Storage Area Networks (SAN) and Networked Attached Storage (NAS) and Direct Attached Storage (DAS) are generally the storage approaches most IT managers envision when they talk about storage architectures. Today, a forth approach to storage in the form of Clustered Storage has come into being.
Data storage industry is witnessing a paradigm shift in the form of movement toward Clustered Storage architectures. In fact clustered storage has the potential of changing the rules of how data is stored, managed and accessed.
Exponential Growth of Unstructured Data and Digital Content
Data or information recession is never going to happen. Today’s organisations are facing an exponential increase in the amounts of data driven largely by the explosion of unstructured data. It is well-known that power-point presentations, audio, video, images and other large digital files and unstructured data are growing much faster than structured information such as databases, enterprise resource planning (ERP) and mail servers.
Today unstructured data is growing much faster than structured data. PowerPoint presentations, MP3s, Microsoft Word documents, the things that people use on their laptops and desktops -- all of this
Another driving factor for the adoption of clustered storage is the widespread adoption of clustered computing.
Clustered NAS is essentially scaling out of NAS from a single storage node or controller to multiple controllers or storage nodes. Narayanan Balakrishnan, Project Manager-Storage Group, American Megatrends India, says, “ There are a number of types of NAS clusters. It could be a set of clustered NAS heads with shared backend storage. Alternately, it could be an out of the box NAS solution.
It can be an active-passive configuration where one file system is accessed only from one node at a time and the other node(s) take over when the first one fails or it can be an active-active configuration where both nodes can simultaneously provide access to CIFS, NFS and other clients.
In addition, the NAS file system might be resident only on node of the cluster. Alternately, the file system itself could be distributed on some set of nodes or all nodes of the cluster. Likewise, the failover during node failures could be seamless or disruptive.
NAS vs. SAN
Both NAS and SAN provide network storage solutions. However, NAS is a single storage device that operates on data files, whereas SAN is a local network of multiple devices that operate on disc blocks. While a SAN normally uses Fibre Channel interconnects, a NAS makes Ethernet and TCP/IP connections.
However, due to the worldwide proliferation of Internet technologies like TCP/IP and Ethernet, some SAN products are making the transition from Fibre Channel to the same IP-based approach NAS uses. And with the rapid improvements in disk storage technology, today's NAS devices now offer capacities and performance that once were only possible with SAN. “These two industry factors have led to a partial convergence of NAS and SAN approaches to network storage,” says Amit Luthra, National Manager, Storage & Networking Solutions Marketing, Dell.
Both SAN and NAS have their own distinct areas of use. “Both are equally equipped to address the storage needs. In fact it should be combination of both. And hence the emergence of unified storage,” says Deep Roy, Consulting Systems Engineer, Technology Solutions Organization – APAC, NetApp. “Unified Storage has in fact ended the NAS vs. SAN debate,” adds Deep.
While NAS is typically less expensive, easy to deploy, integrate and use, uses industry standard protocols like CIFS, NFS, SAN storage cater to high capacity, high performance applications. “Most vendors have adopted the approach of unified storage with both block (SAN) and file (NAS) storage from the storage server. “NAS implementations of today match the advanced feature set of SAN,” says Balakrishnan of AMI. “This proves that both of these have a place and they can coexist,’ adds Balakrishnan .
“Over a period of time, the growth of data has increased exponentially. With the emergence of different types of data forms, such as video on demand digital surveillance, which are classified as unstructured data, Clustered storage has come to the forefront of attention,’ says Akhil Kamat, Country Manager, Storage, Systems & Technology Group, IBM India/SA.
Requirements for Setting up the Cluster
The driving force in choosing a type of NAS cluster is the application it is used for. Depending on the performance, availability and cost considerations the required NAS cluster solution must be carefully chosen. Depending on the type of Clustered NAS, the prerequisites for a solution may vary. For example, NAS gateway clusters might require a separate set of NAS head nodes distinct from the backend block storage, which may or may not be clustered as well.
Clustered file systems (CFS) offer a practical way to respond to big storage problems such as the proliferation of low-cost servers, application data growth and the need to deliver better application performance. A CFS pulls together and shares the excess storage capacity that's often available but hidden on storage networks. In doing so, a CFS increases storage utilization rates, delivers performance typically found only in high-end arrays and gives users an economical way to scale their architectures.
There are three ways to deploy a CFS: on storage servers, NAS gateways and hosts. Any server in the cluster can access any block of storage managed by the cluster. Most CFS also integrate the volume manager with the file system. This allows the CFS to break large files into blocks called extents, and to stripe those extents across different storage arrays to improve I/O performance.
Driving the movement to clustered storage is a dramatic decrease in the price- performance curves of industry-standard hardware components. This trend is part of the continual movement toward the promise of Moore’s Law: over time, companies are getting higher computing power for a lower cost and realizing the economics of commodity hardware. The low cost of commodity hardware components has made the merits of clustered architectures affordable. Google is a prime example of how clustering has leveraged the price-performance curves of industry-standard hardware to realize industry-leading performance and reliability at a fraction of the cost of traditional, custom-built systems. On average, a single query on Google reads hundreds of megabytes of data and consumes tens of billions of CPU cycles. To handle this “high performance computing” workload, Google’s architecture features clusters of thousands of commodity class PCs, off-the-shelf components, with fault-tolerant software. This clustered architecture achieves superior performance at a fraction of the cost of a system built from fewer, but more expensive, high-end servers.
Trends Driving the Movement to Clustered Storage
Clustered Storage Becoming Mainstream
Clustered NAS solutions are still nascent. With growing requirements of storage, file systems of larger sizes, clustered NAS solutions are becoming important from scalability, performance and availability standpoints. But, still a number of file systems, technologies and protocols are in the process of evolution. So, it might take some time till this gets completely in main-stream.
“Though the need for clustered storage is definitely there as customer’s scalability requirement is growing day by day, it is interesting to see that it is not gaining that much of popularity as it should have, “ says Deep.
Clustered technology driving huge business benefits:
Vendors offering Clustred Storage
The very name of NAS brings NetApp to the mind. The company introduced unified storage systems more than a decade ago. The company’s continued leadership and unwavering commitment to innovation have resulted in an architecture unified not only across protocols and disks, but also in storage efficiency, processes, data management, and data protection.
NetApp Data ONTAP 8 offers a unified scale-out storage solution for an adaptable, always-on storage infrastructure to accommodate today's virtualized infrastructures. You can deploy not only NAS storage and file-based protocols, but also the SAN protocols-Fibre Channel, iSCSI, and FCoE-in a shared scale-out data storage system.
IBM has its Scale-Out NAS or SONAS family, which is designed to embrace and deliver cloud storage in the petabyte age. SONAS can meet today’s storage challenges with quick and cost-effective IT-enabled business enhancements designed to grow with unprecedented scale.
Clustered storage means different things to different people. Common characteristics while defining clustered storage
SONAS can also deliver storage services that make the supporting technology almost invisible. It allows applications and services to be uncoupled from the underlying infrastructure, enabling businesses to adjust to change quickly. As a result, SONAS can easily integrate with your organization’s strategies to develop a more dynamic enterprise.
“We have designed our entire strategy around clustered file system is that people are able to grow right from small requirement to large one. We have customer who have deployed terabytes of information on clustered file system on Scale-Out NAS or SONAS family,” says Akhil.
Shift to Clustered Storage architectures is well neigh a revolution which is underway in the data storage industry. It is becoming the dominant new storage architecture benefiting the customers in a large way. The bottom line is that Clustered Storage solutions are gaining momentum. Its adoption is increasing at an exponential pace.
Clustered storage is gaining prominence at a time when enterprises are focusing on what to do with large amounts of unstructured file data. Big data, vast amounts of unstructured information, is being talked about. Clustered NAS, with its ability to store very large volumes of data that is all visible via one file system, is suitable for this.
2015 © American Megatrends India Pvt Ltd. ALL Rights Reserved.