With data explosion here to stay, discovering new methods to control the gargantuan problem is paramount.
does having more data than you can manage have you pulling your hair out? Do you often find yourself searching frantically for a file the CEO has requested for? Does your inability to quickly retrieve an electronic document gives you sleepless nights?
If these scenarios sound familiar, CIOs might find information lifecycle management (ILM) the answer to their prayers - a strategy that leverages information , risk reduction and improving business agility. HP has a set of ILM solutions that synchronises business and IT to capitalise on change through a cradle-to-grave concept that intelligently manages data into useful information based on relevance to the business for easy and secure retrieval.
What is Information Lifecycle Management (ILM)?
ILM is a strategy for managing data from creation to archival and finally, deletion. All data moves through different stages as it ages: First data tends to be highly accessed and/or frequently updated. gradually, it becomes occasionally retrieved for viewing or updating. Eventually, the data is considered archived, as its content is fixed and no longer changing.
As data moves along its lifecycle, it passes across a hierarchy of storage devices, such as disk, tape, virtual tape and optical devices depending on its relevant business value. the concept behind ILM is to have information stored in the 'right place' depending on its business value to the organisation. In a best case scenario, it means: the least expensive storage media available that meets the operational needs for that piece of information, at that particular point in time.
Can ILM help me save cost and reduce my capital investment?
Absolutely. ILM extends the concept of centralised management of diverse storage resources for data that is created and then used by perhaps hundreds of applications and thousands of users. through ILM, the cost of managing information is measurable. For example, the cost of implementing ILM can be easily justified by the cost savings in the storage devices themselves.
In many cases, the storage capacity requirement for older data (referential data) grows faster than the need for new data (operational, revenue generating data). the exceptions are new and/or rapidly expanding businesses where, highly accessed data is created in much greater volumes than older data. Once the organisation understands the dynamic of its data - what percentage is highly accessed/active (operational), less active, and inactive (referential) - and how that will change over time, it can determine the cost savings achievable by automatically migrating referential data to the appropriate storage devices as the data ages.
As an organisation's overall capacity requirements increase, growing the lower (cheaper) end of the storage spectrum using cost-effective devices instead of the high end (more expensive) storage could save the organisation huge amount of capital, up to 60-80 percent. In addition, by storing inactive data on reliable but cheaper secondary media such as tape or optical disks, the costs of continually backing up the referential data can be eliminated.
What are the potential problems and challenges faced when adopting ILM?
Firstly, effective ILM relies on proper definition for management of data movement policies. Every application and every type of data created and/or used needs to be examined. Organisations must decide how data is used and where it should be stored at various points of its lifecycle.
Secondly, some ILM solutions do not recognise existing storage partitions and configurations. in some cases, all existing data will be moved from the devices it currently resides on so that the device can then be added to the ILM storage pool, and subsequently the data will be copied back. Fortunately, ILM solutions can handle the assimilation in the background, seamlessly moving all data and resources into its domain.
Thirdly, a typical storage network is designed to centrally manage various kinds of storage devices whereas the communication between application servers and their assigned storage resources remains decentralised (via a switched network). By pooling all storage network resources into a single virtual repository, ILM will need a virtual file system that controls the entire storage network and keep track of where everything is stored. this means that the ILM system can potentially becomes a single point of failure and bottleneck. proper design and architecting is a crucial part of a successful ILM journey.
What is the right approach to ILM?
To ensure your ILM implementation achieves its full potential, HP has identified five phases to the journey:
1) Data discovery and Classification
2) Storage Tiering
3) Data Movement by Policy
4) Continuous Information Availability
5) Application-Aware ILM
Data discovery and classification begins the process. During this phase, HP helps organisations get a bird's-eye view of their data - developing a clear understanding of its level of importance according to business value an d how the data ages within its lifecycle. By the end of the process, it will be possible to leverage on the discoveries to create more effective policies.
The next step is to place data into storage tiers using options such as online storage disk arrays or Network Attached Storage (NAS) devices for operational data; low-cost disk Serial Advanced Technology Attachment (SATA) systems or HP's hybrid Fibre Attached Technology Adapted (FATA) drives for active archive storage; and tape backup storage for reference data.
Data movement by policy is the critical third phase, where information is moved automatically based on relevance. Organisations can improve total cost of ownership by matching the business value of the information at each point in its lifecycle to the cost of its storage platform.
Next, business can focus on fulfilling the need for continuous data availability (both operational and referential). By indexing referential data, users can quickly and easily perform fast search and retrievals and allow archives to be restored to meet business and audit/regulatory compliance service levels.
The last stop on the ILM road map is the adoption of application-aware ILM solutions which address issues such as messaging, databases, Enterprise Resource Planning (ERP), and Customer Relationship Management (CRM). Completing the full circle: data created by specific applications would be archived and accessed by the right people and systems, managed within the policies that have been created specific to each class of data based on its relevance to the business.
Can you share the story of a successful ILM implementation?
NASDAQ, the largest electronic stock market in the US, is a prime example. It has to handle tens of thousands of emails daily and in accordance to regulations, must store such records and business correspondence for at least five years and be able to retrieve records within two days.
NASDAQ decided to implement the HP Storage Works Reference Information Storage Systems (RISS) - a high performance IT archive and retrieval solution. In the past, NASDAQ staff had to manually download tapes to disk before combing through data. RISS allows for automatic archiving of recent emails on slower, less expensive disks. The system is also able to index, search and retrieve embedded e-mail or in email attachments. HP RISS now allows NASDAQ to meet e-mail retention request within hours.
Is Backup an ILM Strategy?
Organisation are being inundated with information that must be stored, protected, and retrieved, often very quickly due to internal corporate governance, business requirements and compliance. Combined with regulatory mandates like HIPAA, Sarbanes-Oxley, and Basel 2, companies everywhere are discovering a painful lesson: backup isn't even archiving, much less ILM, In the HP view of ILM, data capture, management, and delivery are as critical because it is not only about how information is stored but how information is put to work for the organisation.
Yet, most organisations persist in relying on their backup process for what amounts to archiving. they make a full backup and then they make daily incremental backups. he backups are regularly shipped off site for long term storage. Organisations that want to leverage this cast pool of backup data may suddenly find it is far more difficult and costly than expected to recover it. Data that has been backed up to tape cannot be searched, accessed and retrieved until it has been restored - a lengthy process. Backed up data simply is not stored, managed and maintained in a form that allows for easy, real-time access to data as archiving and ILM provide.