Network Computing

Ever since launch of the PC, and increasingly more now, organizations have been dealing with users who bring technology into the company that they had started out using personally. One of those is the Drobo storage device, which gives users the ability to set up a redundant array of independent disks (RAID) system by mixing and matching a variety of disk drives. Faced with the fact that as many as a quarter of its sales were to small to midsize businesses rather than to consumers, the company is announcing three models of its device--two eight bays and a 12 bay--that are actually intended for business use.

Jeffrey Cochran, senior enterprise network engineer for the Nebraska Book Co., says his company, which sells textbooks and operates college bookstores, uses Drobo devices for more flexible storage needs, such as Websites, compared with the EMC SANs the Lincoln company also uses.

"It sounds a little funny, because we have these itty-bitty Drobos in a rack next to the EMCs," he says.Cochran says he can "go out and buy a couple of terabytes for a couple of hundred bucks," while buying 4TBytes for the EMC SAN could cost him up to $30,000. While the Drobo will never replace EMC because of the space and speed the SAN can provide, "for lower-level administrative tasks or storage, Drobo certainly fits the bill perfectly," Cochran says.

New features in the Drobo devices include thin provisioning to help SMBs deal with capacity and utilization planning, says Kevin Epstein, VP of marketing and product management for the Santa Clara, Calif., company. Users can tell an application such as Microsoft Exchange that the device has 20TBytes, but put in just 8Bytes, and the device will let the user know when it's getting full, he says.

With the new products specifically targeting the SMB space, Drobo is looking to move into a growing market that has historically made do with either stripped-down enterprise offerings or higher-end personal storage products -- neither of which really satisfied the SMB storage demand: higher-end features and capacity with an ease of use and price point similar to consumer products, says Liz Conner, senior research analyst for storage systems and personal storage at IDC, a Framingham, Mass., consultancy.

She was particularly interested in the 12-bay model because that is a "sweet spot" for SMBs. Though she noted that other vendors already have 12-bay versions on the market, features such as management tools, data tiering and 24/7 support really tailor it to small businesses, which typically have small or minimal IT staff and may not have someone with storage-specific knowledge, she says.

The new models of the devices start at $2,199 and consist of an eight-bay file-sharing Drobo with remote backup and an eight-bay iSCSI-attached storage area network Drobo, each of which are available now. A 12-bay iSCSI SAN version with expanded redundancy, support for thin provisioning and deprovisioning, and data-aware tiering is expected to be available in the second quarter.

See more on this topic by subscribing to Network Computing Pro Reports Research: 2011 State of Storage (subscription required).

Dell introduced on Tuesday a sweeping new lineup of laptop, desktop and workstation computers designed to welcome the "consumerization of IT," while still serving enterprise demands for support of virtualization, cloud computing, security and other IT priorities.

The computer maker unveiled 39 new products or related IT solutions for business users to be released during the next year, including a revamped line of Dell Latitude laptops, a new OptiPlex family of desktop computers and new Dell Precision workstations. The strategy behind the development of these new products is to incorporate designs and features that today's knowledge workers have come to expect in computers because they grew up with them, says Steve Lalla, VP and general manager of Dell's commercial client products group.

"In some instances, employees have been bringing their own laptops, tablet computers or smartphones into work, a phenomenon called the 'consumerization of IT,'" Lalla said at a launch event held in San Francisco. Dell has found that enterprise customers want to give workers lightweight and stylish laptops, and allow social networking apps and collaboration capabilities, but still manage and secure their IT.

"It's about how to bring all of this new technology together, how to contemplate what compute gets done at the end point, what compute is done in the data center and how to tie that all together seamlessly," he says. All products incorporate Dell Data Protection security, including automatic data encryption and Remote Data Delete for the laptops in the event a machine is lost or stolen.

The laptops and desktops have been designed for, among other things, convenient maintenance by IT staff, adds Ken Musgrave, director of industrial design at Dell. Even though Latitude laptops come in different sizes with different-sized screens, they all have the same-size keyboards for simpler replacement if necessary. The OptiPlex desktop computer comes with the machine and the monitor built as one unit to avoid cabling hassles. Also, the laptops are protected against scrapes and scuffs with a cover made of the same material as the liner used in pickup truck beds, Musgrave says.

"[Dell's] new lineup is probably one of the strongest that they've had in years," says Tim Bajarin, principal analyst with the research firm Creative Strategies, "and most likely will help them maintain a  pretty strong position in the enterprise space." In the past, enterprises acquired technology that workers used only to get their work done, Bajarin says, but that has changed as people use technology for their work and personal lives simultaneously.

"Now, with the whole issue of people taking their work home and the device being a central part of their personal as well as business lifestyle, [enterprises] have no choice," he says. "They can control the corporate entity part of the technology, but they have really got to give flexibility  to [workers] to let them use these devices for consumer purposes, as well."

Dell also teased the upcoming release of a new tablet computer with a 10-inch screen that will run Microsoft's Windows 7 and be targeted at business users. It already markets two models of the Dell Streak tablet that runs Google's Android operating system.

Dell said pricing for the Latitude laptops starts at $859, for the OptiPlex desktops at $650 and for the Precision workstation line at $840. More details on availability of the new products is expected in the next few weeks.

See more on this topic by subscribing to Network Computing Pro Reports Informed CIO: Mobile Device Security (subscription required).

Falconstor Software has added functionality to its Continuous Data Protector (CDP) and Network Storage Server (NSS) data replication products that make it easier for organizations to use those products for disaster recovery. Moreover, the RecoverTrac software is heterogeneous among platforms, networks, arrays and hypervisors, meaning that organizations would not need to have identical implementations in different parts of the company to be able to restore data, as is typically required today.

Data can also be recovered from physical to virtual, physical to physical, and physical to virtual infrastructures. In addition, there is no per-server or per-client charge for the functionality.

With the announcement, Falconstor is differentiating between disaster recovery and remote replication, says Bobby Crouch, product marketing manager for the Melville, N.Y., company. Remote replication is simply a remote copy of the data, without the application and the infrastructure that enables users in the organization to make use of the data.

"When there's a disaster, customers don't care that you have a remote copy," he says. "They can't access the application that uses that data." In addition to providing copies of data, the software enables users to resume servers, storage, networks and applications, and works with most operating systems, virtual machines and networks. This is required because business applications increasingly have an awareness of transaction integrity, with applications such as SAP behaving differently from applications such as Oracle, says Crouch.

The software has three modes: Recover, or recovering data; Test, or a non-disruptive testing function that lets users create recover jobs and refine them to ensure that they're correct; and Clone, which creates a tertiary copy of the data that can be used for functions such as data mining and analysis, Crouch says.

RecoverTrac means that the disaster recovery features in Falconstor's CDP and NSS products are now more practical and easier for organizations to use and implement because many disaster recovery solutions today are not really ready to do turn-key disaster recovery, but instead require testing or professional services to set up, says Jerome Wendt, president and lead analyst at the Datacenter Infrastructure Group, an Omaha, Neb., consultancy.

The product is particularly appropriate for two sets of users: existing FalconStor users who have already implemented Falconstor's CDP and NSS, as well as users who are willing to implement NSS and CDP in their environment to use RecoverTrac, he says. On the other hand, there is nothing to prevent FalconStor from taking and applying RecoverTrac's technology and using it to manage other storage vendors' storage virtualization and CDP/replication offerings to achieve the same thing, though Wendt was not aware of any plans on FalconStor's part to do so.

The RecoverTrac functionality is included in Falconstor's CDP and NSS products now for no additional charge. Those products range in price from $2,000 per terabyte to $6,000 per terabyte, depending on volume. RecoverTrac is integrated both with VMware's Site Recovery Manager and Microsoft's Cluster Adapter, and supports both the VMware and Microsoft Hyper-V hypervisors.

See more on this topic by subscribing to Network Computing Pro Reports Strategy: Deduplication for Disaster Recovery (subscription required).

At first glance, desktop virtualization is the poster child for the problem of satisfying an increasingly mobile workforce with increasingly scarcer resources. However, despite the proliferation of vendors entering the VDI market--including Citrix, MokaFive, Pano Logic, Ring Cube, Symantec, Oracle and VMware--the reality has been both technically challenging and more expensive than expected.

RingCube Technologies is tackling what it calls the "VDI ghetto"--namely, personalization--with a new solution that expands an organization's VDI user base, enables faster and more efficient VDI rollouts, reduces VDI infrastructure costs, and lowers a VDI project's cost per user while increasing its return on investment (ROI). vDesk VDI Edition for VMware View and Citrix XenDesktop is available today with pricing starting at $95.

RingCube's new offering addresses the current practice of restricting personalization to keep costs and complexity down, as well as limiting the technology to structured task workers. The company says that there are almost 600 million enterprise PCs in use, but today's VDI can address only 10 to 20 percent of them, ignoring campus (30 to 40 percent), mobile (30 to 40 percent) and remote worker (10 to 20 percent).

The biggest challenge to widespread VDI use has been user acceptance, which has left most enterprise VDI implementations in pilot mode, says RingCube. What vDesk VDI Edition does is provide the personal desktop experience on top of VDI for reduced costs. The company says that it is leveraging the infrastructure that VMware, Citrix and Microsoft provide today, adding a layer to their offerings.

This integration with existing VDI implementations is a primary strength of vDesk VDI Edition, says analyst Steve Brasen, Enterprise Management Associates. The ability to customize for individual users makes the VDI investment more applicable to business requirements while simplifying management and reducing infrastructure costs. The product also broadens the appeal of the VDI platforms with which it integrates.

"Traditional VDI implementations are valuable only in specific use cases--enterprises that include a workforce that utilize a common system environment. It has not been particularly useful in environments where end users require extensive customization in their desktops to achieve the disparate requirements of their individual roles. vDesk VDI Edition specifically addresses this problem by allowing individual users to overlay environment settings and customizations on top of a common desktop environment. This provides a "best of both worlds" scenario, achieving minimal storage requirements while allowing flexible user environments. This provides a distinct advantage to VDI platforms that integrate with vDesk VDI Edition over those that are unable to achieve similar capabilities. RingCube cleverly orchestrated this solution so that it enhances existing VDI solutions, rather than competing against them.

George Hamilton, an analyst with Yankee Group, says the company is "able to support a cool model but maintain customization for the end users. It allows you to extend pooled resources for tethered works."

Brasen says there are a number of drivers for desktop virtualization adoption, but the three that are most common are workforce mobility (users need to access their data from multiple different workstations, such as work laptops, home desktops and mobile devices); simplified client management (with all user environments managed on a single centralized platform, administrative tasks are significantly reduced, and processes for patching, configuration, system updates, backups and other maintenance practices are far less time-consuming and simpler to implement; and cost reduction (since physical endpoint requirements such as CPU, memory and disk size are minimized, substantial capex savings can be achieved). Additionally, the reduced client management requirements diminishes administrative operating expenses.

While he thinks the VDI market is too immature for this announcement to have significant impact, at least on the other vendors, Hamilton says they're seeing more organizations looking to get out of the desktop-support business. "You can buy a device for $500 to 700, but then have to spend $2,000 to $3,000 or more to support it." It becomes further complicated when employees buy their own machines, including Apple Macs. "Even my own organization is starting to do that. We're using a combination of cloud-based and local installations." This evolution is starting to get more traction, he says, with users getting PC support from a variety of sources. "So this is just another option on how I can deliver apps to a machine I don't necessarily own or manage."

See more on this topic by subscribing to Network Computing Pro Reports ">Research: VDI Adoption 2010 (subscription required).

Solera Networks has introduced traffic classification and identification with deep packet inspection, including highly detailed application information and visualized geolocation, to its network analysis platform. Solera OS 5, supporting the DeepSee suite of tools, also features an improved database engine for better performance and dynamic updating of dashboard displays.

Solera is among a handful of vendors that capture, store and analyze all network traffic. These capabilities are generally focused on security, but have considerable value for network operations as well, as they help ops teams determine the cause of network outages and performance issues.

"The goal is to catch an incident before anyone sees a problem, before it impacts a user," says the security administrator for a large government contractor. "But, if there's an incident or a machine is acting slowly, you can immediately go back--we're currently configured to go back a full month--to trace the problem to the point of origin."

This class of tools is designed to literally see everything that goes on across the network and enable enterprises to spot problems and investigate issues quickly. Solera describes its capabilities as network forensics. Forrester Research has labeled it network analysis and visibility (NAV), maintaining it is essential to enforce a "zero trust" approach to enterprise security (trust no one, see everything). Without this ability to capture, store and analyze many terabytes of network data, enterprises have to rely primarily on manual log review and "snapshot " packet capture that doesn't provide historical data and may not "see" malicious activity, such as a botnet "phoning home" to a command-and-control server.

This kind of capability is designed in large part to dramatically reduce time to resolution of security and network incidents, getting business systems back on line and fully functional.

Full network analysis and visibility has become increasingly important in the face of what Solera characterizes as next-generation threats, such as Stuxnet, advanced persistent threats (APTs), bots, sophisticated malware and massive insider incidents, such as WikiLeaks.

"Advanced persistent threats is the whole reason to be for network forensics," says Pete Schlampp, Solera VP of marketing and product management. "Once they get on your network, they have multistage and multivector capabilities and can morph identify." Network forensics allow organizations to analyze the changes over time, identify the root causes and remediate.

The new DPI capabilities enable Solera to identify 500 applications, which it organizes into 28 families. Solera says that it extracts some 5,000 descriptive details to support its analysis and reporting. The 5.0 engine automatically generates high levels of detail about applications. For example, previously you had to deconstruct an e-mail message to obtain the address and other information. Now, you can automatically extract information such as sender, recipient, subject line and attachments from Gmail.

The geolocation feature creates visual maps of traffic between IP addresses, enabling operators and analysts to quickly begin to identify and address issues.

"The geolocation piece has increased productivity insanely," says the defense contractor security administrator. "Before, I had to load another tool to trace where the IP was. Now it's all integrated in one." Data can be exported in a file to Google Earth.

The new database engine release supports what Solera says is real-time classification of network traffic on an enterprise scale. Solera says that one customer's average network utilization on a one Gbit network produces 36TBytes of data in a month. Dashboard displays are updated dynamically to reflect current traffic or frozen to begin an investigation.

Solera also provides native integration for several products to move from alerts generated into investigations and provide additional data. Current partners include Sourcefire and SonicWALL for IPS; Palo Alto and QOSMOS for DPI/classification; Splunk and ArcSight for SIEM; and lgo management and FireEye for malware analysis.

See more on this topic by subscribing to Network Computing Pro Reports Research: WAN Security (subscription required).

Lancope's StealthWatch 6.0 network behavioral analysis tool features granular application awareness, flexible grouping of network assets for reporting and analysis, and relational mapping for network visualization.

Network behavior analysis plays on both the security and network operations sides of IT by collecting and analyzing network flow telemetry via Netflow, sFlow, JFlow, etc., to identify and remediate the cause of anomalous activity, such as traffic spikes, performance degradation, and communication with unexpected IP addresses that might indicate botnet activity or data exfiltration.

"The same sort of instrumentation points and same sort of measurements can be used quite effectively in both realms," says Jim Frey, research director, enterprise management, for Enterprise Management Associates.

On the security side, the addition of application awareness, through deep packet inspection, helps enterprises identify potentially malicious activity and applications, such as peer-to-peer or social networking sites, that may be banned or restricted by corporate policy. On the network side, fine-grained application awareness enables ops teams to determine if reports of "slowness" are caused by network problems or application issues, such as authorized or unauthorized video streaming, an issue with an authorized business application or a malicious program that needs to be referred to security.

"Is it the network or the application? Everyone points fingers when users report 'slowness' in something," says Joe Yeager, Lancope product manager. "It's always the networks that are blamed, but the networks are only responsible 20 percent of the time." Understanding the cause of performance issues saves organizations from throwing bandwidth capacity at what appear to be network issues but may be related to applications or a faulty DNS server.

Another major enhancement is the ability to assign network assets to any number of "host groups for reporting and policy management," instead of inflexible zones. Previously, StealthWatch required that a device be assigned to a single zone. So, for example, an Microsoft Exchange Server could be assigned to an Exchange Server zone or a New York zone or a sales and marketing zone, etc., but not more than one.

Zones are appropriate for security analysis and reporting, says EMA's Frey, but the flexible grouping is far better for network operations. Groupings can be used to tailor reports by organizing data in a way that reflects business operations.

"With the more flexible, business-oriented grouping, data can be presented, viewed and studied in a way that is consistent with the way the organization ... is organized," he says. So, he adds, reports can be used effectively for collaboration within IT teams and to work with people in the supported end user community.

The relational mapping feature produces customizable diagrams to show network flow between assets based on network topologies, different parts of the network, groupings, and so on. Frey believes this will help network and security personnel spot problems more quickly.

"Humans are good at pattern recognition in graphical image consumption; they've never been particularly adept at processing and consuming tables of data," he says. "When you present data in way that's visually impactful, there's a huge productivity and efficiency improvement."

Quick recognition improves time to resolution, saving money by bringing production systems back on line or to peak performance. It also gives organizations the option of putting the screens in front of their support teams. "There are big savings with first-call resolution on the help desk side."

Lancope has also improved its performance and data storage capability, replacing its database engine through an OEM agreement with Vertica. Entry-level pricing for StealthWatch starts at $55,995.

See more on this topic by subscribing to Network Computing Pro Reports Best Practices: Performance Management (subscription required).

Data management appliance vendor BridgeSTOR has released a new application-optimized storage (AOS) appliance that shrinks data in Microsoft's System Center Data Protection Manager (DPM) backup and recovery system. The new appliance provides data deduplication and compression capabilities that DPM doesn't offer on its own, and BridgeSTOR says it can shrink data stores to as small as 10 percent of their original size.

While other data deduplication products are expensive high-end systems, the BridgeSTOR appliance is targeted at small to midsize businesses, or branch offices of larger companies, with pricing that starts at $20,000, says John Matze, CEO and founder of BridgeSTOR. He says he has visited customers that have reduced their server count down to one rack through virtualization but still have four or five racks of storage hardware.

The DPM-based appliance brings to four the number of AOS product lines from the company. It already offers appliances that support data management from VMware, Symantec Backup Exec and Microsoft Windows Unified Data Storage Server.

In addition to data compression and deduplication, which is the deletion of multiple copies of the same file, the AOS appliance offers thin provisioning, which adds storage capacity in small increments only as needed, and optional data encryption for security purposes. "We can optimize customers' storage requirements so they can use half the storage or a third of it. That is a substantial dollar savings, not only in purchasing the storage but having to pay the cost of heat, electricity and maintenance," Matze says.

The appliance is based on an HP Proliant DL 160 server and ships with Microsoft System Center DPM pre-installed, along with the appropriate Windows Server 2008 operating system. DPM also supports Microsoft Exchange for e-mail and SQL for databases.

To illustrate how much AOS can compress data, BridgeSTOR says, a storage system with 3.5TBytes of physical capacity appears to DPM as 35TBytes of virtual capacity, expandable to 10.5TBytes of physical storage and 105TBytes of virtual capacity.

BridgeSTOR says it achieves that level of data reduction with its exclusive Virtual Storage-Advanced Data Reduction (VS-ADR) technology. It combines in-line, ASIC-assisted block-level data compression and deduplication with "disk on-demand" capability so the customer only has to add disks as needed. This also saves money, Matze explains, because the price of disks continues to decline so there's no sense in buying large numbers of disks at today's prices when the prices will likely go down tomorrow.

The storage appliance market serving SMBs underwent some consolidation last year when BridgeSTOR competitor Ocarina Networks was acquired by Dell and Storwize was acquired by IBM. Other firms in the space include ExaGrid Systems, Nimble Storage and GreenBytes.

See more on this topic by subscribing to Network Computing Pro Reports Research: 2010 Data Deduplication (subscription required).

Security information and event management (SIEM) vendor SenSage has opened up its data to third-party business intelligence (BI) tools and dashboards, enabling organizations to leverage security data with business data analytics and create highly customized reports. The latest release (version 4.6) supports Open Database Connectivity/Java Database Connectivity (ODBC/JDBC) APIs, allowing integration of SenSage's data warehouse with BI products through SQL queries and/or the BI tools' more user-friendly query wizards.

"What I call vendor promiscuity," says 451 Group analyst Andrew Hay, "opens up opportunities for people to extract the data they need to get the job done and not be constrained by the vendor interface."

While SIEM vendors are known for their ability to import data from many sources and, in some cases, to open data to technology partners, they have not allowed their data to be pulled in on-demand using open standards. Hay thinks more SIEM vendors will move in this direction as long as they can protect their intellectual property.

This type of open-standards approach to information sharing, he says, opens up many possibilities in addition to porting data into BI tools, dashboards, and other reporting and analysis tools. For example, third-party vendors could leverage SenSage's scalable data warehousing capabilities if they lack their own. Or, enterprises could create Web applications to present highly focused dashboards to present, say, security information to a particular business division.

"Opening up the architecture to business intelligence tools is an opportunity to take what has been most useful in the business data analysis universe and bring it to the security universe," says SenSage president and CEO Joe Gottlieb. "It's a data mining problem. This brings the state of the art of data mining to security."

Bringing security data into business intelligence should enable enterprises to bring more granular context to security analysis to better assess risk based on the potential business impact of a threat.

The 4.6 release also provides:

• Expanded interoperability through industry APIs, with the ability to accept alerts from third-party products like IBM Tivoli and HP OpenView, leveraging open APIs such as SNMP as well as proprietary APIs such as Check Point LEA;
• An updated analytics installer and log adapters with new views and source-specific reports;
• Database storage for reporting history;
  
• The ability to audit security administrator changes;
• Improved SNMP Sender and Retriever, which acts as a bridge and can put all data onto a syslog stream; and
• Support for RedHat 5.5
  

"This is an indication that the SIEM space is maturing," says Hay. "We can push a SIEM-style or LM product to the operational and hands-on security people and present them with this high level abstraction layer for business-focused risk and compliance."

See more on this topic by subscribing to Network Computing Pro Reports Research: WAN Security (subscription required).

A company called Gizmox has recently introduced Instant CloudMove, a set of tools for converting client/server-based software applications to run in Web- and cloud-based environments and on mobile devices.

Gizmox says that the conversion allows enterprises to run applications that reduce CPU workloads to 50 percent and bandwidth consumption to 10 percent of what they would be as rich Internet applications (RIA). Instant CloudMove also serves 10 times more end users than other application virtualization and streaming solutions and offers better security, the company says.

Instant CloudMove leverages Gizmox's existing Visual Web GUI (VWG) 6.4 platform, which helps virtualize client-based applications written in Microsoft .NET and Visual Basic 6. That makes it possible for the apps to run in any HTML- or HTML 5-enabled Web browser.

The main driver for Instant CloudMove is the fact that enterprises have invested heavily in their client/server applications, but they now have to adapt to new business models to deploy those apps through the Web and onto mobile devices, says Navot Peled, president of the Tel Aviv, Israel-based Gizmox. "We offer a path to port those applications to the Web and the cloud," he says.

Instant CloudMove reduces CPU cycles by eliminating the need to change and re-deliver objects of the application from the server, Peled explains. The only things that move on the network are the packets of data that go back and forth as a person uses the application. "The best way to describe it is that we are not baking the cake; all we are doing is sending the recipe to bake the cake yourself," he says.

Security is also improved with cloud-based delivery, Peled adds, because the end user sees only the part of the application that he or she needs to see to perform a particular task, and no data from the application is stored on the client machine. With such security in place, highly sensitive e-banking and medical files applications can safely run in a cloud environment.

Gizmox is also targeting mainstream business applications such as enterprise resource planning (ERP) and customer relationship management (CRM) for its conversion tools. Because the apps can run in any HTML browser, workers can access them from a variety of mobile devices popular today, such as notebook and tablet computers, as well as smartphones.

Gizmox is offering a free Assessment Tool that generates a report detailing how close the application is to matching the Web/cloud and mobile version. At that point, the customer can complete the conversion without specialized expertise or training, the company says.

See more on this topic by subscribing to Network Computing Pro Reports Strategy: Bringing APM to the Cloud (subscription required)

The only pure-play storage resource management (SRM) vendor in the visionary category of Gartner's "Magic Quadrant for Storage Resource Management and SAN Management Software," Aptare is looking to further separate itself with significant enhancements (Feature Pack One) to its unified storage management product suite, StorageConsole.

The biggest innovation is an agent-less host data collector that applies multiple methods in order to capture storage capacity metrics from the hosts and heterogeneous storage systems. The collector "learns" and optimizes its data collection for subsequent scheduled collections.

There are hundreds of enhancements in Feature Pack One, but the two major ones are related to capacity and backup management, says the company, which focuses on the global 50,000.

The capacity enhancements include storage array support for EMC Symmetrix VMAX and Hitachi VSP/HP StorageWorks P9500, capacity and forecasting for EMC Virtual (Thin) Volume Pools (Clariion, DMX and VMAX), and chargeback for replicated volumes. On the backup side, there is now advanced support for EMC Data Domain that includes capacity and predictive analysis, as well as reporting on EMC Avamar replication processes.

There are a number of strengths to StorageConsole, but the ability to provide this information in an "agentless" manner is really forcing change in the landscape, says Bob Laliberte, senior analyst, Enterprise Strategy Group. "Organizations don't want to dedicate resources to managing an SRM solution and updating agents; rather, SRM solutions need to be easy to install, simple to maintain, and provide actionable information where and when it is needed so organizations can make intelligent decisions regarding their storage infrastructure."

Enterprise Strategy Group is seeing more organizations consuming SRM as a service, leveraging the information collected without having to worry about deploying and maintaining software, he says. "Aptare is definitely one of the leaders in these areas and continues to set the bar for intelligent and agentless information collection, reporting flexibility and choice of consumption model."

The two biggest storage management trends going forward are the increased use of server virtualization technology and managing data growth, says Laliberte. "They're not unrelated, especially when you consider that virtualized server environments need a networked storage environment in order to take advantage of all those great mobility functions like vMotion, DRS, high availability, etc."

Organizations need to be able to manage their storage environment more effectively to remain competitive in their respective markets, he says. "Server virtualization has brought tremendous advances for provisioning new virtual machines, minutes or hours instead of days and weeks. The storage environment needs to be able to respond to these new demands, and the only way they can do that is if they have the correct and current information about their storage environment. Organizations are also turning to storage virtualization technologies, deduplication, etc., to drive greater efficiencies. So it is critical that organizations have the insight and visibility into these virtualized [thin provisioned or fully virtualized]and deduplicated environments so they can make the appropriate decisions on where to provision new storage, when to order more and how to optimize existing investments."

You can read more about IPv6 with Network Computing Pro Reports Strategy: Storage Virtualization [subscription required].

Today, the Internet Assigned Numbers Authority ( IANA) will be passing out the last five Class A address spaces to the Regional Internet Registries (RIR), which in turn allocate IP addresses to organizations within their specific geographic regions. This marks the end of the IPv4 central pool, and the countdown to IPv4 exhaustion begins in earnest. The American Registry of Internet Numbers (ARIN)--which handles allocations for Canada, the United States and many Caribbean islands--is forecasting that it will run out of address space in the July-to-October time frame.

Other RIRs may exhaust their IPv4 address space sooner or later, depending on their regional demands. However, IPv4 exhaustion is not a cause for panic--the Internet will not fail. In fact, you may hardly notice the event horizon. However, it is time to start planning your IPv4-to-IPv6 migration. Notably, John Curran, CEO of ARIN, notes that ARIN is starting to see service providers requesting IPv6 address blocks sooner and more frequently.

When an organization requests an address block, it has to justify the request and demonstrate that it is using the addresses that it has already been assigned. This helps to manage the available address pool more effectively. ARIN uses a number of methods to verify the justifications, but the process can take time. Curran says service providers are requesting new IPv6 blocks sooner in an attempt to have address space to meet demand for their existing and new customers.

The initial impact most organizations will see with IPv4 exhaustion is turning up new services on the Internet. As the address space is exhausted, organizations, service providers, hosting providers and cloud providers will have to develop strategies to support both IPv6 users and services and IPv4 users and services. Some of those strategies include running dual-stack IPv4 and IPv6 hosts and networks, performing IPv6-to-IPv4 translation, tunneling and the use of proxy servers to facilitate a migration to the new address space as software allows. One IT admin who wished to remain anonymous notes, "We have done this before with [Novell's] IPX to IP. It's not going to be a big deal. There will be bumps along the road. But we'll get over it."

While migrating from one addressing system to a new one may be old-hat to some, there is the potential for applications such as SIP, FTP, gaming protocols and peer-to-peer protocols (which use dynamic address and port assignments encapsulated in a the protocols control channel) to be adversely affected. The problem is that for complex protocols to survive a translation, some device, which is usually the nearest router, has to be protocol-aware and translate not only the IP addresses in the packets, but also the IP addresses residing inside the protocols. Waiting for these application-level gateways to be developed could cause service disruptions while service providers transition to IPv6 and equipment vendors catch up with software updates.

On the server side, new services that are IPv6-only won't be able to rely on the client side of the connection to perform IPv4-to-IPv6 translation since it will take time for service providers to bring up translation services. Running your own translation gateway ensures that users, whether on IPv4 or IPv6 networks, will be able to reach you, but you may loose valuable tracking information. And the translation gateway becomes a potential bottle neck.

Whether you are a large, multinational enterprise or a small to midsize business, you will have to migrate to IPv6. The migration will take time as existing network devices like application servers, printers, network cameras, VoIP phones--anything IP connected--is upgraded to support IPv6. If you start planning now to ensure that IT equipment and services are either supporting IPv6 or that your vendors have a well-defined IPv6 support roadmap ready, you can ensure the inevitable transition will be smooth.

You can read more about IPv6 with Network Computing Pro Reports Strategy Session: IPv6 [subscription required].

Carnival Cruise Lines has swapped out aging EMC Symmetrix storage (and Sun servers) for Dell EqualLogic SAN storage and PowerEdge servers, racking up savings of 7,000 hours per year for its shipboard IT staff. Symmetrix is a great platform, says Carnival's Doug Eney, VP of IS engineering, but it was time to upgrade and EMC's prices were too high. "We also looked at Clariion [servers], but didn't feel it was as resilient as we need. EqualLogic was able to meet our needs." Other storage companies that were evaluated included LeftHand, 3Par, NetApp and Compellent.

"We want to make sure things are smooth-sailing, and a calm IT environment aboard ship is critical," he says. "When something goes wrong at sea, you can't call on EMC or Dell technicians to fix it." Reliability and maintainability are essential, adds Eney, and everything is tested rigorously before it is put on ship.

Dell's storage arrays and servers have been deployed on the ships and Carnival's shore-side data center in Miami, Fla. The company has 22 cruise ships, 19 of which have been upgraded, with the remaining three due to be fitted out this year.

The time savings was only one of the benefits of the upgrade, says Eney. Carnival was able to reduce the number of servers--typically anywhere from 13 to 22 physical servers per ship--down to just two PowerEdge servers. Initially they deployed high-end four-socket PowerEdge R900 servers, but the workload could be supported on less-costly two-socket Dell platforms such as the PowerEdge R710, Eney says.

The proof of concept convinced the company that the EqualLogic PS6010XVS array--which hosts hot data on solid-state drives while pushing cold data, the least-used data, to the slower spinning disks--would save 60 to 80 percent compared with other high-end enterprise storage options under consideration.

"It's more than exceeded our expectations ... at a fraction of the cost of upgrading to Solaris and Symmetrix," says Eney. "We want to provide a great vacation experience, but at the same time we have to get cost out."

Dell is the revenue and unit market share leader for iSCSI storage, and is now perceived by IT professionals as the iSCSI storage market leader, according to the recent IT Brand Pulse iSCSI storage brand leader survey. According to Dell and IDC, it accounts for 33.8 percent of iSCSI storage revenue, more than its two closest rivals (EMC and HP) combined. Over the last six quarters, EqualLogic has posted year-to-year growth rates ranging from 33 to 78 percent.

Dell is also taking over Carnival's data warehouse, says Eney. The cruise line has been adding more functionality to the data warehouse, but its existing implementation either couldn't do the work or was cost-prohibitive.

Three years ago it was 300GBytes and now it's more than 20TBytes. Currently running the warehouse on both platforms, it will cut over to Dell on Feb. 11. "We're actually lowering our costs; and maintenance annuities we're just washing away."

See more on this topic by subscribing to Network Computing Pro Reports Informed CIO: Data Center Consolidation

The Greenplum division of storage vendor EMC is offering a Free Community License version of its EMC Greenplum Database software, which allows software developers to build new applications to deal with the explosion of so-called "big data" that businesses and other enterprises have to try to manage. The community license is based on code from Greenplum's massive parallel processing (MPP) database product, and includes the open-source MADlib library of analytic algorithms and Alpine Miner, a data mining modeling tool.

As companies build databases of ever-expanding amounts of data, they need more tools to analyze it and make business decisions based on those findings. Eventually, the databases hit a limit on how much they can scale, says Luke Lonergan, chief technology officer and VP of EMC Data Computing Products Division and co-founder of Greenplum, which EMC acquired in July 2010.

Lonergan gave an example of a company that introduces a new product that quickly becomes popular and all of a sudden they've got 1 million visitors to their site within a month or two. "What does an operation do when they get hit by the scale truck?" Lonergan asks.

Big data applications require "scale-out" technology, he says, which keeps up with demand as enterprises add more servers and storage hardware, and need database analytics software that keeps up with the data. The community license is to be used only for research; a commercial license is required to deploy an application in production or for commercial purposes. Greenplum's commercial- and community-licensed database software is based on the open-source PostgreSQL database software project, to which Greenplum has been a contributor.

The MADlib library offers tools that provide mathematical, statistical and machine learning methods for structured and unstructured data. MAD stands for "magnetic, agile and deep." Alpine Miner is a visual data mining tool from a company that Greenplum incubated within its own company, Lonergan says. Its chief advantage is that it can run right in the database engine as opposed to a situation where a small amount of data is copied from the database and tested in a separate workstation, saving several steps in the modeling process.

"What we have assembled with Community Edition are best-of-breed tools that are the right categories here for building big data applications," Lonergan says.

Greenplum's database product is based on "massively parallel processing architecture," where data is partitioned into segments in different servers. It is called a "shared nothing" environment because there is no disk-sharing of data. Instead, all communications between servers are via a network connection. This is in contrast to "shared disk" or "shared everything" environments for online transaction processing such as Oracle or Microsoft SQL Server relational database systems.

News of the EMC Greenplum Community License release was announced at a database technology conference being held this week in Santa Clara, Calif.

See more on this topic by subscribing to Network Computing Pro Reports Research: 2010 State of Database Technology

Quest Software has announced a new version of its vRanger Data Protection Platform for virtualized systems that integrates backup and replication by incorporating the capabilities of the company's vReplicator. Other features in version 5.0 include a simplified, lower-cost version of the product aimed at the lower end of the small to midsize business market of 250 or fewer employees, with additional features included in a higher-end vRanger Pro package.

Other new features include native cataloging, which makes it easier for users to find and restore individual files, and improved speed and reduced network traffic through support for Fibre Channel.

Joe Wakeman, an information technology specialist with Webster City Community Schools, a K-12 public school system in Webster City, Iowa, says he was particularly interested in the combination of backup and replication products. "It saves me a lot of time," he says. "I only have to manage one product, rather than two that don't talk to each other."

The combined products, which Wakeman has been using since November, have also saved staff time in managing backups, he adds. The school system has about 1,600 students and 250 staff members, and he is responsible for 800 computers. The organization, which has been using vRanger for about three years, regularly backs up about 20 servers to an off-site location. An active bitmapping feature also lets him save as much as 50 to 60 percent of disk space.

The lines between backup and replication are blurring, with users setting up "tiered recovery" systems in the same way that they have set up "tiered storage" systems, where some data is more important than others, says John Maxwell, VP of data protection products for the Aliso Viejo, Calif., company. vRanger and vReplicator use the same binary code for their replication functionality, Maxwell says.

This summer, Quest expects to release version 5.2, which will incorporate a virtual appliance to the replication engine in a VMware server, followed by similar functionality for backup by next year, according to Maxwell. While Microsoft's Hyper-V hypervisor is on the roadmap, it is not a priority, he adds.

Other new features include file-level recovery for Linux (it was already available for Windows); support for Network File System, which means that network attached storage devices can be used as targets for backup and replication; and support for File Transfer Protocol repositories.

Barb Goldworm, president and chief analyst for Focus LLC, a Boulder, Colo., consultancy, says she was particularly interested in the cataloging feature, due to its potential for improving performance. The catalog adds value because it helps users recover files more quickly. It works by allowing users to store information about the files during the backup for fast recovery later, without slowing down the backup itself, says Goldworm.

The product is available now and has been in limited release, with 700 customers, for the past three weeks. vRanger Standard starts at $399, while vRanger Pro starts at $699. vReplicator will continue to be sold separately for the rest of the year for $299.

See more on this topic by subscribing to Network Computing Pro Reports Strategy: Storage Virtualization

VMware is rolling out new cloud consulting services that the virtualization vendor claims can get a private cloud running for a customer in 30 days or less. Also available, as of Monday, is a VMware vCloud reference architecture, which the company describes as a blueprint to help customers quickly create a vCloud solution that has been proven in other enterprise cloud deployments.

VMware says it is targeting these services at companies that understand the value of cloud computing but are unclear as to how to make it happen. "Everyone we've worked with that started out with preconceived notions dramatically changed their expectations of what cloud can do for them [after meeting with VMware]," says Matthew Stepanski, VP of VMware's Technical Services Cloud Practice.
 
The main new consulting offering is the vCloud Accelerator Service, which helps a customer plan, design, configure and deploy a private cloud, Stepanski says. The Accelerator Service brings together a variety of VMware services, including vSphere for managing virtualization in the data center, vCloud Director for pooling together virtual infrastructure resources, vCenter Chargeback for managing how to bill various lines of business for access to cloud services, and vShield for securing cloud environments.

VMware is just one of many IT companies going after the cloud computing market with various services and product offerings. HP, for instance, last week unveiled its HP Enterprise Cloud Services-Compute offering, which is designed to help customers deploy private clouds.

Acknowledging the competition, Stepanski believes VMware's legacy in virtualization gives it an advantage among others because virtualization is at the core of cloud computing. "You can have virtualization without cloud, but you cannot have cloud without virtualization," he said.

But like virtualization, in which companies initially deployed the technology in application testing and development environments before deploying it in more risky production environments, some customers are also deploying cloud computing cautiously and in limited areas. To address that, VMware offers a "controlled introduction," Stepanski says, even though VMware is confident its cloud environments could work in production environments easily.

VMware's vCloud Jumpstart service provides customers the opportunity to evaluate a non-production installation of the VMware vCloud solution. The company also shares with customers case studies of other cloud deployments that closely match the deployments that VMware thinks would work best for them.

At the company-hosted VMworld 2010 conference last summer in San Francisco, VMware CEO Paul Moritz said businesses are likely to first adopt private cloud computing within their own data centers before fully embracing public cloud computing, going outside for their computing needs from providers such as Amazon, Google and Rackspace. 

See more on this topic by subscribing to Network Computing Pro Reports Strategy: Private Cloud Blueprint

With almost 10,000 customers, including three-quarters of the Fortune 100, Fluke Networks' AirMagnet is releasing the latest version (9.0) of AirMagnet Enterprise, its wireless intrusion prevention system (WIPS) for managing and securing enterprise WLANs. Wireless communications is exploding, and security is a huge and growing problem. The company says that by the end of the year, 70 percent of all new worldwide voice and data client-to-LAN connections will be wireless, with 802.11n becoming the de facto standard (with 50 percent of unit shipment in 2011, 63 percent in 2012 and 75 percent by 2014, according to Infonetics).

Richard Stiennon, chief research analyst, IT-Harvest, says that Wi-Fi is becoming a required service for owners of retail and public places to attract customers. "It is becoming the primary way to offer guest access in offices and hotels. Along with ubiquity comes threats as attackers target individuals--think FireSheep--and networks over Wi-Fi."

The main enhancement to version 9.0 is automated threat protection technology that enables organizations to dynamically update and defend the network when new threats emerge, says Fluke. No other independent or integrated WIPS platform supports such an automated process with immediate deployment and threat database update (DTU) capabilities, according to the company. The capabilities are made possible by a new, extensible security event detection definition engine that enables rapid development and modification of new threat detection signatures.

In addition, AirMagnet Enterprise's policy signature database is now separately loadable and no longer embedded in the server software. DTU also allows zero-day protection from new threats via immediate and automatic installation of new signatures when available.

Other 9.0 changes include: enhanced rogue-on-wire detection that adds two new methods to increase detection accuracy of rogue access points and speed, as well as reduces configuration complexity; increased accuracy of device location and more flexible sensor density deployment with Cisco AP data; PostgreSQL open-source database support; and expanded Microsoft server/database version support.

Set for availability on Feb. 7, a complete bundle, including sensors, server and console software starts at $10,000. The AirMagnet Enterprise Server base license lists for $5,995, while the AirMagnet Enterprise Sensor, 802.11n and dedicated Spectrum Analysis radio lists for $1,395. The software upgrade is free to existing customers under contract.

Stiennon sees the two biggest strengths of AirMagnet 9.0 as the incremental updates that decouple the WIPS engine from signatures and the enhanced sensing of rogue access points over the wired network. "This (update) avoids downtime as new configurations are loaded to block new attacks, very similar to the way AV and IPS for wired networks work." As for the enhanced sensing of rogue access points, any sensor on a LAN segment can pick up the telltale signatures that an unauthorized Wi-Fi access point exhibits.

The competitive landscape is of course dominated by Cisco, with Air Defense/Motorola being another stand-alone WIPS, says Stiennon. "Airmagnet continues to offer the differentiating spectrum analysis capability in addition to these new features. Spectrum analysis is a great debugging tool for sensor and access point placement, as well as detection of other strange radio-wave activity such as 3G."

See more on this topic by subscribing to Network Computing Pro Reports Research: Wireless Nation 2011

Targeting the midmarket, DataCore Software is addressing what it calls the biggest barrier to widespread adoption of virtualization: storage. The infrastructure ISV says that its SANsymphony-V storage virtualization software enables data centers to use existing equipment and conventional storage devices, instead of the rip-and-replace approaches being proposed to support desktop and server virtualization projects.

Two years in development, the software uses adaptive caching and performance-boosting techniques to absorb wildly variable workloads while simultaneously removing storage as a single point of failure and disruption. Available in five models, software licenses for a fully redundant, high-availability configuration start at less than $10,000, including annual 24-by-7 technical support.

The company's objective is to be the logical third dimension of a virtualization strategy by combining the first two dimensions, server and desktop virtualization, with the virtualization of storage assets. The business benefits of a stable software infrastructure spanning multiple models and manufacturers across generations of hardware are as compelling for servers and desktops as they are for storage, states DataCore.

By decoupling the virtual infrastructure from the underlying disks, SANsymphony-V frees customers from hardware vendor lock. And by instituting best practices through the use of automation and guided workflows, it reduces complexity and operational costs. The benefits include faster application response by auto-tuning "mega caches" and selecting the best I/O paths, high availability through physical separation with full auto-failover and recovery capabilities, optimal disk space utilization through thin provisioning and pooling, and continuous data protection to rapidly rollback in time and recover workloads and virtual machines.

While somewhat dismissive of the "purveyors" of server hypervisor technology (saying they are "peddling fluff"), Jon Toigo, CEO and managing principal of Toigo Partners International LLC, gives good marks to DataCore's SANsymphony-V. He says the software provides a layer of functionality over the block storage infrastructure that simplifies the three main tasks of storage administration as conceived today: capacity management, performance management and data protection management.

For capacity management, SANsymphony-V provides a way to virtualize all block storage capacity (Fibre Channel-attached and iSCSI-attached) into a pool that can be easily connected to servers. Then, once virtual volumes are associated with servers, SANsymphony-V allocates actual physical storage to the server's virtual disk on an as-needed basis, watching the size of the capacity pool and notifying administrators well before any shortfalls arise. "That's thin provisioning, and it is provided much more efficiently on the storage virtualization layer than it is on an individual array controller."

The software also helps to optimize the performance of the virtualized storage infrastructure, by caching all reads and writes to the physical storage, but also by managing and load balancing traffic across available traffic paths. "Most of this is done behind the scenes, and we saw a 3-times improvement in storage performance after it was virtualized. That means you get brand-name storage performance out of simple JBODs [just a bunch of disks]. That could translate into significant cost-savings to companies by eliminating the need to buy more expensive rigs and by breaking vendor lock-ins."

For the third task, data protection management, Toigo says SANsymphony-V provides rock-solid protection for data at three layers. "By ticking a checkbox next to a virtual drive, you can initiate an I/O logging function that captures every single write made to the volume. You can use this log to recover back to any point in time--as granular as a write function-- which provides outstanding protection at the data layer."

Overall, Toigo says this product is crucial for anyone who is thinking about virtual servers because it fixes most of the I/O and storage mapping problems that arise in server virtualization efforts but are rarely addressed by the initial server virtualization strategy. "It will also be key as more companies embrace desktop virtualization. ... The market potential for VDI [virtual desktop infrastructure] is significantly greater than server virtualization--400 million PCs waiting to be virtualized--but to do so, the storage infrastructure must be affordable, scalable and resilient. With SANsymphony, I think you get all three."

See more on this topic by subscribing to Network Computing Pro Reports Strategy: Storage Virtualization

Embotics has announced version 3.7 of its V-Commander virtualization management software, which includes integration with VMsphere for improved configuration management and automated remediation of non-compliant virtual machines (VMs) that do not have the appropriate types of metadata defined. Other new features include customizable self-service VM request forms, and enhanced capacity and performance management capabilities, including bottleneck and rightsizing reports.

The company had said last August that it planned to ship a version supporting Microsoft's Hyper-V by the end of 2010, but while it has developed that software internally, it is not expected to ship until the first half of this year.

The initial plan was for Embotics to offer general availability of support for Microsoft's Hyper-V by last year, but based on customer requirements, the company decided to invest in capacity features instead and defer release of Hyper-V support, says Jason Cowie, VP of product management for the Ottawa, Canada, company. Embotics had believed 2011 would be the year of mass adoption of Hyper-V, but now believes that will be deferred to 2012, he says.

The VMsphere integration, provided through a VMware Infrastructure Client, means that vSphere users can launch V-Commander in context, so they no longer need to go into VMware Virtual Center (now known as VMware vCenter Server) to perform configuration functions, Cowie says. Instead, the company has been able to marry configuration and change management into Virtual Center using the same interface, he says.

Capacity management reports help prevent the "virtual sprawl" that can happen in virtualization environments. This virtual sprawl makes it difficult for an organization to track where its assets are, according to Cowie. In many organizations, for as many as 10 to 20 percent of the VMs, no one has a good understanding of where they are, their current state and whom their owners are. Cowie related the story of one organization that discovered it had $1 million in virtualization assets.

Right-sizing reports include information such as summary, storage, memory and CPU, while idle VM reports show information such as low CPU, disk I/O and network I/O. Bottleneck reports include data such as CPU, memory, disk and network.

"Embotics has done a nice job," says Chris Wolf, research VP for Gartner, a Stamford, Conn., consultancy. "Infrastructure as a service is highly complex, with numerous dependencies such as capacity management and lifecycle management. Embotics' integrated solution is attractive to customers because it can address those management dependencies under a single umbrella. Embotics had been a bit ahead of its time, and virtualization management requirements are now starting to catch up to the solution offered by Embotics."

The new version of the software is available now for prices starting at $299 per year per host CPU socket, or $649 per socket for a perpetual license, Cowie says. The company's users are split fairly evenly between the two licensing schemes, he says. About 20 to 30 percent of Embotics' users are enterprise-sized, while 60 to 80 percent are midmarket-sized, he says. The company has about 30 employees.

See more on this topic by subscribing to Network Computing Pro Reports Hyper-V Takes on VMware

Symantec has released two major components of its appliance strategy for backup and storage. The NetBackup 5200 integrates backup, deduplication and storage on one box; the FileStore N8300 is designed for cloud-scale storage of unstructured data.

The 5200 option complements Symantec's existing NetBackup and PureDisk deduplication software, as well as the NetBackup 5000 and 5020 deduplication appliances announced earlier this year. The FileStore N8300 is a hardware version of its FileStore soft appliance. Symantec also offers SaaS-based backup and storage.

"The primary pain point for us has been our data retention period," said Tommy Meek, systems operations manager for CBeyond, an Internet communications provider for small- and medium-sized businesses. "The real issue was storage capacity. What was our option, purchase more disk or increase tape libraries?"

CBeyond, which just purchased two 5200 appliances (one each for primary and remote data centers), currently uses NetBackup as well as PureDisk deduplication software. Effective deduplication, he said, is a key. "With shorter retention, we ran a greater risk of not being able to recover critical data."

The NetBackup 5200 features built-in source and media deduplication and 32 TB of deduplicated storage per node. It can use either inline (before data is written to disk) or post-process (after written to disk) deduplication methods.

The 5200s, Meek said, will help CBeyond eliminate tape backups. "We want to eliminate the cost of tape maintenance and off-site storage completely," he said. The move to appliance-based backup, deduplication and storage was based on ease of implementation.

The FileStore N8300 is a NAS appliance built for cloud and large-capacity enterprise environments. "Enterprises need scale and performance for new workloads, such as digital media, cloud and virtualization," said Yogesh Agrawal, VP of Symantec's FileStore Product Group. "85 percent of data is unstructured. They need to manage growth and cost."

The N8300 can scale capacity up to 1.4 PB; Symantec claims it gains linear performance with the addition of each up to six nodes. It makes optimal use of disk space, avoiding fragmented "islands of data," Agrawal said, as it can scale independently in performance and storage capacity, so for example, enterprises do not have to purchase additional storage because their compute capacity can't support their existing environment.

The appliance includes on-board Symantec AV scanning and Storage Foundation data tiering technology, so organizations can make cost effective decisions on the type of storage that is used for particular data classifications. The N8300 fully integrates with NetBackup, as well as Symantec's Enterprise Vault data archiving product.

The 4-U NetBackup 5200 is priced at $59,995 per node. The FileStore N8300, also a 4-U appliance, costs $109,000 for a "typical" two node, 48 TB configuration.

GigaSpaces Technologies, which specializes in products that address scalability in the application stack, is now shipping eXtreme Application Platform (XAP) 8.0, which enables organizations to transition their existing infrastructure into more modern systems such as distributed architectures, virtualized environments and the cloud. The company says its Same Data, Any API capability promotes openness and interoperability, supporting all common interfaces for accessing data, such as Memcached, JPA, JMS, Document and the "highly efficient" native object-oriented API.

The ability to choose the best API for the use case at hand and operate on the same data, regardless of the APIs chosen, significantly decreases the learning curve associated with adopting the XAP technology. Also new with 8.0 is its support for continuous scaling, which allows for rolling upgrades, continuous deployment, quick introduction of complex features and real-time changes to the applications data model, business logic or even underlying platform with zero downtime.

Scalability has tended to be introduced as an afterthought, says GigaSpaces, with the philosophy that "we'll worry about it when we get there." With the emergence of cloud computing, scalability must be thought about from day one, and this situation has resulted in what the company calls the "scalability tsunami." Combine that with the recent trend of having to do more with less, and scalability becomes even more critical.

That's where the latest enhancements to XAP come in--the openness and scalability, multi
tenancy and improved efficiency. It deals with the challenge getting to better agility and reducing the time between developing a new feature and putting it into production, says the company. Version 8.0 also simplifies operations and administration with its elastic middleware capabilities that encompass the entire application stack, says GigaSpaces. By identifying problematic or erroneous situations and reporting them to the user, the new, simplified Alerting API enables real-time problem isolation and prevention. In addition, the new multisite deployment support makes deploying and managing geographically distributed applications easier.

"Through the introduction of JPA, Memcache, REST, etc. APIs support, XAP 8.0 becomes more appealing for mainstream developers that so far have not considered the product or similar products from competitors such as Oracle, VMware/Gemstone, IBM and Terracotta  due to the complexity and lock-in risk of the native XAP APIs," says analyst Massimo Pezzini, VP and Gartner Fellow. "Although other vendors support some standard APIs--for example, Oracle supports JPA for its Coherence product--none support Memcached."

The new functionality is significant, but the main challenge for XAP is adoption from system integrators, ISVs, cloud service providers and SaaS vendors, which may not be willing to adopt technology from a relatively small vendor like GigaSpaces, says Pezzini. According to the 10-year-old company, it has only 350 customers globally.

Pezzini thinks the key for XAP adoption could be Memcached support, something the other vendors don't provide. "This is an extremely popular open-source distributed caching platform--as an example, Facebook is a massive user of this technology. By supporting the Memcached APIs, GigaSpaces has now a powerful tool to up-sell to the Memcached installed base."

Another factor is that large software vendors (such as Oracle, Microsoft, IBM and VMware/SpringSource) see these technologies as key enablers for a variety of other strategies, such as BPM, ESB, application servers and cloud. XAP has been utilized primarily by financial services companies, especially to support trading types of applications, but the last three years have seen growing adoption in verticals such as Web commerce, online gaming and cloud/SaaS, as well as in commercial banking and other segments to support application scenarios such as mainframe offloading.

"This growing adoption, including increasingly by mainstream enterprises, is calling for a standardization. Although there is no credible standardization industry effort taking place, APIs like JPA, Memchached, Ehcache are rapidly emerging as de facto standards for DCPs [distributed caching platforms or memory data grids]", said Pezzini.

See more on this topic by subscribing to Network Computing Pro Reports Strategy: Calculating APM Costs