NetApp Data ONTAP 8.1: Efficient, Flexible, Scale-out NAS and SAN Storage

Introduction

Background

The demands on and for storage are increasing rapidly. The challenge to effectively control how vast amounts of data are created, stored, and accessed is intensifying. To address data growth without interrupting business operations, rapid deployment of storage and IT resources to meet increasing demand becomes a function of scalability.  The amount of data businesses need to store is skyrocketing, which drives corresponding growth in overall data storage costs in the form of storage systems, floor space, power, cooling, and the people required to manage it all. With IT under constant pressure to find ways to reduce costs, taking a long hard look at the storage environment makes sense. Managing data growth was cited as a top priority by one quarter of those IT managers surveyed in ESG's 2011 spending intentions survey, making it the number two priority for IT managers and only slightly behind increasing the use of server virtualization.[1]

IT is starting to feel the pressure of the sheer amount of data it has to store. "Spreadsheet management systems" cannot track what data lives on which LUNs when there are tens or hundreds of thousands of LUNs in the networked storage environment. Separate storage growth forecasts based on block and file protocols are not useful when, at the end of the day, all anyone wants is storage capacity to use when and where they need it. What's more, standalone silos with sub-50% utilization rates create too much waste in terms of management, floor space, power, and cooling.

Server virtualization is also driving more complexity in the storage environment-disk LUNs need to be mapped to virtual machines and administrators need to carefully monitor how many virtual systems are sharing a single LUN if they are to avoid performance bottlenecks.

NetApp Data ONTAP 8.1

NetApp is addressing these challenges with its Data ONTAP 8.1, designed to deliver an on-demand, highly efficient, flexible, and scalable single storage operating system to help users manage data, application, and scale-out storage infrastructure growth. Data ONTAP 8.1 operating in Cluster Mode takes multiple NetApp storage systems and creates a massively scalable unified storage platform to provide increased operational flexibility and efficiency with a goal of eliminating planned and unplanned downtime for both enterprise and service-oriented infrastructures.

Data ONTAP 8.1 provides scale-out storage that scales to petabytes of capacity and gigabytes per second of throughput using any combination of NetApp FAS or V-Series storage systems, configured in pairs for high availability. With Data ONTAP 8.1, NetApp provides a host of enterprise class capabilities that help storage administrators effectively manage their increasingly large and complex storage environments.

• Always-on Scalability, Availability, and Resource Balancing. As storage nodes are added to the system, all physical resources-CPUs, cache memory, network IO bandwidth, and disk IO bandwidth-can be easily kept in balance. Data ONTAP 8.1 systems enable users to add storage (Up to 11 PB in a four node cluster, up to 50 PB in a 24 node cluster) and move data between storage controllers and tiers of storage without disrupting users and applications. This enables administrators to increase capacity while balancing workloads, and can reduce or eliminate storage IO hot spots without the need to remount shares, modify client settings, or stop running applications.
• Unified Storage Efficiency. NetApp provides storage efficiency technologies for production and backup datasets that include block-level data deduplication, compression, and thin provisioning. These technologies can be deployed individually and in combination for both SAN and NAS, allowing customers to reduce the capital costs associated with storage.
• Multi-vendor Virtualization. Data ONTAP 8.1 fully supports NetApp V-Series Open Storage Controllers to virtualize traditional third-party storage arrays and incorporate their storage capacity into a Data ONTAP 8.1 cluster.
• Secure Multi Tenancy. Using Vserver, Data ONTAP 8.1 provides secure, protected access to groups of servers and applications, allowing organizations to provide IT service consumers with dedicated administration, IP addresses, exports, storage objects, and namespace.
• Integrated Data Protection. NetApp provides on-disk snapshot backups using capacity and resource-efficient Snapshot technology. Customers can reduce recovery time objectives (RTOs) and improve recovery point objectives (RPOs) across all of the storage in their environment. NetApp and its partners-VMware, Citrix, CommVault, and others-have integrated their data protection technologies to deliver unified and streamlined business continuance models for consolidated and virtualized environments.
• FlexClone. Using NetApp hardware-accelerated provisioning technologies, customers can instantly create clones of production data sets and VMs in order to meet the requests of a dynamic infrastructure without requiring additional storage capacity. Clones can speed test and development, provide instant provisioning for virtual desktop and server environments, and increase storage utilization. This capability is integrated with a number of offerings from NetApp partners including Microsoft, VMware, Citrix, and SAP.
• Unified Management. NetApp OnCommand data management software offers effective, cost-efficient management of shared scale-out storage infrastructure to help organizations optimize utilization, meet SLAs, minimize risk, and boost performance. By offering a single system image across multiple storage nodes in a Data ONTAP 8.1 cluster, NetApp enables organizations to automate, virtualize, and manage service delivery and SLAs through policy-based provisioning and protection.

ESG Lab Validation

ESG Lab performed hands-on evaluation and testing of Data ONTAP 8.1 in a NetApp facility in Research Triangle Park, North Carolina. Testing was executed with a focus on operational efficiency, on-demand flexibility, and always-on availability.

Getting Started - Operational Efficiency

ESG Lab began with a NetApp Data ONTAP 8.1 Cluster consisting of four NetApp storage controllers as shown in Figure 3. Two FAS 3270 storage controllers were configured , configured in an HA system (an HA system is defined as an active-active dual controller configuration)with NetApp disk protected using RAID DP technology, while two V3270 open storage controllers, also configured in an HA system, were installed in front of a third-party modular array with 110 146 GB 15K RPM drives.[2]

The first step to providing services using NetApp Data ONTAP 8.1 is to create a Vserver. A Vserver is a secured, virtualized storage server with its own administration, IP addresses, exports, storage objects, and namespace independent of physical location. A  Vserver can exist on any node in the cluster that provides a physical resource-LUNs, volumes, and network interfaces. Figure 4 shows a simplified representation of a Vserver; physical nodes house FlexVols and the  Vserver presents volumes and LUNS out to client systems over virtual interfaces. It's important to note that volumes can be created from any node in the cluster and can be served to clients from any node in the cluster. A physical interface can host multiple virtual interfaces, serving up storage for multipleVServers.

ESG Lab Testing

ESG Lab launched the Vserver setup wizard from the NetApp OnCommand System Manager, which guides users through creating a new Vserver from start to finish, assigning storage capacity, enabling protocols, and configuring interfaces. As shown in Figure 5, two interfaces per controller were assigned to this Vserver, for a total of eight interfaces.

The Vserver was created with all protocols enabled: NFS, CIFS, iSCSI, Fibre Channel, and Fibre Channel over Ethernet (FCoE).

Next, a 30 GB iSCSI LUN was created using the Create LUN Wizard.

Finally the LUN was mounted by a virtual server using the Microsoft iSCSI initiator and the volume was formatted. Files and folders were copied to the new drive letter. The total elapsed time from the first mouse click in the NetApp System manager to a virtual server utilizing storage from the cluster was less than ten minutes.

Next, ESG Lab created a snapshot of the volume using NetApp SnapDrive, and then deleted the files and folders created in the previous step, as seen in Figure 8. The Windows recycle bin was then emptied, permanently deleting the files and folders.

To restore the deleted files and folders, ESG Lab again used SnapDrive and navigated to the LUN with the lost data.

As shown in Figure 9, a volume or LUN can be restored to a previous state with two clicks. Once ESG Lab clicked "Restore Disk," a warning was presented to ensure that any applications or services using the LUN were stopped prior to restoring. When ESG Lab clicked "Yes," the LUN was immediately restored, and the deleted files immediately reappeared.

Why This Matters Poor utilization, increasing complexity, rising costs, and the need to improve the availability and recoverability of IT services are driving a growing number of organizations to make major commitments to server and storage consolidation initiatives. ESG Lab has confirmed that NetApp Data ONTAP 8.1 can be used to create a scale-out cluster providing massively scalable storage under a single management interface from a single system image with easy-to-use common tools and techniques. Storage capacity requirements and management complexity are also rising as a growing number of applications-and users-rely on server virtualization. ESG Lab has confirmed that a centralized pool of heterogeneous storage supporting a consolidated mix of servers and applications can be easily virtualized and managed using the NetApp OnCommand management interface. Using NetApp SnapDrive, ESG Lab was able to create on-demand and scheduled snapshots of multiple volumes in seconds. This enabled ESG Lab to restore an entire volume to a previous point in time after a complete data loss.

Enterprise Non-stop Operations

ESG Lab validated the non-stop capabilities of Data ONTAP 8.1. Data ONTAP 8.1 is designed to enable users to manage, upgrade, and service their storage infrastructure non-disruptively over the life of their data. NetApp Data ONTAP 8.1 Cluster nodes are configured in fault tolerant pairs, with each node in a pair providing failover services for its partner. While a cluster may contain any combination of NetApp controller models that support Data ONTAP 8.1, the nodes in a storage failover pair must be the same model. All nodes in the cluster, configured in fault tolerant pairs, are managed as a single system.

ESG Lab Testing

For this test, ESG Lab used a six-node NetApp Data ONTAP 8.1 cluster (illustrated in Figure 10) and upgraded two of the nodes to the newest version of Data ONTAP 8.1 while a virtual machine running Windows 2008 Server ran a SQL Server workload continuously against a volume on an aggregate behind the systems being upgraded.[3] The nodes were named sjlab-01 through sjlab-06. Nodes sjlab-03 and sjlab-04 were selected for upgrade.

A server running SQLIO was used to simulate a live SQL Server by running an OLTP workload against a real SQL database. Figure 11 shows Perfmon running on the server before the upgrade was started.

ESG Lab began by downloading the new system image to the first node to be upgraded, sjlab-03. Once the image was downloaded and marked as the default, a reboot of the node would load the new code. First, ESG Lab modified the storage failover policy to disable auto-giveback of storage. This is a NetApp best practice, to prevent automated fallback until the upgrade is confirmed to be successful.

Next, ESG Lab initiated a storage failover triggering a takeover of sjlab-03's storage by sjlab-04. Figure 12 shows the iSCSI disk properties after the failover; it's important to note that the server has not had any interruption in service and still sees multiple paths to the storage through multiple nodes in the cluster.

Once sjlab-03 finished rebooting and the new code was verified, a storage failover was initiated to have sjlab-03 take over all storage from sjlab-04, and the upgrade process was repeated on sjlab-04. After the operation was complete and the code was verified, all storage was restored to the appropriate node.

Finally, ESG Lab confirmed that both nodes were now running the new system image, and the SQLIO workload was still running on the Windows 2008 virtual machine, as seen in Figure 13.

Why This Matters As storage environments grow in size and complexity, so too does the impact of data outages. More than half of IT managers surveyed by ESG indicated data availability as a major driver in choosing to deploy scale-out networked storage.[4] Regardless of the number and types of hardware failures that may occur during the life data on disk, managers, employees, and customers expect their data to be available. ESG Lab has confirmed that Data ONTAP 8.1 can provide an always-on storage environment able to   operate through planned maintenance and unplanned faults thanks to a robust, integrated highly available architecture combined with robust cluster services.

On-demand Flexibility and Performance Optimization

Next, ESG Lab examined NetApp protocol-independent flexibility by moving volumes between different disk aggregates without interrupting access or activity, as depicted in Figure 14. End-users would want to perform moves like this for multiple reasons: to balance load, for controller upgrades, or to move data to a different tier of storage.

ESG Lab Testing

Using OnCommand System Manager, a volume was selected for movement from an aggregate on a FAS3270 system to an aggregate on a V3270 system in the same cluster. The 100 GB volume was in use by a VMware vSphere server and contained approximately 30 GB of datastores used to store virtual machine disk (VMDK) files. The virtual machines were running at the time of the move.

As can be seen in Figure 15, the only information required to start the volume move are the names of the Vserver, the volume to be moved, and the destination disk aggregate.

When ESG Lab clicked "Move," the system immediately began copying data to the new volume on the new aggregate. While the copy was in progress, ESG Lab accessed all of the virtual machines attached to the datastore in the volume being moved.

In less than five minutes, the volume move was complete, as seen in Figure 16.

ESG Lab experienced zero interruptions in service before, during, or after the move.

Why This Matters Data migrations using traditional methods cause application downtime and can have significant business impact. Most customers can tolerate very little downtime without impacting revenue. The majority of respondents to a recent ESG research survey could not tolerate more than one to four hours of downtime before experiencing a significant impact to their businesses (23%). Nearly as many said that they could not tolerate more than an hour (22%) and many couldn't tolerate any downtime at all (14%). ESG Lab was able to configure and execute a volume migration quickly and easily using the NetApp OnCommand System Manager and moved a volume between storage platforms with no interruption to IO, which translates to no application downtime. Data ONTAP 8.1 provided on-demand storage mobility with zero overhead to the vSphere server and zero impact to running applications.

ESG Lab Validation Highlights

• ESG Lab was able to configure a new Vserver, create a volume, and present it to a server in less than ten minutes using the easy and intuitive NetApp OnCommand System Manager.
• SnapDrive enabled the fast and easy creation of capacity-efficient on-demand and scheduled snapshots of multiple volumes presented by the cluster. Full disk restores from snapshots were completed with two clicks in seconds.
• Data ONTAP 8.1 was able to move volumes between nodes quickly and completely non-disruptively.
• ESG Lab confirmed always-on availability by performing a rolling upgrade of multiple nodes in the cluster, with no disruption to data access.

Issues to Consider

• While a Data ONTAP 8.1 cluster can grow to 24 nodes when running NFS and CIFS, customers are limited to four nodes in a single cluster when running SAN protocols. Considering that NetApp supports a wide range of platforms from their low-end to their very largest systems in a single cluster, this will not be an issue for any organizations except those with the most extreme requirements. A cluster consisting of just four FAS 6000 series nodes could scale to 11 PB of capacity and many gigabytes per second of throughput. In discussions with NetApp ESG learned that the roadmap includes plans to qualify and support larger heterogeneous scale-out configurations in the future well beyond the qualified configurations available today.
• Adding or removing cluster nodes is a fairly complex, manual process and best practices recommend that it only be performed by skilled and trained personnel. To be fair, adding or removing cluster nodes are significant events and not something users are likely to perform on a regular basis, but some automation of these procedures (error checking, load balancing, etc...) would be a useful enhancement.
• Moving volumes around on the back end to balance capacity and workload is also a manual process at the time of this writing. While automation of back-end data movement to address resource and capacity requirements within the cluster is important, so is taking the time to implement this type of automation correctly. NetApp is working to integrate the capabilities of the OnCommand suite with Data ONTAP 8.1 to provide policy-based maintenance and management including load and capacity balancing.

The Bigger Truth

Large enterprises are looking to improve business responsiveness while continuing to reduce costs. IT departments are extending their virtualized infrastructure to a computing service-oriented infrastructure which demands non-disruptive operations with the highest levels of operational efficiency. In a perfect world, data centers would look quite different than they do today. They would be centrally managed, with applications, servers, and storage working in concert to provide reliable, cost-effective computing services to the business. Changes would be dynamic and fluid, with automation driving workflow. Most organizations, however, have to deal with the reality of an IT infrastructure far from this nirvana.

Even so, organizations are starting to recognize significant cost benefits by adopting a server virtualization strategy that allows IT to dynamically provision workloads for a variety of tasks-for production as well as test and development environments. It stands to reason that these same virtualization gains can be achieved in storage. Significant investments are already sunk into storage infrastructure that is costly to manage and grossly inefficient

The server world began to attain true reductions not only in capital expenditures but in operational costs by virtualizing discrete server operating systems that could only run on separate hardware in the physical. Provisioning services for workloads became quick and economical. But as virtualization began to proliferate, demands on the storage infrastructure became more severe. Storage became the bottleneck to truly automating provisioning of compute services to meet demand. Server administrators don't want to think about the datastore and how it's managed; they just want the storage when they need it. And storage administrators want to make datastores easily available to clients and applications without the overhead of managing cumbersome storage infrastructure.

Using NetApp with Data ONTAP 8.1, users can manage all the heterogeneous storage in their environments as a single pool using a single interface with common tools. Every NetApp V-Series or FAS system-whether primary or deep archive-runs Data ONTAP. Data ONTAP provides a consistent user interface and powerful storage efficiency technology. All storage can be available to applications and users over either SAN or NAS protocols.

ESG Lab quickly and easily virtualized a centralized pool of storage supporting a consolidated mix of servers and applications over multiple protocols using Data ONTAP 8.1. One-click access to powerful storage functionality was provided by NetApp OnCommand System Manager. Using both a NetApp FAS system and a third-party storage array behind a V-Series open storage controller, ESG Lab gained access to high performance, easy-to-manage scale-out storage virtualization.

If your organization is struggling to keep up with data growth, keep costs in check, and increase the availability of consolidated, virtualized business applications, ESG Lab recommends a serious look at the benefits that can be realized from virtualizing your storage environment with NetApp Data ONTAP 8.1. ESG Lab has confirmed that NetApp can bring a flexible and efficient service-oriented model to heterogeneous storage environments while reducing complexity and delivering a shared IT infrastructure foundation for nonstop operations, operational flexibility and efficiency, and on-demand IT services.

Appendix

---

[1] Source: ESG Research Report, 2011 IT Spending Intentions Survey, January 2011.

[2] Detailed configuration information can be found in the Appendix.

[3] As of this writing, Data ONTAP 8.1 clusters running SAN protocols are restricted to four nodes with six-node configurations scheduled for support in early 2012.

[4] Source: ESG Market Report, Scale-out 2.0: Simple, Scalable, Services-Oriented Storage, June 2010.

ESG Lab Reports The goal of ESG Lab reports is to educate IT professionals about emerging technologies and products in the storage, data management and information security industries. ESG Lab reports are not meant to replace the evaluation process that should be conducted before making purchasing decisions, but rather to provide insight into these emerging technologies. Our objective is to go over some of the more valuable feature/functions of products, show how they can be used to solve real customer problems and identify any areas needing improvement. ESG Lab's expert third-party perspective is based on our own hands-on testing as well as on interviews with customers who use these products in production environments. This ESG Lab report was sponsored by NetApp.