By James E. Gaskin

Moving is always a hassle, and moving data is no different. So when talking about data migration, get organized and test processes before committing.

Smaller businesses keep data on servers and workstations until something forces them to upgrade. “The triggering point for moving to shared storage is almost always virtualization,” says Bob Laliberte, senior analyst at the Enterprise Strategy Group (ESG) research firm. “You need networked storage in order to take advantage of server mobility features.”

Managing secure access when moving from server to network storage doesn’t have to be difficult. “If you’re moving from a Windows server to a network-attached storage [NAS] device running Windows Storage Server, your Active Directory tree may provide enough security,” says Greg Schulz, founder and senior analyst at The StorageIO Group. “Enable encryption on the NAS, at least, and this type of data migrationis almost painless.”

Moving from NAS to a storage area network (SAN) adds complexities. Because security and access controls can be stripped off in the transfer, “you need to double check your security and access-control settings when you move from a file-based NAS to a block-based SAN,” says Schulz.

The other major data migration target is a public or private cloud. In ESG’s most recent spending survey, for example, increasing virtualization and managing data growth were the top two topics. “[And] the cloud and SaaS were our two fastest climbers in importance between 2010 and 2011,” says Laliberte.

That doesn’t mean moving data skyward is easy. “There are big issues getting data up to a cloud,” says Schulz. “If you don’t have too much data, you can transfer it over the network. But for large amounts, you’ll need to transfer it to disks, and then send those disks to your cloud provider.”

The same advice is true when the data comes back down, says Laliberte. “Getting a single file from a backup is easy. Getting a server’s worth of data may take four days over the network, so cloud providers will send your data on a hard disk. The best data migrationmedium for large amounts of data may still be FedEx.”

Moving data to a cloud is more complicated than from a server to networked storage. “Do a pilot,” says Schulz. “Many times, especially with a cloud, I have to copy data to the cloud and verify the data before deleting the network version.” Schulz’s advice? “Move to cloud storage in steps. Start a cloud backup project and copy local data to the cloud. When up and running, you’ve migrated your data.”

Luckily, migrating data from cloud service to cloud service is rare. “Changing cloud providers is a royal pain,” says Schulz.

Source:  ChannelPro SMB -http://www.channelprosmb.com/article/28170/Data-Migration-Best-Practices/?highlight=data+migration

Cloud functionality comes about by deploying a set of smart and compatible virtualization, storage, security, and infrastructure automation technologies. Working together, they implement six foundational delivery principles:

1. Secure separation ensures the physical or logical separation of basic services such as servers and storage in use by customers, or “tenants.”

2. Service assurance automates provisioning of services and enables reliable performance to tunable service-level agreements.

3. Service provider in control supports the management of the cloud’s operations, including integration of provisioning, support, billing, and reporting systems.

4. Tenant in control supports the customer’s management of on-demand services, access to support, and transparency into utilization and billing.

5. Security and compliance employs identity and access management, encryption and data loss prevention, enterprise key management, and other security methods to protect the tenants’ cloud environments.

6. Data protection employs backup, recovery, replication, and data redundancy methods to ensure the tenants’ business continuity.

The first four principles enable multi-tenancy, the flexible yet secure sharing of technology resources in a cloud. The last two focus directly on the protection of the data and applications hosted there for customers.

Cloud configuration is an overlapping three-phase process. First, gain the efficiency and often dramatic cost savings of virtualizing technology resources, from servers and storage to the applications, user interfaces, and other business services being hosted. Second, maintain control through automation of access and security and the overall operation and management of the virtualized environment. Third, provide customers with choice by provisioning a variety of business services, many of them offered in partnership with third parties.


The effort of cloud configuration depends, of course, on where the company is today. All of the following are steps toward successfully implementing a cloud:

• Consolidating servers, storage, networks, and other technology resources

• Virtualizing technology resources, including information and applications

• Automating technology resource and security management

• Organizing and provisioning IT offerings as business services

• Structuring and managing IT as a shared services organization

• Building standard interfaces with compatible service providers

• Making effective use of selected public cloud services

Cloud computing is a natural extension of all these improvement initiatives, as well as the architecture for integrating them.  Cloud is the means of provisioning a comprehensive range of technology-based services with much the same efficiency and discipline that is demonstrated in provisioning a variety of network services.

Data deduplication is becoming known for being the smart way to compress data.  Data deduplication is a data storage method of reducing data storage needs by eliminating redundant data.  Only a single instance of the data is actually retained on the data storage media, such as disk or tape.  Redundant data is replaced with a pointer to the unique data copy.

Data deduplication offers the great benefits of low storage space requirements, which will save money on storage expenses and make for longer disk retention periods.  These benefits provide better recovery time objectives (RTO) for the organization.

Before an organization embraces data deduplication as a process, here are four simple, but smart tips for smarter data deduplication:

1. The integrity of the data being deduped is far more important on the data storage device compared to data stored on any server because the data storage device will hold the only copy of data.  Implement quality data storage with redundant hard drives and power supplies, such as EMC Data Domain or EMC VNXe.
2. Prepare for growth today by procuring a data storage solution that is scalable or easily upgradable.  Dedupe requires more overhead than traditional servers and data loss should be a major concern for administrators.
3. Stay on top of the dedupe process by being knowledgeable of resources.  Having trained personnel in-house or contracted is essential for any data storage process.
4. Test regularly for data integrity from your data storage.

Here’s seven great examples of how small businesses are using private clouds, an approach once only considered by large enterprises:
Applications updates: A regional specialty paper goods manufacturer found the software to accurately keep up with constantly changing sales tax rates and rules in the half-dozen states where the company sells products. The vendor keeps the software up-to-date, and the company can expand to new states effortlessly.

Customer collaboration: An accounting firm was expanding its CFO services to its SMB client base, but accessing and dealing with the wide variety of clients’ financial applications was proving infeasible. The answer was a cloud-based portal for ongoing communication and realtime financial updates. “CFO outsourcing” quickly grew to be over half the firm’s business.

Business continuity: After thieves stole a recycling company’s entire office computing setup of eight PCs, the company bought eight more the next morning and was back in business by noon. Information, applications, access—nothing had to be recreated because it was all running and backed up in the cloud.

Business communications: To compete with much larger firms, a risk management consultancy needed to use excellent communication and collaboration tools. A cloud-based suite—from messaging and e-mail to shared workspaces and online meetings—provides on-demand capability while reducing technology costs by thousands of dollars a year.

Business scalability: When a workforce scheduling and management firm moved to a hosted application for its sales force, it anticipated a 20 percent reduction in associated technology, training, and personnel costs. What it also got was steady IT costs even while the firm tripled its business.

A fresh start: When one of its products surged in popularity, a provider of multi-player games for mobile phones “hit a wall” when its own modest server facility was overwhelmed. The solution—in terms of revenue generation, customer satisfaction, and technology—was rapid conversion to pay-by-use cloud infrastructure.

Start-up: A company running online marketplaces for artisans and their products faced three challenges—geographic reach (the 10 employees are scattered across the U.S., and they are bringing local artisans to a national market), image storage (hundreds of thousands of images of arts and crafts), and payment processing (many relatively small payments called for automation, not clerks). For a company like this, cloud infrastructure and applications was by far the best way to start and grow a business while minimizing the co

There’s always been a fundamental tension in information technology management: we want the computing environment to be robust and secure, but we also want it to be flexible and accessible. With cloud computing, it’s now possible to have it both ways. Cloud architecture represents a better way to configure and manage IT resources of all kinds, from servers and storage, to information and applications, to productivity tools and user interfaces.

Technology assets and their capabilities are well-defined, modular, and connectable. Interface methods are standardized and published. Virtualization enables physical devices to be efficiently and securely shared, and heterogeneous technologies to work together. All of these resources can be managed as an efficient and flexible pool, shared (as authorized) across the business, its customers, and its partners. Very importantly, these resources can also be consumed differently—as understandable business services that people can access on demand and often via self-service through a standard browser interface.

Businesses of all sizes enjoy more transparency into the services they consume and can often pay according to actual usage. The cloud approach enables better performance on multiple fronts simultaneously: cost, manageability, information access, new capability deployment, coordination and collaboration, business continuity and security, business innovation and growth. It’s the best and smartest way for companies of all sizes to get the best mix of capabilities, performance, and cost.

EMC, the global leader of data storage and management, guarantees 25% more efficiency for all EMC VNX series unified data storage arrays compared to competitive offers.  Here are the simple details:

• Based on out-of-the-box best practices
• No need to compromise availability to achieve efficiency
• No caveats on types of data you must use
• No need to auto-delete snapshots to get results

If EMC data storage is not more efficient, we’ll match the shortfall.  Here’s why we’re so confident in EMC:

When EMC Corp. rolled out its VNX and VNXe storage platform this year, it shone a bright light on unified storage systems and intensified its competition with NetApp.

NetApp and EMC aren’t the only vendors to offer unified storage — most major vendors have systems that support multiple protocols — but NetApp’s FAS and EMC’s VNX are the most talked about. They’re also the most scrutinized and serve as the models that most people refer to when they discuss what is and isn’t true about unified storage platforms.

Unified storage systems allow one-box access to both block and file storage, typically over Fibre Channel (FC) or Ethernet networks. NetApp first offered multiprotocol storage in 2002 when it added SAN or block capability to its NAS filer platform. All NetApp FAS systems from entry level to enterprise support file storage via CIFS and NFS and block storage over Fibre Channel and iSCSI.

EMC merged its midrange Clariion SAN and Celerra NAS platforms into the VNX. EMC has a separate enterprise SAN Symmetrix family but the VNX put the two largest pure-play storage system vendors into a head-to-head battle with systems that cover most of the overall data storage market.

Let’s take a look under the hood to see how the NetApp and EMC unified storage products compare.

NetApp FAS series

NetApp’s FAS scales from the FAS2020 with 68 disk drives to the FAS6280 with 1,440 drives. Each system comes with NetApp’s Data Ontap 8.0 operating system (OS) that includes all NetApp’s advanced data services, including compression, data deduplication, disaster recovery and snapshot capabilities.

The FAS2000 series includes the 2U 2020, the 2U 2040 and the 4U 2050 targeted at mid-sized companies, remote and branch offices, and large enterprise departments. The series’ maximum supported drives range from 68 to 136 SAS, FC or SATA drives, and includes two PCI Express (PCIe) expansion slots.

NetApp’s FAS3100 series are all 6U chassis with eight PCIe expansion slots and 64-bit controller architectures. They’re targeted at enterprise applications and technical computing. The 3100 line supports between 420 and 840 SAS, FC and SATA drives.

The FAS3200 product line features a wide range of speeds and feeds. Each 3200 product includes four PCIe expansion slots; in addition, the 3240 and 3270 can add eight more PCIe expansion slots for greater I/O capabilities. Without the additional expansion slots, each unit is a 3U form-factor and supports from 240 to 960 drives.

NetApp’s largest FAS product line is the 6200 for large enterprises. The 6210 is a single 6U unit with two controllers, 48 GB of RAM, a 3 TB Flash Cache, eight 10 Gigabit Ethernet (GbE) connectors and eight PCIe expansion slots. It holds up to 1,200 drives. The 6240 and 6280 are each dual-enclosure 12U units, hold up to 1,440 drives and include eight 10 GbE connectors and 24 PCIe expansion slots. The 6240 comes with 96 GB of RAM and 6 TB Flash Cache. The 6280 has 192 GB of RAM and an 8 TB Flash Cache.

NetApp also has its V-Series, which virtualizes storage from other vendors, including EMC, to add file capability.

EMC VNX

The EMC VNX product line comes in three flavors: VNXe for small- and medium-sized businesses (SMBs), VNX for mid-sized organizations and enterprises, and the VNX Series Gateway (VG) that provide NAS capability for EMC SANs. All the products in the VNXe, VNX and VG product lines use EMC’s Unisphere unified management software.

Both the VNXe3100 and VNXe3300 include up to two block-storage processors, file deduplication, compression, thin provisioning, and snapshots. The 2U VNXe3100 holds from six to a maximum of 96 SAS or Nearline-SAS (NL-SAS) drives. The 3U VNXe3300 includes up to 120 SAS or NL-SAS drives.

All the VNX series products support Flash, SAS and NL-SAS drives, as well as two block storage processors, virtual provisioning and EMC’s SAN Copy data copying technology. The VNX5100 includes up to 75 drives, the VNX5300 holds up to 125 drives and the VNX5500 holds up to 250 drives; the VNX5700 holds up to 500 drives, while the VNX7500 holds up to 1,000 drives.

The VNX Series Gateway provide file access to existing VNX, Symmetrix and Clariion storage systems. Both the VNX VG2 and VG8 support file deduplication and compression, virtual provisioning and EMC’s SnapSure snapshot technology. The VNX VG2 supports up to two X-blades and 256 TB of total usable capacity. The VG8 supports up to eight X-blades and 1.8 PB of usable capacity.

While NetApp’s FAS systems run DataOntap, VNX runs the Clariion Flare and Celerra DART operating systems although EMC hasn’t spelled out exactly how the two OSes co-exist in one box.

Commonalities include single-pane-of-glass management

NetApp and EMC also provide single-pane-of-glass storage management for their unified solutions. Analysts say a single management interface is one of the main attributes that makes a true unified system.

“Whenever people talk about multiprotocol, the first things they think about is management overhead and consistency of utilization across the protocols,” said Ashish Nadkarni, a senior analyst and consultant at Taneja Group. “Am I going to be sitting around with five interfaces all utilizing different physical components of the same system?”

EMC’s Unisphere supports the VNXe and VNX product lines, as well as EMC’s Clariion, Celerra products and RecoverPoint/SE replication product. Unisphere’s single sign-on web interface works in both physical and virtual environments and plugs into an existing environment without modification.

NetApp’s OnCommand management software includes System Manager for single storage systems and Operations Manager for multiple NetApp systems. OnCommand controls, automates and analyzes storage infrastructure. NetApp also provides Virtual Storage Console (VSC) plug-ins for managing FAS systems within VMware Inc.’s vCenter management platform.

“We’re not hell bent on having administrators come to our console to manage storage,” said Chris Cummings, NetApp’s vice president of product and solutions marketing.

When EMC launched the VNX family, NetApp executives said it wasn’t a true unified storage platform because it runs the Celerra and Clariion software stacks separately on the Intel Corp. Xeon quad-core processors. But Taneja Group’s Nadkarni said that‘s not much of an issue for customers. “NetApp started from a single code basis and branched out into other areas,” he said. “EMC is now started from the other side and is coming to a converged solution. From a customer’s prospective, it’s not that relevant anymore.”

Different philosophies on storage tiering

EMC and NetApp also take a different approach to storage tiering across their product lines.

EMC leverages solid-state drives (SSDs) and Fully Automated Storage Tiering (FAST) software for automated tiering while NetApp creates what it calls a virtual storage layer.

FAST is policy-based, application-aware software that moves data between SSDs, Fibre Channel and SATA drives. It moves frequently used data to higher-performing drives, such as SSDs and FC drives, according to user-defined policies. FAST can run automated or in user-approval mode; it works at the sub-LUN level by moving volumes across disks.

“Automated tiering is a key capability to allow customers to not only dynamically meet their performance needs but to do so in a cost-effective manner,” said Jon Siegal, director of product marketing for EMC’s unified storage division.

While EMC’s approach to tiering is the traditional one taken by storage vendors, NetApp approaches it differently. NetApp’s virtual storage tiering does not use tiering software as EMC and others do. NetApp has a Flash Cache PCIe-based solid-state storage device that speeds performance on specific applications, such as data warehousing. It also supports SSDs in the array as other storage vendors do, but recommends that only for customers who need all reads and writes to be fast instead of peak performance on specific applications.

NetApp’s Cummings said Flash Cache identifies hot data at the block level in real-time and uses the virtual storage tier to move additional resources to support that data. He said that avoids having to move data around, which could create potential backup and file deletion problems.

“We look at this virtual storage tier and say it’s simpler, cheaper and you don’t run into these types of problems,” Cummings said.

Clustering unified storage

Unified storage systems haven’t been known for their clustering and high-availability capabilities. NetApp allows you to cluster your multiprotocol system in pairs using a global namespace at implementation for business continuity, performance needs and load balancing, while the VNX series can run in a two-node mode.

The NetApp clusters can be separated geographically depending on your networking capabilities. During implementation you indicate whether you want to scale up your system or scale it out with clusters. Then you define your clusters. “So you can scale it up or scale it out,” Cummings said. “Running scale out in a unified [system] is going to add a bunch more value to customers.”

The VNX series can run with two nodes for mirroring. EMC’s Siegal said the VNX family is built as a scale-up architecture for now, and pointed customers to EMC’s recent Isilon Systems Inc. acquisition for a scale-out model.

While EMC’s and NetApp’s code base and auto-tiering technologies are markedly different, neither has been proven to be better than the other.

“Don’t worry about splitting those hairs,” said Greg Schulz, founder and senior analyst at StorageIO Group. “Focus on your needs when choosing a multiprotocol system. How is the performance? How is it managed? How’s the availability?” He advises people considering the systems to compare the technology differences to their specific needs to decide which way to go.

- written by Todd Erickson, News and Features Writer, TechTarget

One doesn’t have to look long to find out why an ever-escalating number of companies recovering from the recession are turning to virtualization: cost. Instead of one server running one instance of an operating system, now several such instances — or even several operating systems — can be launched that act as complete and whole servers unto themselves. To the hosting machine, each virtual machine appears like an application, and the base hardware is powerful enough to run several of these applications at the same time.In addition to its obvious use, many more clever ways have been found to utilize virtual machines. They are making it possible for people with tablet computers to access their desktops — meaning work can be performed in real-time from literally anywhere. Virtual machines can mimic physical servers, be recovered from backup and then be restored to the same or even different hardware. They are contributing greatly to the rock-bottom prices now being offered by hosting companies, and are making cloud computing extremely cost-effective. It is certain that many more innovations are headed our way as well.

But as companies continue to virtualize, speed must be maintained between hardware hosts and virtual machines. Virtual machines activities must also be coordinated.

A critical factor to access speed is I/O reads and writes — a factor so important that if not properly addressed can create considerable and widespread problems. Fragmentation, a condition in which files are split into thousands or tens of thousands of pieces (fragments), causes many additional I/Os to be required in simple reading and writing of those files. Performance and hardware reliability are seriously hampered by fragmentation, and I/O bottlenecks occur as a result as well.

Instead of the single layer of hardware-based systems, an I/O request must pass through multiple layers in a virtual environment. Hence, fragmentation has a more profound effect than it used to, and can even hinder the ability to launch and run further virtual machines.

Once the solution to fragmentation, traditional defragmentation won’t fully address I/O issues in virtual environments. Beyond fragmentation, I/Os need to be prioritized, coordinating shared I/O resources. Another condition peculiar to virtual environments that results from fragmentation is virtual disk “bloat.” This condition occurs when virtual disks are set to dynamically grow but don’t then shrink when users or applications remove data.

The solution is virtual platform disk optimization technology which, fortunately, is now considerably advanced and available. This solution prevents a majority of fragmentation before it even happens, resources are fully coordinated, and a compaction feature makes it possible to eliminate virtual disk “bloat.”

To truly realize the considerable cost savings offered by virtualization, all enterprises should ensure their virtual platforms are properly optimized.

SOURCE: Bruce Boyers Marketing Services

All businesses have at least one thing in common: they create data.  In fact, most corporations double the amount of data created each year. This level of data growth presents the challenge of protecting that data from disasters and human error, efficiently managing it during its lifecycle and storing for compliance with regulatory requirements.

For most organizations, the operational, legal and compliance challenges of email necessitate going beyond simple backups. Increases in data leads to increases in costs – especially regarding time and personnel.

The real danger beyond costs then comes when a company treats backup as a single solution for both data protection and data retention, resulting in highly ineffective and inefficient data management. This is where the fundamental difference between data backups and data archives comes to light, which is simple: Speed.

Data backup stores information that can be rapidly accessed, whereas data archiving is used to store important data that’s not used on a regular basis but still needs to be retained and managed for business purposes (think, emails for compliance).  Data backups are intended for disaster recovery, for high-speed copy and restore to minimize the impact of failures, human error or disaster.

Data archiving is for discovery, to effectively manage data for retention and long-term access and retrieval. With archiving, the ability to search for information by keywords and metadata trumps recovery speeds.

Mixing data backups up with archiving will negatively impact recovery time (RTO) and recovery point objectives (RPO).  Same is true for using backups as archives.  Imagine how long it would take to sort through 2 years of emails just to find an email from several customer orders.

To combat this issue, the two capabilities can be applied together in a unified platform to optimize the cost and improve the overall effectiveness of data management. Data backup is more efficient in environments that have an effective archiving solution, and archives leverage the data backup for its own data protection needs.

Several vendors offer solutions that do a fantastic job of handling data backup and archiving, such as EMC’s Data Domain and EMC’s VNX unified platforms.  These data management solutions will make a data administrator’s life easier by streamlining both processes, decreasing recovery time, increasing data protection and optimizing recovery points.

The World's Simplest Data Storage