X-IO Technologies today expanded its iglu blaze fx enterprise platform with an all-flash version that scales from 4.3 TB to 466 TB in an array.
The iglu 800 series uses all Toshiba enterprise MLC solid-state drives (SSDs) and includes X-IO software features such as a new stretched cluster technology with synchronous mirroring, snapshots, replication, and data-at-rest encryption.
The vendor claims the 800 can deliver 600,000 IOPS with 366 TB of flash capacity. The stretch clustering provides disaster recovery for data centers up to 100 kilometers (6.2 miles) apart. Stretch clustering can be used with any iglu storage system.
Customers can federate 32 pairs of iglu controllers.
For the all-flash version, X-IO upgraded iglu controllers with double the number of CPUs and cores and up to three times as much memory of its hybrid arrays. The 800 series includes Intel Xeon E5-2680 v3, 12 cores, and 64GB of memory.
The iglu 800 all-flash arrays start at around $120,000 for 28 TB of capacity.
One feature missing is data deduplication, which has become popular in all-flash arrays because it expands the amount of effective data that can be stored on a system and reduces price per GB. Ellen Rome, X-IO’s vice president of marketing, said the vendor is concentrating on performance and dedupe can slow it down.
“We’re positioning this for really high performance environments where they might not benefit from dedupe,” Rome said. “We’re running high performance databases, small block loads, things like that.”
The latest object storage systems are often misunderstood by potential buyers. This may be because these systems, capable of storing extremely large numbers of object and files, serve more than one type of environment. The messaging for one environment can leave an impression that the object storage system is not an applicable solution for another.
Based on Evaluator Group’s work with clients, we have discovered several ways to apply object storage. It helps to understand that in a majority of IT environments, a parallel IT organization has evolved. So you have the traditional IT group tasked with continuing to run current operations – keeping the lights on, if you will. Proficient in current operations, this group struggles with demands for more capacity and greater productivity. The second IT group in the parallel IT organization is the one charged with changing the delivery of services by creating and deploying a private or hybrid cloud.
Object storage can be used in both of these groups. For the traditional environment, object storage serves as a content repository to directly access information that may have been on primary or secondary systems before.
The system can meet the growing capacity demands by using replication and versioning, altering and simplifying the data protection model. This type of system adds value by also serving as a target for retained copies of backups and an online archive. The following diagram illustrates the traditional IT usage of object storage systems.
Use of object storage in private or hybrid clouds is understood but is somewhat hard to depict in a diagram. In general, object storage in a private cloud is a separate system used for many of the same purposes as traditional IT but with different access methods. The major difference is that object storage in private/hybrid clouds is the target for newly written applications that often deal with mobile devices and distributed access. A content repository is another major use case and many times is coupled with file sync and share software. As with traditional IT environments, object storage can serve as online archive and retained backup targets in the cloud. The following diagram shows use of object storage in private/hybrid clouds, representing it as a separate system (logically or physically) from the compute/storage node instances federated together for creating the cloud environment.
Object storage systems are really multi-dimensional with many uses in different environments. In the parallel IT groups that have evolved in organizations, object storage systems can be applied as solutions for growing capacity demands and solving data protection issues from the growth. As this dual applicability becomes better understood, expect more deployments.
(Randy Kerns is Senior Strategist at Evaluator Group, an IT analyst firm).
NetApp’s latest earnings report shows the vendor is still in a hole with a lot of digging to do before it returns to revenue growth.
NetApp reported another quarter of declining revenue Wednesday and laid out plans to reduce its workforce by 12%. NetApp’s overall revenue of $1.39 billion was down 11% from last year and four percent from the previous quarter. Its product revenue of $750 million was down 19% year-over-year and eight percent from the previous quarter. The vendor did turn a profit of $153 million, but that was down from $177 million a year ago. NetApp revenue was below its previous guidance of $1.4 billion to $1.5 billion. Its guidance of $1.35 billion to $1.5 billion for this quarter was also below expectation as financial analysts expected roughly $1.51 billion.
George Kurian, who took over as NetApp CEO in June 2015, said he has completed a formal review of the company and developed a turnaround strategy. The plan includes concentrating on growth areas of the market, cutting costs, and trying to build value in the company through share repurposes, dividends and long-term investments.
On the plus side, NetApp is making progress selling the clustered DataOntap (CDOT) operating software that requires customers to do a disruptive migration. It also reported a strong uptick in all-flash arrays, not including the SolidFire all-flash systems NetApp acquired for $870 million last December.
“NetApp does not need to completely reinvent itself …” Kurian said, adding the vendor is in a transformation that will require taking “significant steps to streamline the business and further advance our pivot to the growth areas of the market.”
The growth segment of NetApp’s products consists largely of CDOT, its all-flash arrays, E-Series performance arrays and OnCommand Insight management software. NetApp is phasing out its OEM business and the DataOnTap 7-Mode product being replaced by CDOT.
The plan is to return to growth by late 2017.
Kurian said NetApp is looking to reduce costs by $400 million annually, with half of that coming from cutting around 1,500 jobs. He said most of the layoffs will occur this quarter.
Kurian said CDOT is now running on 24% of NetApp’s installed FAS arrays, including nearly 80% of FAS arrays bought last quarter. The number of customers who bought CDOT last quarter increased by about 60% over last year.
NetApp executives said their all-flash revenue increased about 60% last quarter from the previous quarter, to around $150 million.
The focus on growth areas at the same time as NetApp makes cuts might not leave much opportunity to jump into new technologies. For instance, NetApp apparently has no plans to come out with a hyper-converged system. Kurian said NetApp can solve customers’ problems with its current products, such as SolidFire and its FlexPod reference architecture program with Cisco.
“We see what customers really want [from hyper-convergence] is essentially simplified provisioning and operational management, like our relatively simple pay-as-you-go building block architecture,” Kurian said. “And you will see address those customer needs with both the SolidFire scale-out architecture, as well as exciting new innovations in the FlexPod lineup.”
With 2015 in the rear-view mirror, FalconStor CEO Gary Quinn says the storage software vendor has completed its transition phase and is ready to reverse its years-long streak of losing money.
“We believe that FalconStor has moved from its transition phase from when I first took over the company in July 2013, and we are now on a normal operating pattern in 2016 and beyond,” Quinn said Tuesday during FalconStor’s earnings call.
“Normal” for FalconStor means its FreeStor data protection and storage management software is fully in the market and subscriptions are coming in. It doesn’t mean the vendor is flush in sales or money, though. For the final quarter of 2015, FalconStor reported revenue of $9.4 million, down from $11.8 million from the previous year. The vendor lost $1.3 million in the quarter compared to a loss of $2.1 million a year ago.
For the full year of 2015, FalconStor’s $48.6 million in revenue was up from $46.3 million in 2014 and its loss of $1.3 million compared to a $6.1 million loss for 2014. FalconStor finished the year with $13.4 million in cash.
Quinn and CFO Lou Petrucelly said the company’s goal is to at least break even for this year. FalconStor has suffered through years of losses and turmoil, including the 2011 suicide of founder ReiJane Huai in the wake of fraud charges.
FalconStor executives claim more than 170 customers are using FreeStor in production. Quinn said FalconStor has 0.1 percent of the software-defined storage market as defined by IDC. But he expects that to grow significantly and predictably as FalconStor moves from perpetual licensing to a subscription model. Because the subscription pricing is deferred revenue, the switch resulted in lower revenue over the last few quarters but Quinn said it will bring growth in the long run.
However, FalconStor is losing a steady revenue stream from an OEM deal with Hitachi Data Systems (HDS). HDS sales of FalconStor virtual tape library backup software regularly accounted for more than 10% of FalconStor revenue, and came to 34% in the fourth quarter of 2014. HDS is now selling its own disk backup product from its 2014 Sepaton acquisition, which will relegate FalconStor software to occasional deals through HDS.
“I would not view them as a contributor going forward,” Quinn said of HDS. “It’s really more opportunistic on a one-off deal here and there where our technology is better than Sepaton or we have existing installed base customers who like our technology and are renewing it.”
Violin Memory’s latest all-flash arrays haven’t received the warm reaction in the market the vendor hoped for, so it’s releasing bundled systems that will make it easier and cheaper to deploy.
Violin launched its Flash Storage Platform (FSP) 7700 with data reduction and protection features last year, hoping to make a renewed push into the all-flash world. But sales have been slow with only $12.5 million in total revenue and $6.3 million in product revenue in the third quarter of 2015. So this week Violin came out with what it calls Starter Kits to simplify sales.
Violin calls these the Violin Scalable Starter Kit and the Stretch Cluster Starter Kit. The Scalable Starter Kit includes two FSP 7700 array controllers, two Brocade Fibre Channel switches, two all-flash arrays with 35 TB each, and Violin Concerto OS7 and Symphony Management Kit. The Stretch Cluster kit adds a Stretch Cluster license that automates recovery of critical applications and data by using two data centers.
Customers can scale beyond 70 TB by adding flash drive shelves.
This is the first time Violin has bundled its arrays with switches. “Before, you would have to buy a 7700 system, switches and a storage shelf as discrete purchases,” said Keith Parker, Violin’s director of product marketing. “Now we’ve packaged everything together in a kit.”
The starter kits can save customers considerably. The list prices are $470,000 for the Scalable Kit and $840,000 for the Stretch Cluster kit. That’s about half the price as it would cost to buy everything separately, Parker said.
Asigra is packaging its Cloud Backup software on the Oracle ZFS Appliance for cloud providers who would rather not purchase and maintain their own hardware.
Toronto-based Asigra for decades have sold backup software only to managed service providers. As cloud backup grows in popularity, MSPs find their hardware infrastructure costs rising to keep up with capacity demands, Asigra executive vice president Eran Farajun said.
Farajun said Asigra has been looking to reduce costs for service providers for several years, beginning with a recovery license model. “Our partners said, ‘We like that but we have other costs in the stack, and that’s the hardware,’” he said. “We said, ‘We’re not a hardware company, we’re a software-company.’ They said ‘Go into your cave in Toronto and try to figure out a way to lower costs.’”
Farajun said Asigra looked a low-cost hardware and found the ZFS Storage Appliance. “We didn’t know Oracle even made storage,” he said. “We did trials and got good performance. We joined their OEM program. That gives us access to Oracle products and aggressive prices, and we can pass that on to our partners.”
He said for providers, buying Asigra backup on the Oracle appliances is “cheaper than cobbling together your own system. Some try using commodity hardware with CentOS, fully open-sourced, then they run into trouble and there’s nobody to call for support.”
Asigra actually does cobble together its own appliance for smaller partners. Last year it began bundling Cloud Backup on Supermicro appliances with FreeBSD open-source software. Farajun said the Asigra appliance is scales to 100 TB, while the Oracle ZFS appliance begins at 100 TB.
Is Commvault back on track?
The backup vendor snapped a string of four quarters of year-over-year revenue declines when it reported of $155.7 million last quarter, up 2% from the previous year. More impressively, its software revenue of $71.4 million increased 24% from the previous year.
While Commvault continued to make money during its sales declines, its $13.2 million income last quarter was its highest take in a year.
Now we’ll see if Commvault can gain momentum with its its new platform launched in October. Wall Street analysts expect revenue of $156.3 million this quarter, a small uptick from last year.
Commvault spent the past year revamping its sales and marketing teams before the new data protection and management platform launch. Discussing the earnings Wednesday, Commvault CEO Bob Hammer said that transformation period is over.
Hammer said the latest software platform was a controlled release to select customers, with broader distribution expected this quarter. But he said the changes made to that release played a key role in attracting customers. Commvault reacted to claims that its software was too expensive and complicated by altering its pricing model and making it available in smaller bundles.
“As we went into this transformation, we had to segregate the product line, pricing, packaging … “ he said. “We had to make some really significant underlying changes to the platform to align to the cloud..
“We experienced significant increases in the amount of large enterprise deals and had higher than normal close rates,” he added.
Commvault’s enterprise deals of more than $100,000 in a quarter represented 54% of its software revenue, and increased 33% from the previous quarter.
Hammer said more enterprises are moving backups to the cloud, which hurts storage hardware vendors but is a positive for Commvault because it helps with the process. Commvault software use the cloud as a repository, just as it would if the backups were going to disk or tape.
“As customers migrate to the cloud, adopt new IT infrastructures and deploy newly architected applications, they are looking for a strategic partner who has technology and services that can help them make the transition,” he said.
“AWS might have hundreds of different data centers, so we can move the data in those clouds, orchestrate those infrastructures, spit up compute network storage in the cloud, tie to a given application and manage and index all that data. So now customers have a complete picture of what the heck is going on with the data no matter where it resides. And we can federate across the different AWS silos or [Microsoft] Azure silos or between Azure and AWS or Azure and OpenStack, et cetera.”
Another thing helping Commvault is the largest backup software vendors are going through their own changes. EMC is getting acquired by Dell, Veritas just spun out from Symantec, Hewlett-Packard split up and IBM has been in a storage revenue freefall.
“I’m not going to get into a specific commentary on Veritas or what’s going on with EMC or some of the other larger competitors of ours, HP or IBM,” Hammer said. “But, clearly we’ve gotten significantly out in the front of every one of those competitors from a technical standpoint, a services and support standpoint, and we’re going to create additional distance between us and these competitors. We are on a much firmer foundation right now … and every one of the competitors I just mentioned has what I call significant architectural underlying issues that will take time and money to address.”
Commvault still has to worry about Veeam Software, which has been cutting into Commvault’s business from below. Veeam, a private company, said it had a 55% increase in bookings revenue in the fourth quarter of the year. Veeam said its bookings revenue for 2015 totaled $474 million. Commvault’s revenue for calendar 2015 was $581 million.
David Goulden, CEO of EMC Information Infrastructure, teased new product launches to the VMX, VNX, DSSD and Data Domain product lines Wednesday during EMC’s earnings conference call. Flash will play a major role in all of the primary storage arrays, including at least one new all-flash platform.
“Flash is one of the megatrends that’s changing the infrastructure business forever,” said Goulden’s boss, EMC CEO Joe Tucci.
Goulden didn’t give away too many details, but he said VMAX enterprise and VNX midrange arrays would be “re-archtected” for flash. The VMAX and VNX are legacy arrays developed for hard disk drives but were retrofitted to accompany solid-state drives (SSDs) when they became available for primary storage. Goulden said new all-flash versions will include substantial changes to take full advantage of new flash technologies.
In an attempt to set a record for using the word “flash” the most times in a sentence, Goulden said this quarter EMC will introduce “a new flash-optimized all-flash VMAX that will significantly change the way flash is deployed in high-end primary storage.”
It’s unclear if EMC will re-name VNX or add another midrange product. While Goulden referred to VNX by name several times on the call, he also talked about a “mid-tier storage family’ coming in the second quarter, “which will change the use cases for flash in the mid-tier.” Whether that is VNX or another family remains to be seen.
DSSD is a new product, with technology acquired when EMC bought Andy Bechtolsheim-founded startup DSSD in 2014. EMC has previewed server-based DSSD at events over the past year but the system has not been released. Goulden said DSSD will launch this quarter, calling it a “quantum leap” in flash. He said DSSD will deliver “mind-blowing performance, bandwidth and latency for high-performance business applications like Hadoop analytics and ultra high-performance databases.”
Goulden did not mention any updates to XtremIO, which is the market leader in all-flash systems with $1 billion worth of sales in 2015. He did mention integrated copy data management (ICDM) added to XtremIO last year that puts database copies on primary storage.
EMC will also release a software-only version of the Data Domain disk backup platform and new converged products from VCE. The vendor has been in development with the virtual Data Domain for several years but wasn’t sure if it should cannibalize its popular hardware backup system.
Goulden said EMC and VMware have developed “a new next-generation hyper-converged appliance family” and VMware will make an announcement in February.
Despite any backup or hyper-converged developments, 2016 is shaping up as the year of all-flash for EMC. Or as Goulden said, a shift to “all-flash, all-the time ‘ in the midrange and high end.
“Flash is not about a single product,” Goulden said. “It’s a key technology across our portfolio. We really think that the technology has advanced to the stage with the latest 3D NAND technology and things like 3.8 terabyte drives. Of course you need to architect your system to optimize to use something that big and that fast, which is why we talked about the re-architecting. We can really come to market with a complete family of VNX, VMAX, XtremIO, DSSD, leveraging this latest technology and basically use all-flash all the time for primary storage.”
Product overlap is nothing new to EMC, but how will it explain to customers which all-flash platform to use?
“DSSD is going to address a whole new class of workloads,” Goulden said. “XtremIO and VMAX are playing in broadly similar markets, but with different attributes. The mid-tier line fits underneath that. So we really have the market exceptionally well covered and of course we’re leading with an all-flash agenda.
Gridstore closed a $19 million funding round today with plans to substantially expand engineering and sales of its all-flash hyper-converged systems built for Microsoft Hyper-V.
The round brings Gridstore’s total funding to $45.5 million over three rounds. While $19 million is small compared to some of the storage funding rounds over the past year, Gridstore CEO George Symons said it is more than enough to fund Gridstore’s growth plans.
“I’m a believer in not going out and raising stupid amounts of money, even if you can,” he said. “If the goal is to go public in a year and lose large amounts, that’s the way you do it. Our goal is to build a business.”
He said the funding is enough to allow Gridstore to more than double from its current 38 employees by the end of 2016. Symons said most of the growth will hit the vendor’s two largest departments, sales and engineering.
Gridstore has already hired a chief strategy officer and chief financial officer – two positions it never had before. Former Del CTO James Thomason will fill the CSO role. Kevin Rains is the new CFO. Rains also worked at Dell as director of operations for its Enterprise Software Group. Gridstore also hired Phillip Lavery as VP of sales for the Americas and Nariman Teymourian, most recently senior vice president and general manager of the Converged Systems Division of HPE, replaces Imation CTO Geoff Barrall as Gridstore’s chairman.
Gridstore started selling storage appliances for Microsoft in 2013 and added all-flash hyper-converged systems in 2014. Symons said Gridstore revenue grew 343% last year, with new new customers “almost exclusively hyper-converged. It surprised me how quickly it happened and how completely it happened. I don’t think we had more than one or two new standalone storage customers last year.”
Symons said Gridstore will drop its storage-only system but will expand its hypervisor support by adding KVM. As for market leading hypervisor VMware, Gridstore will continue to leave that space to other hyper-converged players – including VMware.
“We’ll continue to watch that market,” he said. “Just jumping in as another hyper-converged infrastructure player doesn’t seem to be the right move for us.”
Atlantic Bridge Capital led the funding round with previous investors Acero Capital, GGV Capital and Onset Ventures participating.
NexGen CEO and founder John Spiers said the deal “came about serendipitously” at the suggestion of a venture capitalist firm. While working on a funding round, NexGen talked to one of Pivot3’s investors who suggested the two explore a merger. Spiers met with Pivot3 CEO Ron Nash and the two agreed they would be better as a larger company. The vendors’ investors decided how to allot the shares in the new company.
“It’s really a stock swap, there’s no real purchase price,” Nash said.
Nash will be CEO of the new Pivot3. Spiers will become Pivot3’s executive vice president and Chief Strategy Officer and fellow NexGen founder Kelly Long will be a CTO in the company. Nash said all NexGen’s employees will join Pivot3.
Pivot3 sells its VMware-based vStac OS on all-flash and hybrid hyper-converged systems combining storage and compute. It also sells a system optimized for video surveillance and high-density blades that fit in a rack to support up to 9,000 virtual machines.
“We saw in NexGen a unique opportunity to expand upon what we have been doing in hyper-convergence,” Nash said. “We saw that we can be boarder in terms of the range of performance we can offer to IT shops. And they have a lot of storage functionality that makes us deeper.
“NexGen gets to combine with us, we have more international reach and additional products. It’s a great combination all around.”
Nash said NexGen’s 86 employees will join Pivot3’s 130 to expand the headcount to more than 200. He said Pivot3 has more than 1,600 customers and NexGen has around 400, giving the combined company more than 2,000 customers. Austin, Texas-based Pivot3 will keep NexGen’s Boulder, Colorado, office.
Nash said Pivot3 will continue to sell both companies’ current products, and they will port software features to the others’ platforms. The NexGen brand will continue for now, but there will be new products combining both vendors’ technologies.
Among NexGen’s software features, Nash said Pivot3 is most interested in quality of service and dynamic provisioning.
“There is a new wave of technology breaking out,” Nash said. “In a software-defined data center, you will have some hyper-converged, some storage-only and some cloud products – all based on a commodity x86 processor with software on top of it.”
This is the second time NexGen has been acquired. PCIe flash vendor Fusion-io bought NexGen in 2013 for $119 million. After SanDisk subsequently acquired Fusion-io, SanDisk spun off NexGen in January, 2015, putting it back in control of original founders Spiers and Long.
“We weren’t a good it for SanDisk’s business strategy because we competed with their OEMs,” Spiers said. “Our growth plans [for 2016] were to triple the size of our sales force, hire key marking positions and add key people in Europe and Asia. Pivot3 brings all that to the table.”
“We think this new wave of technology is going to knock out several incumbent IT companies,” Nash said. “We think we have a chance to be one company that pops up to replace them.”