As enterprises demand more speed and agility from their infrastructures, the performance advantages of flash are critical for addressing the requirements of next-generation businesses. In building an enterprise-ready all-flash solution, we knew our architecture had to solve some of the technical challenges that hadn’t been successfully addressed in the past. This included resiliency, data protection, and scalability.

One of the critical areas we’ve been able to address with a breakthrough solution has been in extending the endurance of flash technology. Endurance is a critical factor in making flash enterprise ready. In fact, one vendor recently released a study in which 77% of respondents said that they believed that the endurance of flash memory is the key to the widespread adoption of SSDs in the enterprise.

How have we been able to achieve new levels of endurance? It all starts, of course, with SPEAR (Scale-Out Performance and Resilience Architecture), which is the foundation for our recently introduced fourth-generation K2 all-flash storage array. As a true scale-out architecture, SPEAR enables us to spread the load very intelligently to maximize flash endurance. Here are some of the keys:

1. SPEAR utilizes a system-wide write cache that eliminates hot spots.

2. By using cache for writes, SPEAR is able to minimize the number of writes and distribute writes across all flash media in the system. The system never writes to the same place twice.

3. SPEAR’s efficient management of metadata eliminates writes to the flash during updates and facilitates space-efficient snapshots that do not incur additional writes.

4. K2 deploys the highest enterprise-grade SSDs with advanced flash management capabilities that optimize the endurance of the flash.

5. Writes in K2 are always sequential to the flash media.

6. SPEAR uses advanced error correction

By building all of these features and functions into SPEAR, Kaminario has been able to prolong the endurance of flash far beyond the spec limits of flash manufacturers, and beyond the limits of almost every competitor building all-flash arrays. When it comes to endurance for enterprise-ready flash arrays, we continue to lead the industry.

Check out our white paper on SPEAR Architecture.

Every flash array vendor out there yammers on about high performance, including us. That’s what we’ve come to expect from flash arrays. SSD flash is quickly becoming mainstream technology at the enterprise level because of its performance and affordability.

The reality is, it’s not all about high performance in I/O. Enterprise storage requires the ability to support a mixture of workloads, with consistent, predictable IOPS, bandwidth and latency.

Let’s look at some real world examples:

• Any Business: It’s the end of the fiscal year. The sales team is hammering their CRM trying to close business and finance is running a massive number of queries to get reporting ready for end of quarter. Which business unit could afford to lose performance? How about… Neither.

• Retail: Sales order entry transactions are constantly competing with accounts receivables and payables entries for performance. Do you want to get paid, or do you want more sales? That’s an easy one! Both.

• Manufacturing: Assembly line transactions are constantly competing with accounts receivables and payables. Slowing down an assembly line can cost millions of dollars, and often times, lost opportunities. Not really an option, right?

The choices are loss of productivity, which ultimately results in loss of profits in the immediate term, and future growth, a.k.a. losing your edge in the marketplace. Neither of these sound good to us, and definitely don’t sound good when considering the impact on your business.

This is why the Taneja Report results were so exciting for us. The testing group showed that doubling the number of K-Blocks doubles I/O and bandwidth performance while keeping latency consistently low (see the below table for specific results). There were no IOPS that impacted overhead in scaling the Kaminario K2 with a read-heavy workload.

The Taneja Report also tested latency during snapshots, showing that latency stayed consistent and predictable when snapshots were executed, with a near zero impact on executing I/O. Plus, the snapshot was completed in less than one second. Multiple loads were running at the same time, yet the impact on latency was also negligible. Even in the event of a hardware failure, performance remains consistent and predictable.

I urge you to consider consistency in performance across workloads as a top tier requirement when looking at storage performance. Considering what you have to lose, consistency really is a no-brainer at this point.

To check out the report, click here.

It is no secret that scalability has long been the elephant in the room for most solid-state storage vendors. The fact that most solid-state vendors can’t scale creates multiple storage silos driving up operational overhead and costs. Scalability and predictive performance are essential to maintaining mission critical storage access, yet few have been able to architect their way to an effective solution. We keep hearing rumblings from the big guys about their scale-out solutions, but no GA yet. Meanwhile, Kaminario has been providing customers with true scale-out architecture since 2010. The fact is Kaminario is the only enterprise scale-out all-flash array vendor on the market today.

In the newly released independent report by Jeff Boles, senior analyst with the Taneja Group, “Kaminario K2: The Truly Enterprise-Ready SSD Storage Array,” the firm evaluated Kaminario’s newest scale-out all-flash storage array (K2), testing performance, scalability, storage capabilities and resiliency in databases and virtualized environments.

The report validated Kaminario’s scale-out architecture and its ability to deliver customers with consistent, unsurpassed performance in any mixed workload. The tests also confirmed K2’s ability to provide ready-for-enterprise features on top of the most sophisticated resiliency approaches on the market. As noted, “…better than linear progression of performance in terms of both IOPS and bandwidth, with 1ms or less of latency”, and following a node failure concluded that, “Kaminario K2 effectively maintains mission critical storage access, making disk, node or subsystem failures inconsequential.”

So how did the Taneja Group think Kaminario faired against everyone else clamoring for a seat at storage performance table? Pretty well it would seem, as the firm cited Kaminario as “a force to be reckoned with” amongst its competitors.

We’ve worked incredibly hard to make K2 the best product on the market today. We knew we created something to be proud of, and are extremely pleased to see that the Taneja Group agrees. We urge you to judge for yourself. To request a copy of the report, click here.

Our team is ecstatic about the results! We hope you will be, too.

Today is an important day, not just for Kaminario but, we believe, for the entire enterprise storage industry. Today is the day we announce the Kaminario fourth-generation K2 all-flash storage array. What is so important about our fourth-generation product? Here are a few key points:

Breakthrough TCO: With K2 v4 we have been able to increase the density by more than 500% while cutting the price in half. Why is this so significant? Lowering the cost of enterprise-ready solid-state flash storage has been one of the long-time goals of the industry. If you believe, as we do, that solid-state is the future of Tier One storage, then it must be cost-efficient for all applications, not just those requiring the highest levels of performance. And now it is. What’s more, Kaminario’s TCO breakthrough extends the viability of solid-state to a new class of mid-size customers who can now afford unprecedented levels of performance, resiliency and ease of use from their storage infrastructures.

Killer Performance: Even before today’s announcement Kaminario delivered by far the best performance in the industry, with world record SPC-1 results for sustained performance and price/performance. With K2 v4 we now have 400% more read/write bandwidth than our previous versions, with consistently low latency (120-microsecond writes). Topping it off, we still offer industry-leading IOPS performance. K2 v4 delivers significant across-the-board performance advantages over any other solution for any combination of workloads, be they OLTP, OLAP or virtualization. It’s a level of performance consistency that is unparalleled in the industry.

Bulletproof Resiliency: Our proven SPEAR Scale-Out Architecture maintains data integrity through any type of failure. For most workloads the performance levels of our solutions will be minimally impacted—usually less than 10%. We even guarantee that degradation will max out at 25%. What’s more, our snapshots are the most efficient in the storage industry, enabling instant restore and recovery from any snapshot with no impact on the performance of the production environment. Snaps can be taken in just milliseconds, 20 times faster than any legacy SAN system on the market today. We make snapshots sexy.

These are major advances in solid-state that have been more than five years in the making. They move storage technology to a whole new level. Don’t just take our word for it: Take a look at the new architectural white paper or better yet, the independent report.

On a personal note, it is an important day for me, as well as for the entire team here at Kaminario. From the beginning we envisioned a new era in enterprise storage driven by solid-state flash and a true scale-out architecture. As of today, that vision is a reality. What’s next? We can’t wait to tell you…

Earlier this year, InfoWorld’s Paul Venezia penned Good-bye – and good riddance – to spinning disk. In it, he laments the continued use of “ancient storage technology” and foresees “a post-storage world” where the industry is “devoid of the painfully outdated yet ubiquitous spinning disk.”

His take fits our view that hard disks are headed the way of the floppy though he says that in some scenarios, HDDs will be used like the way tape is used today. I suppose the alternative is simply recycling.

Venezia says “rethinking centralized storage is a necessary part of this transition [to all SSD].” He is right, but you also have to rethink how you evaluate SSD storage solutions too. Innovation is making hardware dumb and software smart. SSD hardware in the enterprise is fast becoming a commodity and prices are falling in line with that movement. More important than the media itself when comparing your SSD options, software capabilities should trump hardware. Focusing too much on hardware and not enough on the software platform wrapped around it introduces greater risk of quick obsolescence and the inability to adapt to changing requirements.

Additionally, hardware comparisons are nice but not as valuable as how well a system scales, or how well it supports high availability or data protection when it comes to planning for long-term ROI. This is the reason why Kaminario has designed its K2 with cost-effective off-the-shelf servers, enterprise-grade solid-state drive (SSD) media and the SPEAR Storage Operating System. We have learned that it is software that drives consistently high IOPS and throughput, and low latency to deliver 3-D performance that includes all three of these aspects behind fast storage access to applications and databases. We also have built-in to each K2 intelligent high availability and data protection software, called DataProtect, which increases the availability, resiliency and reliability of all of the data stored on every K2 storage array.

Venezia believes that “storage will be so available, cheap, persistent, and indefatigable as to be largely ignored and taken for granted.” If that is the case, then the all-Flash SSD era will fulfill general-purpose storage needs rendering storage to be like a utility provisioned with ease thanks to seamless storage management software.

While I understand that the transition to all-Flash in the enterprise data center is not happening overnight, I raise my glass to Veneezia’s call for the end of spinning disks. Their time has come to fade into the night and make way for the next generation of storage innovations. At Kaminario when we talk with our customers and partners about this transition we end with the simple motto that describes the direction that we’ve taken as “Storage Done Better.”

(photo credit)

An interesting Gizmodo article entitled Hot Flash: The Secret To SSDs That Will Outlast The Universe discusses a new discovery claimed by Taiwanese Flash manufacturer Macronix that by periodically applying intense heat to MLC Flash for milliseconds they can increase its life from 10,000 write cycles to more than 100 million. How does it work? Flash writes erode transistor insulation slowly until eventually the chip is no longer viable. Add a quick jolt of 800 degree-centigrade heat and you can actually heal the insulation for another 5 to 10,000 writes.

No doubt this is an important discovery, but the truth is that the life limitations of flash are greatly exaggerated even today. Flash manufacturers and solution vendors have come up with a raft of technologies, including wear leveling, advanced error correction, and sophisticated digital signal processing, to ensure your Flash device can last almost as long or longer than a typical hard disk.

Kaminario builds on the life extending technologies offered by flash manufacturers with its advanced SPEAR operating system. SPEAR distributes writes across the K2’s Flash blocks and across entire Flash nodes, ensuring that no single Flash block or Flash card will be rendered obsolete by excessive writes. Not only does this lengthen Flash life, it also ensures superior flash scalability and performance.

The K2’s DataProtect feature takes further pressure off of production Flash by backing up data to less expensive secondary Flash or Hard disks, ensuring that data protection has no impact on Flash wear. Finally, the K2’s advanced snapshot technology is designed to minimize the use of Flash writes by snapshots.

Flash may one day outlast the universe. What we can appreciate today, however is that Flash is a fast, durable technology that can boldly take your performance-dependent applications where they have never gone before—at warp speed.

In 2013, MLC Flash as general purpose storage will be a major area of industry attention because users are demanding many of the same features they are used to getting with HDD technology but with the added capabilities of MLC Flash. Customers must have scalability and data protection as well as performance to make the case to jump to SSD.

Focus on Scalability
Scalability is a cornerstone element of a general purpose storage strategy and a solid differentiator for Kaminario. However, we view it a little differently than other vendors:

Linear Scalability: Grow Capacity Not Complexity
There is a difference between adding capacity and scalability. Adding capacity is stacking storage hardware while scalability is intelligently absorbing storage. We view scalability as linear and driven by powerful software like SPEAR to enable distributed workloads managed from a single console. Linear scalability increases operational efficiencies and ease-of-use. Simply adding capacity means multiple management consoles, more complexity and more points-of-failure.

Linear Scalability Enables Consistent Performance of Mixed Workloads
Our view of true scalability provides users with consistent latency as IOPS and bandwidth are scaled up. Users should not have to pay a performance tax for consolidating applications or deploying data protection features. There should be no performance surprises.

So, for example, you could be running an OLTP and an ERP solution then add an analytics database. Linear scalability using SPEAR means that your latency should remain constant as you scale up your IOPS and bandwidth. There is not much point in high storage utilization if the quality of the performance goes down. Two years ago, EMC sponsored a group that put 26 people inside a single mini cooper automobile. That’s high utilization, but how well does that car perform on the road? How can the driver see anything?

When you evaluate SSD offerings, look closely at how scalability is really defined and what effects on performance it really has.

What this Means?
Organizations are embracing SSD storage more and more to tackle their information management challenges. It continues to be true that big performance is not enough to meet these requirements. Concurrently, the competition among vendors serving these customers is heating up. Vendors are staking their ground regarding how they want to position themselves. Some are tying their fortunes to cloud arrays. Others may focus on specific applications like VDI. The K2 is general purpose on purpose. What the others do not have is SPEAR, a true, enterprise-class Scale-out architecture that includes the intelligence and flexibility to offer advanced data protection and linear scalability features in addition to low latency and fast read and write speeds.

The full promise of SSD benefits lies with general purpose storage. That is what the market is demanding and will be the next giant leap forward for SSD based storage.

(photo credit)

Tom Coughlin and Jim Handy, respected storage analysts, conducted a survey on how many IOPS are really needed for several popular enterprise applications to better understand real-world IOPS requirements. They asked respondents about their critical storage applications and how many IOPS those applications needed. More than 80 percent of their initial respondents said that their key app was a database, an OLTP solution or a Cloud storage service. About half of the respondents said their IOPS needs were in the 1k to 100k range. Coughlin and Handy note that other factors like latency contribute to overall performance, not just speed. You can view research details from a presentation that Tom did at the SNIA Storage Developer conference, How Many IOPS is Enough? or get the full report at How Many IOPS Do You Really Need?

Measure Consistent Performance
I am bringing this up now as a way to comment on how Kaminario looks at performance. We do not see performance as a point in time such as timing a runner in a single forty-yard dash. True performance is measured by consistency of speed and results over time and as capacity and applications are added. You shouldn’t necessarily buy an all-Flash array based of the performance of a single app because there are so many variables that can come into play down the road. This is one of the reasons that Kaminario is taking the general purpose storage approach with the K2. We think it is smarter to evaluate a Flash storage system in the context of your entire storage environment and future app needs. You don’t want to buy a system that gives you great value up front but falls short in mixed workload environments. You don’t want to have to buy a separate storage system when you introduce a new application or need to drastically scale up capacity.

3D Performance: Latency, IOPS and Throughput
When Kaminario designs its K2 All-Flash SAN Arrays, we view performance through a lens we call “3D” performance. You get the best results when you have the best combination of ultra-high throughput, super low latency and a large number of IOPS. So if we put together a similar survey, we’d definitely want to account for all three of those performance dimensions. And if possible, we’d want to take high availability features into account too. Real-world environments often have mixed workloads that have different needs, e.g. OLTP applications are very latency sensitive, analytic applications are both throughput and latency sensitive, and something like VDI is more IOPS sensitive.

More Research Needed
What I saw from Coughlin and Handy’s initial results affirms these distinctive K2 capabilities:

• You get 3-D performance – high throughput, low latency, more IOPS
• You get smarter management and cost-efficiency that are perfect for general purpose storage
• You get the most linear scalability for performance and capacity on the market
• You don’t have to sacrifice data protection to get ultra-fast performance

I applaud them for taking on this interesting research to further understanding about storage performance especially when it confirms enterprises’ requirements for 3D storage performance.

(photo credit)

A November 30 Fusion-io blog entry, Cisco and Fusion-io Tackle Big Data with Oracle NoSQL, highlights blazing NoSQL big data performance achieved by a Cisco Unified Computing System (UCS) configured with server-based Fusion ioDrive2  SSD. The blog and linked Cisco Solution Brief discuss the importance of near-real-time performance when running operations on user profile data in an e-commerce transaction scenario. Fast performance is key to e-commerce customer satisfaction and can be difficult to achieve with the widely fluctuating workloads typical of a busy e-commerce site.

Fusion-io and Cisco make a compelling argument for SSD in near-real-time big data scenarios and for server-based SSD in particular. However, it’s important to remember that for business critical big data applications, you need scalability and bulletproof reliability as well. There’s another solution that is both more reliable and more efficient than server-based SSD: SSD arrays such as the Kaminario K2.  Here are some reasons why you should consider an SSD array, and, specifically, a Kaminario K2 with its Scale-out Performance Storage Architecture (SPEAR), for your near-real-time big data needs.

Efficiency By nature, server-based SSD cannot be shared across a single server, which means you can find yourself in situations where one server-based SSD card is maxed out running data operations, with the resulting performance hit, while another server card sits underutilized, without any easy way to balance the I/O operations load. By nature, an SSD array is shared, which means you have a giant super-fast storage pool that can be sliced and diced across CPU’s, databases, and big data applications at will. Not only do you get more consistent performance, you have a more cost efficient storage solution as well.

Scalability A server array architected as a scalable clustered SAN, such as the Kaminario K2, gives you a shared storage pool that can be scaled to hundreds of terabytes in a single array, and even more as you add arrays, which, in the case of the K2, can be managed together as a single unit. In a server- based solution you can only scale to the limit of each server’s PCI bus. And the best thing about the K2’s SPEAR design is that you scale performance as you scale SSD, so you never take the performance hit that can come with SSD on a shared bus.

Availability and Data Protection Server-based cards can end up being single points of failure that  slow down performance and possibly even cause an outage if one of them drops out for any reason. The Kaminario K2’s highly fault tolerant SPEAR design continues delivering super-fast performance across your shared SSD storage even if one or more cards fail. For data protection, the K2’s DataProtect provides super-fast high-volume snapshot and data replication capabilities that protect your data with no impact on overall performance, a claim that can’t be made with server based SSD solutions and even with the overwhelming majority of array based SSD solutions available today. Check here for more information on how the K2’s high availability and data protection features work.

So with a shared Kaminario SSD array, you get the best of both worlds: super-fast performance with airtight reliability. As they told you in kindergarten, sharing is better.

2012 was the year SSD started moving from niche status into the enterprise mainstream, building up the seasoning and enterprise-class features it needs to take off in the years ahead. It’s obvious that enterprise IT is taking notice, feeling more comfortable with the technology, and becoming more aware of its benefits and best uses.

Here are some of the trend highlights we noticed over the past year.

Sales continued to grow – SSD sales continue their upward trajectory. According to IDC, the market for enterprise SSDs will continue to grow to $5.5 billion in 2015. SSD shipments in general reached 12.9 million units in the first half of 2012 and are expected to reach 28 million in the second half, according to HIS iSuppli Memory and Storage Service.

SSD continues to close the price gap with HDD – Flash pricing continues to become more competitive with hard disk storage, and the gap is narrowing. SSD prices continued to fall this year, with the bare media cost falling well below $3 per gigabyte in 2012, compared with almost $9 per gigabyte in 2010. Adding to the lower cost are the cooling and real estate advantages of an SSD array compared with hard disk storage.

MLC takes center stage – Enterprise SSD vendors are settling on MLC as the flash media of choice for the enterprise. MLC has a shorter life than the considerably more expensive eMLC and SLC flash, but advances in wear leveling, digital signal processing, ECC, and a variety of other endurance-enhancing technologies have enabled vendors to offer less-expensive MLC-based SSD solutions with a life span that approaches that of the average hard disk. Meanwhile, array vendors have started to add their own enhancements, such as Kaminario’s SPEAR in its K2 storage systems, which distributes reads across flash blocks and data notes, extending the life of each individual block.

IT gets it – IT always needs some time to get up to speed with a new, up-and-coming technology. It seems as though IT is fully embracing SSDs. As the technology has matured and IT has moved past the fear and toe-in-the-water testing phases, it’s clear that those in IT have become comfortable with and excited about SSD’s potential. A recent IDC study found that 72 percent of IT managers surveyed said their organizations either use or plan to use SSD in their data centers within 12 months. Similarly, an InfoPro survey indicated the growth rate for SSDs would reach 36 percent this year, mostly for I/O-intensive databases, ERP, and CRM applications. Several large companies have made SSD an essential part of their general-purpose storage infrastructure.

SSDs alone are not enough – The inclusion of SSD technology in a legacy storage system that was built for HDD technology has proven to not be enough to take advantage of the higher performing media. Many of the larger vendors in this market are working with products whose storage operating systems were built 20-25 years ago and were not designed for all-SSD, performance, but rather for slower HDDs. Innovators in SSD storage architectures, such as Kaminario, developed new modern storage operating systems from the ground-up specifically for SSD performance.  As a result, all features of these systems work along with the speed of SSD and are not “slowed down” to work at HDD performance rates.

Scale-out is the way to go – Scale-out architectures that can grow along with your companies’ information processing requirements is what is required in dynamic enterprises, rather than having to do a rip and replace or forklift upgrade to the storage system as needs change. The performance of SSDs in a scale-out architecture is also more complementary with faster CPUs and networking interfaces, so that the system is streamlined for higher performance from end-to-end, rather than having a mismatch of underutilized high performance systems and networks waiting for slower storage systems to finish processing.

New features are worthy of an enterprise contender – SSD has moved past its reputation for nuthin’ but raw speed, adding data protection and general-purpose features that make it a true enterprise storage contender. Kaminario has led the way by being the first to add enterprise-class RAID capabilities, lightning-fast snapshots, and replication to its K2 solutions with the unique performance and architectural characteristics of SSD in mind.

Enter the big boys – The storage big boys have taken notice of SSD’s maturity and acceptance and have started to make their moves. There were several big acquisitions this year, including EMC’s acquisition of XtremIO and IBM’s announcement that it would acquire Texas Memory Systems, not to mention Apple’s acquisition of Anobit and SanDisk’s gobbling up of FlashSoft.

Hybrid hokum appears – Several major legacy storage vendors, including NetApp, EMC, and IBM, have also started offering hybrid solutions that supposedly offer the best of both worlds: SSD performance with the seasoned, enterprise-class storage management features the major players have offered for years. As previously mentioned when talking about how ‘SSDs alone are not enough’, these solutions often suffer from the limitations of legacy storage architectures with controllers and management features designed with slow hard disk storage in mind. (See Hybrid Storage: Best of Both Worlds? Not!)

All in all, things are moving along very nicely, with enterprise SSD building up the head of steam it needs for its eventual takeoff. I won’t say 2013 will be “the year of SSD,” because everyone else will and that will probably jinx it, but I won’t be surprised if it happens.