|
|
The Challenges of Cloud Storage Delivering an Information-Centric Cloud Storage Infrastructure Types of Cloud Storage Solutions Hybrid Storage – A new class of Storage Solution With the onset of cloud computing, and more specifically cloud storage, organizations see potential in being able to reduce cost and complexity for these key applications. The promise of capacity on-demand – being able to pay only for what is used and for what is transferred – is appealing, especially given that the price of (most) public cloud storage services are pennies on the dollar compared to traditional on-premises data center storage. The economies of scale and shared infrastructure allow service providers to deliver cloud-based storage at extremely low price points compared to that of a traditional infrastructure. But cloud-based computing and storage go far beyond this benefit by extending storage capabilities through the use of advanced API sets.
For IT customers, cloud-based storage is scalable, accessible, manageable and even more distributable, than a traditional storage infrastructure. Once the locality of data becomes irrelevant, users can integrate data from anywhere. And when the location of data is no longer important, it is easy to scale performance by distributing or moving data across any system according to demand. Exponential data growth particularly coming from unstructured data, is forcing companies to look at new, cost-effective ways to move their information from expensive primary disks to archive storage. Additionally an explosion of interest in large-scale, elastic and cloud-based data storage and processing suing open-source Hadoop extract new value from both complex and structured data is being implemented both in-house and using cloud-based infrastructure. Now cloud storage is being adopted as a reliable platform for long term archive needs given that new efficient tools are now available from vendors allowing users to quickly and securely move corporate information to the cloud.
A typical example of emerging VertiClouds© reveals tremendous benefits of using cloud storage services. The U.S. healthcare system spends $10–15 billion each year on redundant radiological exams that stem from physicians' lack of access to patients' prior imaging exams. Medical Imaging brings with it a significant and never ending demand for storage. The cloud computing and storage infrastructures and Vertical Cloud© platforms create an opportunity to address the significant need for providing a consistent set of networked healthcare services for radiology, cardiology and patient records transport and management to enable sharing diagnostic imaging information. By using a 10GbE network from end-users to Cloud, ultimately supports thousands of hospitals, imaging centers and hundreds of thousands of physicians and millions of patients. In turn, users benefit from a virtualized, secure, reliable and elastic nature of cloud services ensures a capability to scale the performance and operation of the network for peak usage periods. Also the pay-for-use cloud storage model provides an affordable model for health care facilities of different sizes. © 2008-10 IMEX Research Cloud Storage changes the game for distributed enterprises by delivering LAN-like application performance over the WAN and enabling a wave of IT consolidation from the branch into the data center or private cloud through optimizing network access for cloud storage deployments with new solutions Now, enterprises are beginning to migrate storage to the public cloud to take advantage of the dramatic operational and cost benefits. Having implemented virtualization in their data centers, organizations are now establishing creating new metrics and goals metrics to make their data centers cloud-ready.
The Challenges of Cloud Storage Given the obvious benefits of Cloud Storage, raises the question as to why haven’t more organizations already adopted cloud storage for on-premises applications and alleviate present expensive storage. The answer – Right now, there are traditional IT barriers which must be overcome when considering integration of cloud storage into today’s production environments. A survey conducted by IMEX Research reveals that both enterprises and SMBs today are concerned about:
Delivering an Information-Centric Cloud Storage Infrastructure Organizations that presently use on-premise storage and wish to selectively federate data to public clouds, would need to create an integration that enables automated searching and tagging of on-premise data and creates a policy-based federation of data to public clouds to deliver a truly information-centric cloud infrastructure keeping in mind the following: Organization Goals for Primary Storage In area of cloud storage, organization goals for primary storage include: Application Performance Optimization using Data Segmentation One of the key requirements to improve performance of data access and computations is to identify application/workload characteristics (IOPs, latency and bandwidth), data usage patterns (including frequency of access, I/Os, age, priority and relationship with other data) as well as data management (data placements, storage device characteristics and locality of reference) and then store the data on appropriate tier of storage and eliminate access hotspots.
Cloud Storage - Ideal for Certain Data Types Cloud Storage - a poor fit in following data types: Transactional Data
Active Corporate Data
Cloud Storage - good for the following data types: Large files with mostly read accesses
Long-Term Storage Files
Geographically Shared Files
Achieving Optimized Cloud Storage using Tiered Data Sets  Key to optimizing I/O performance of storage is understanding the characteristics of Data set volumes and taking advantage of tiered storage on-premise plus cloud storage services to achieve CAPSIMS© goals of
Types of Cloud Storage Solutions Customers need to decide whether or not a cloud storage service, public or private, should be used for a given data set and which cloud storage service class should be used before deciding to migrate to cloud storage services.
Cloud Clones are similar to snapshots in that they are a point-in-time, consistent copy of a series of application-related volumes. Whereas snapshots are typically mounted by a backup server and written to virtual or physical tape, which is then shipped off-site for protection, Cloud Clones on the other hand are stored persistently in the cloud. They are a replacement for VTL and tape libraries and allow one to mount and restore data at any point in time without having the hassle of retrieving tapes. Cloud Clones are de-duplicated and compressed, minimizing cloud capacity and data transfer costs, which makes them more cost compelling than tape as a form of archival storage. Cloud Clones are encrypted using a key the user supplies (or optionally the system can generate for the user) to ensure confidentiality while data sets are stored in the cloud using cloud clones. Hybrid Storage – A new class of Storage Solution
These solutions typically have the following characteristics: These solutions help address the immediate requirement for adding capacity inexpensively. Aside from the cost of the solution hardware, the rest of expense on storage is based on how much capacity the user utilizes in the cloud and how much data they transfer (These rates vary from cloud services providers such as Amazon Web Services, Rackspace etc.) Application Aware TieringThey have application-awareness that helps to address specific storage-related issues that impact the ability of an organization to scale or manage an application. This can be manifest in application-aware tiering, volume data location policies, and even plug-ins that integrate with frameworks provided by the application vendor. They provide primary storage deduplication, which not only helps improve performance when using the cloud, but also minimize the cloud cost (since only deduplicated data would ever be read from or written to the cloud). They also help address some of the concerns with Exchange and SharePoint regarding storage efficiency which is especially important, as Microsoft removed the Single Instance Store for attachments in Exchange 2010 and also for using versioning and extended recycle bins in SharePoint. Deduplication ensures that redundant email attachments are stored in a more space-efficient manner, meaning lower storage capacity consumption They allow users to control the encryption key, and it is never shared with the cloud provider. This increases confidence, as organizations don’t need to worry about what happens if someone gets a hold of their data from the cloud, or if the cloud provider is requested to release the data – or hardware where data may reside – in support of investigation or litigation. Coupled with other security services provided by the cloud provider, such as Virtual Private Networks, roles-based access control, multi-factor authentication etc., cloud storage can be as secure as traditional on-premises storage for certain applications. For these extra security features additional dedicated performance-boosting power is generally needed. Helping simplify data protection is one of the major benefits of Hybrid Storage Solutions. Having a virtually unlimited pool of storage sitting behind an on-premises hybrid storage appliance that provides data deduplication and encryption means that with the right application hooks and validation with data protection software vendors, users can take snapshots - that are not only crash-consistent but also application-consistent - and store them in a space-efficient, secure format on the cloud as an always-on backup and available in case needed for disaster-recovery. So instead of having to continually fetch tape from an off-site vault, users can simply use a cloud backup copy to perform a mailbox restore, object restore or even recover the entire application. Seamless Integration of Data Storage between Data Centers & Cloud Usually, when organizations decide to leverage cloud based storage solutions, they will be forced to rewrite their applications because most of the cloud storage providers give access to their storage using http or some REST based APIs. Traditional applications are not capable of accessing the storage that uses REST based API and it will require a rewrite of the applications or deployment of some sort of a connector. StorSimple mitigates this using hybrid storage solutions that makes cloud storage from multiple vendors such as EMC, IronMoutain, Microsoft, and Amazon, appear like local data center storage that integrates into customers' existing storage and data management tools. Much as these solutions are relatively nascent, many customers are achieving lower total cost of ownership, simplified backup and restore, better disaster recovery, and the ability to confidently scale many applications without concern over performance. Advanced Cloud Storage API enabled Capabilities A cloud NAS gateway that many storage providers offer, provides an important translation between data center network storage protocols (e.g. CIFS/NFS) and a more internet-optimized protocol (e.g.WebDAV) that is less chatty. The cloud NAS gateway provides a viable quick fix to get data moving to the cloud. The advantage of a cloud NAS is that it enables a customer to begin archiving data to the cloud almost as easily as they could send data to a local network mounted disk. This allows for rapid adoption of cloud storage as an archive destination. But Interacting and performing efficient communication to the cloud is just the bare minimum of what an API should do. Advanced cloud storage API sets allow following improved capabilities: Content Search Capability – An API set from within the programming language can provide the ability to deliver context searching during ingestion on the data that is being stored on it and create the index - all done on the back end, by the cloud storage system. Subsequent commands allow the application to query the index and retrieve results. These queries can be against the core content itself, or they can be metadata specified during ingestion and even display the results. This avoids having to issue subsequent retrieval requests to paint the search results. The query API set can return a set or a “page” at a time results, allowing the user browse through very large search results sets. Optimized Retrieval - Advanced API sets can also provide commands to the cloud storage to store the data in a parsed format allowing an organization to retrieve just the components of the data that is needed. Having these items individually addressable provides an important workaround to mitigate the latency of the cloud. A typical example would be to command the API to parse the metadata (header, body and attachments) only. On search only the relevant metadata (header and or body) is pulled across the internet. With relevant email’s header found, associated large attachments can then be brought across the internet. Retention and Compliance Controls - The API set should be enable to set retention and WORM levels within the application itself optionally set by an administrator or user or automatically be activated tied to an internal analysis of the data. Retention commands should include the ability to set length of time to be stored, number of copies to be kept and if the data can be modified and allow for the assured destruction of data, confirming that the data is no longer recoverable. Metadata Tagging - Metadata tagging by an API set allows for keywords or reference points to be set on files to help with search criteria, retention criteria or compliance criteria. Being able to build search optimization, pre-parsing data for retrieval, setting retention, compliance and metadata upfront at the point of archive goes a long way toward those feature sets actually being used. A feature-rich API set allows building classification into the process, enabling an interactive archive to become an upfront, quick task performed consistently will likely increase the accuracy of the classification as the application usually has the most context at that point compared to traditionally classification projects that are very large and involve processing of data long after it was written with specifics of that data forgotten. Security – The API data should be encrypted during transmission and remain encrypted when stored at the cloud storage provider’s facility. Not only does this prevent the infiltration of data in flight, it also insures that only that customer can read their data. The API set should have the ability to store multiple copies of that data not only locally but in geographically disperse locations as well, providing a comfort level in recovery from a failure or data corruption locally as well as globally. |
| Click on the following for additional information or go to http://www.imexresearch.com
|
Now a new class of storage solutions – namely hybrid storage solutions – are emerging that help address these exact challenges. A hybrid storage solution is:
