems; plus Marco is a great pleasure to work with. The Intel booth, not surprisingly, was in a great location right across from VMware.
3. One PB for $475,000 list price with HA: Aberdeen is a well regarded VMware expert with a long history in storage and with NexentaStor in particular. They chose a very direct way to make a splash at VMworld. They ran, in their booth, a full 1PB of enterprise class storage, with HA, full replication, VAAI integration, and the opportunity for SSD based cache and they put a price on it. Only in storage would the fact that they’ll actually price it out and stick by that price make news – but before Nexenta broke onto the scene as the leader of OpenStorage such behavior was unheard of! Ask your legacy vendor to tell you right now how much 1PB would cost. Their answer will be: “it depends.” What does it depend on? The price you pays depends on a complicated algorithm that includes your negotiating ability and their need for a particularly high margin deal at that point in time.
Back to the headline pricing – $475k for 1PB works out to $475 per TB, or approximately 93% less than the list price of legacy vendors. And yet Aberdeen’s solution was based on NexentaStor which, as discussed above, ran VMworld’s Hands on Lab and proved itself to again be able to perform as well as legacy storage while maintaining at least as high a level of availability.
So those are a few highlights of the show at least as I see it from a very Nexenta-centric perspective. What did I miss? Let us know your thoughts. See you in Copenhagen. And Nexenta will be at various shows including this week’sPhoenix VMug, along with our excellent partners Nordisk. Please stop by.
If you stop by booth 658 you can see an unprecedented view in an unprecedented work-load for virtualization, streaming live. We are showing a flexible view into IOPS, throughput, latency and other metrics generated by NexentaStor as it serves up over 5,000 simultaneous multiple VM labs as a part of this year’s hands on lab.
I wrote a little bit about the breakthrough IOPS required as a part of this work load in a recent blog.
In this blog I’d like to share with you another view into VMworld -> the Nexenta Partner Map.
As you can see, many, many partners including Intel (where Nexenta is powering high performance VDI with Login-VSI tests available), VMware, Avere, Aberdeen and others are discussing NexentaStor and open storage with their visitors. And, if you get evidence that you visited each booth, you can win an iPad
Next week NexentaStor will be making history at VMworld. Stay tuned or stop by our booth 658 for more information. Or stop by Intel’s booth, where we will be the only storage solution featured. Or stop by Aberdeen’s booth 760 to see 1PB on the floor with active / active failover and VAAI. Or stop by any one of another 3-4 partners and end users at the show to learn more.
What if you could get superior performance (including random IOPS), air tight data integrity, tight VMware integration, the industry’s best asynch replication, in-line deduplication and compression, and have it all running on industry standard hardware at a small fraction of the cost and without the vendor lock-in of legacy storage? What if you could get enterprise class capabilities without the cost and aging technologies of legacy storage solutions? Well, you can, which is why Nexenta is achieving record growth in terms of paid deployments. See a Register article on the data on our growth from this spring.
At VMworld we’ll answer the questions: “what if the system could pull as much as 1 million IOPS? How about 3gigabytes per second?” (results can vary)
I think the answer will be that countless stalled virtualization efforts, that froze due to high costs of IOPS, can now move forward. If you can spend less than 40k to get well over 100,000 IOPS, and save 85% versus alternatives to get as much as 1 million IOPS then $/IOPS are no longer the problem. The result should be faster adoption. And more load than you’ve ever seen on your network
In IT, we always play pass the bottleneck. Starting next week the bottleneck is firmly off the storage, if you are using a NexentaStor based solution. VMware will prove that we can saturate multiple 10G links with, again, with what we believe will be industry leading IOPS and as much as 3gigabytes of throughput.
Stay tuned. The proof will emerge next week that I/O, one of the fundamental bottlenecks of today’s IT infrastructures, has been solved. Contact marketing at nexenta if you would like to schedule a briefing at the show with one of our experts.
Please note that I/O alone isn’t the solution. The solution is unlimited I/O *plus* enterprise class capabilities. Sorry SSD guys – just I/O alone or I/O plus a file system you developed in your garage (or, more typically, you compiled from Linux), does not make your technology a solution to the virtualization bottleneck. Fast, but unsafe, is just as much a non-starter as slow but unsafe. In data storage if you cannot store the data you are not really in the market. Any start-up with a home grown file system has got 7+ years of product deployments to go before the file system matures to the point that major users will touch it. NexentaStor fixes the I/O bottleneck while improving your ability to protect your data. It is that combination that is needed to move the industry forward.
© 2008 - 2012 Nexenta Systems, Inc.
All Rights Reserved
